View PDF Version

In the Supreme Court of the United States

UNITED STATES OF AMERICA, ET AL., APPELLANTS

v.

AMERICAN LIBRARY ASSOCIATION, ET AL.

ON APPEAL FROM THE
UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF PENNSYLVANIA

JOINT APPENDIX
(VOLUME 1)

THEODORE B. OLSON*
Solicitor General
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217

* Counsel of Record
for Appellants

PAUL M. SMITH*
THERESA A. CHIMARA
DANIEL MACH
KATHERINE A. FALLOW
Jenner & Block, LLC
601 13th Street, N.W.
Washington, D.C. 20005
(202) 639-6000
*Counsel of Record
for Appellees American
Library Association, et al.
CHRISTOPHER A. HANSEN*
ANN BEESON
STEVEN R. SHAPIRO
American Civil Liberties
Union Foundation
125 Broad Street
New York, New York 10004
(212) 549-2500
*Counsel of Record
for Appellees Multnomah
County Public Library, et al.

 

APPEAL FILED: SEPTEMBER 6, 2002
PROBABLE JURISDICTION NOTED: NOVEMBER 12, 2002

UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF PENNSYLVANIA (PHILADELPHIA)

 

Civil Docket No. 01-CV-1303

AMERICAN LIBRARY ASSOCIATION, INC., ET AL., PLAINTIFFS

v.

UNITED STATES, ET AL., DEFENDANTS

RELEVANT DOCKET ENTRIES

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

 

3/20/01 1 Complaint for the Declaratory and Injunctive Relief. filing fee $ 150 receipt # 756926 (jmp) [Entry date 03/20/01]

* * * * *

3/26/01 2 ORDER THAT INASMUCH AS SECTION 561 OF THE COMMUNICATIONS DECENCY ACT OF 1996 REQUIRES A DISTRICT COURT OF THREE JUDGES TO BE CONVENED IN ANY CIVIL ACTION CHAL-LENGING THE CONSTITU-

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

TIONALITY, ON ITS FACE, OF ANY PROVISION OF THAT STATUTE, AND I HAVE BEEN REQUESTED TO CONVENE SUCH A COURT BY THE DISTRICT JUDGE TO WHOM THE PLAINTIFFS' REQUEST WAS MADE (SEE ATTACHED), PURSUANT TO THE PROVISIONS OF 28 U.S.C. SECTION 2284, I DESIGNATE THE HONORABLE EDWARD R. BECKER, UNITED STATES CIRCUIT JUDGE, AND THE HONORABLE HARVEY BARTLE, III, UNITED STATES DISTRICT JUDGE FOR THE EASTERN DISTRICT OF PENNSYLVANIA, TO SIT WITH THE HONORABLE JOHN P. FULLAM, UNITED STATES DISTRICT JUDGE FOR THE EASTERN DISTRICT OF PENNSYLVANIA, AS MEMBERS OF THE COURT FOR THE HEARING

AND DETERMINATION OF THE ABOVE-CAPTIONED MATTERS. (SIGNED BY

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

EDWARD R. BECKER, CHIEF JUDGE THIRD JUDICIAL CIRCUIT) 3/26/01 ENTERED AND COPIES MAILED. (fp) Entry date 03/26/01] [Edit date 09/12/01]

* * * * *

5/15/01 15 ORDER THAT ACTION NOS. 01-1303 AND 01-1322 ARE CONSOLIDATED FOR ALL PURPOSES UNDER ACTION NO. 01-1303; DEFTS SHALL FILE A MOTION TO DISMISS BY 6/8/01; PLFFS SHALL FILE THEIR RESPONSES BY 6/29/01; DEFTS MAY FILE REPLY BRIEF BY 7/13/01 PLFFS SHALL PROVIDE DEFTS WITH THEIR PROPOSED DISCOVERY PLAN BY 5/29/01; DEFTS SHALL PROVIDE PLFFS A COUNTER-PROPOSAL BY 6/15/01; A DISCOVERY CONFERENCE WILL BE HELD 6/26/01 AT 10:00; ONE DAY PRIOR TO

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

SAID CONFERENCE, PATIES SHALL SUBMIT COPIES OF PROPOSED DISCOVERY PLAN(S) TO JUDGES BECKER, FULLAM AND BARTLE; AND ARGUMENT ON MOTION TO DISMISS WILL BE HELD ON 7/23/01 AT 11:00. ( SIGNED BY JUDGE JOHN P. FULLAM ) 5/15/01 ENTERED AND COPIES MAILED AND FAXED. (mld) [Entry date 05/15/01] [Edit date 05/15/01]

* * * * *

6/8/01 18 MOTION by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT FED. COMMUNICATIONS, DEFENDANT BEVERLY SHEPPARD, DEFENDANT INSTITUTE OF MUSEUM TO DISMISS PLFFS' COMPLAINT , CERTIFICATE OF SERVICE. (mld) [Entry date 06/08/01]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

6/8/01 19 Memorandum of Law by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT FED. COMMUNICATIONS, DEFENDANT BEVERLY SHEPPARD, DEFENDANT INSTITUTE OF MUSEUM in support of their MOTION TO DISMISS PLFFS' COMPLAINT, Certificate of Service. (mld) [Entry date 06/08/01]

* * * * *

6/29/01 21 Response by PLAINTIFF AMERICAN LIBRARY, PLAINTIFF FREEDOM TO READ, PLAINTIFF ALASKA LIBRARY ASSOC, PLAINTIFF CALIFORNIA LIBARY, PLAINTIFF NEW ENGLAND LIBARY, PLAINTIFF NEW YORK LIBRARY, PLAINTIFF ASSOC.OF COMMUNITY, PLAINTIFF FRIENDS OF THE PHILA, PLAINTIFF PA. ALLIANCE FOR DEM, PLAINTIFF ELIZABETH HRENDA,

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

PLAINTIFF C. DONALD WEINBERG in opposition to DEFTS' MOTION TO DISMISS PLFFS' COMPLAINT, Memorandum, Certificate of Service. (mld) [Entry date 07/02/01]

* * * * *

7/13/01 23 Reply Brief by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT FED.COMMUNICATIONS, DEFENDANT BEVERLY SHEPPARD, DEFENDANT INSTITUTE OF MUSEUM in support of their MOTION TO DISMISS PLFFS' COMPLAINT, Certificate of Service. (mld) [Entry date 07/13/01] [Edit date 07/13/01]

7/26/01 24 MEMORANDUM AND ORDER THAT DEFTS' MOTIONS TO DISMISS THE COMPLAINTS ARE DENIED. ( SIGNED BY JUDGE JOHN P. FULLAM ) 7/26/01 ENTERED AND COPIES MAILED AND FAXED. (mld) [Entry date 07/26/01]

* * * * *

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

9/10/01 34 Answer to Complaint by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT INSTITUTE OF MUSEUM, DEFENDANT BEVERLY SHEPPARD, DEFENDANT FED.COMMUNICATIONS, Certificate of Service. (np) [Entry date 09/12/01] [Edit date 09/12/01]

* * * * *

11/15/01 37 ORDER THAT IN ACCORDANCE WITH THE UNDERSTANDINGS REACHED AT A TELEPHONE CONFERENCE HELD THIS DATE, THAT TRIAL OF THIS CASE IS SET TO COMMENCE ON 3/25/02; PLFFS' PRELIMINARY PRETRIAL MEMORANDUM WILL BE FILED BY 12/14/01; DEFTS' PRELIMINARY PRETRIAL MEMORANDUM WILL BE FILED BY 12/28/01. (SIGNED BY JUDGE JOHN P. FULLAM ) 11/15/01 ENTERED AND COPIES MAILED AND FAXED BY CHAMBERS. (mld) [Entry date 11/15/01]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

* * * * *

3/20/02 47 Joint Trial Stipulations of all Parties. (FILED UNDER SEAL) (mld) [Entry date 03/21/02]

3/20/02 48 Redacted Joint Trial Stipulations of all Parties, Certificate of Service. (mld) [Entry date 03/21/02]

* * * * *

3/26/02 52 Minute entry re: Civil Non Jury Trial held 3/25/02 DAY 1 Witnesses called and sworn: Motion of Non Party N2H2 for limited intervention: Motion GRANTED. (jpd) [Entry date 03/26/02] [Edit date 03/26/02]

* * * * *

3/27/02 54 Minute entry re: 3/26/02 civil non-jury trial day 2. Geoffrey Nunberg, resumes; witnesses sworn. (mld) [Entry date 03/27/02]

3/27/02 55 Transcript for Trial held on 3/25/02. (mld) [Entry date 03/27/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

3/29/02 56 Minute entry: Non-Jury Trial resumes 3/28/02. (td) [Entry date 03/29/02]

3/29/02 57 Transcript for Trial of 3/26/02. (td) [Entry date 03/29/02]

3/29/02 58 Minute entry re: Non Jury Trial Day 3 held 3/28/02: Witnesses called and sworn. (jpd) [Entry date 03/29/02]

3/29/02 59 Minute entry re: Non Jury Trial Day 4 held 3/29/02: Witnesses called and sworn. (jpd) [Entry date 03/29/02]

3/29/02 60 Minute entry re: 3/29/02 non-jury trial, day 4. Remarks by the Court; witnesses called and sworn in DEFTS' case. (mld) [Entry date 04/01/02]

4/1/02 61 Minute entry re: 4/1/02 non-jury trial, day 5. Remarks by the Court; witnesses called and sworn in DEFTS' case. (mld) [Entry date 04/01/02]

4/2/02 62 Transcript for Trial held 3/28/02. (mld) [Entry date 04/03/02]

4/2/02 63 Transcript for Trial held 3/29/02. (mld) [Entry date 04/03/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

4/2/02 64 Transcript for Trial held 4/1/02. (mld) [Entry date 04/03/02]

4/2/02 65 Minute entry re: 4/2/02 non-jury trial, day 6. Remarks by the Court; PLFFS' case resumes; witnesses called and sworn in PLFFS' case. (mld) [Entry date 04/03/02]

4/3/02 66 Minute entry re: 4/3/02 non-jury trial, day 7. Remarks by the Court; witnesses called and sworn in DEFTS' case; exhibits moved into evidence by Atty BHATTACHARYYA on behalf of DEFTS; DEFTS rest; PLFFS rest. (mld) [Entry date 04/03/02]

4/3/02 67 Transcript for Trial held 4/2/02. (mld) [Entry date 04/04/02]

4/4/02 68 Minute entry re: 4/4/02 non-jury trial, day 8. Remarks by the Court; closing remarks by Attys SMITH and HANSEN

on behalf of PLFFS; closing remarks by Atty BHATTACHARYYA on behalf of DEFTS; rebuttal by Atty SMITH. By the Court: counsel for both

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

sides to submit to court Findings of Facts and Briefs by 4/11/02. (mld) [Entry date 04/05/02]

4/5/02 69 Transcript for Trial held 4/3/02. (mld) [Entry date 04/05/02]

4/9/02 70 Transcript for Trial held 4/4/02. (mld) [Entry date 04/09/02]

4/11/02 71 Joint Post-Trial Brief by PLAINTIFF AMERICAN LIBRARY, PLAINTIFF FREEDOM TO READ, PLAINTIFF ALASKA LIBRARY ASSOC, PLAINTIFF CALIFORNIA LIBARY, PLAINTIFF NEW ENGLAND LIBARY, PLAINTIFF NEW YORK LIBRARY, PLAINTIFF ASSOC.OF COMMUNITY, PLAINTIFF FRIENDS OF THE PHILA, PLAINTIFF PA. ALLIANCE FOR DEM, PLAINTIFF ELIZABETH HRENDA, PLAINTIFF C. DONALD WEINBERG, Certificate of Service. (mld) [Entry date 04/12/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

4/11/02 72 Joint Proposed Findings of Facts by PLAINTIFF AMERICAN LIBRARY, PLAINTIFF FREEDOM TO READ, PLAINTIFF ALASKA LIBRARY ASSOC, PLAINTIFF CALIFORNIA LIBARY, PLAINTIFF NEW ENGLAND LIBARY, PLAINTIFF NEW YORK LIBRARY, PLAINTIFF ASSOC. OF COMMUNITY, PLAINTIFF FRIENDS OF THE PHILA, PLAINTIFF PA. ALLIANCE FOR DEM, PLAINTIFF ELIZABETH HRENDA, PLAINTIFF C. DONALD WEINBERG, Certificate of Service. (mld) [Entry date 04/12/02] [Edit date 04/12/02]

4/12/02 73 Post-Trial Brief by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT FED. COMMUNICATIONS, DEFENDANT BEVERLY SHEPPARD, DEFENDANT INSTITUTE OF MUSEUM, Certificate of Service. (mld) [Entry date 04/12/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

4/12/02 74 Proposed Findings of Facts

by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT FED.COMMUNICATIONS, DEFENDANT BEVERLY SHEPPARD, DEFENDANT INSTITUTE OF MUSEUM, Certificate of Service. (mld) [Entry date 04/12/02]

4/18/02 75 Post-Trial Reply Brief by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT FED.COMMUNICATIONS, DEFENDANT BEVERLY SHEPPARD, DEFENDANT INSTITUTE OF MUSEUM, Certificate of Service. (mld) [Entry date 04/18/02]

4/18/02 76 Supplemental Proposed Findings of Facts by PLAINTIFF AMERICAN LIBRARY, PLAINTIFF FREEDOM TO READ, PLAINTIFF ALASKA LIBRARY ASSOC, PLAINTIFF CALIFORNIA LIBARY, PLAINTIFF NEW ENGLAND

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

LIBARY, PLAINTIFF NEW YORK LIBRARY, PLAINTIFF ASSOC. OF COMMUNITY, PLAINTIFF FRIENDS OF THE PHILA, PLAINTIFF PA. ALLIANCE FOR DEM, PLAINTIFF ELIZABETH HRENDA, PLAINTIFF C. DONALD WEINBERG in response to DEFENDANTS' Proposed Findings of Facts, Certificate of Service. (mld) [Entry date 04/19/02]

4/18/02 77 Joint Post-Trial Reply Brief by PLAINTIFF AMERICAN LIBRARY, PLAINTIFF FREEDOM TO READ, PLAINTIFF ALASKA LIBRARY ASSOC, PLAINTIFF CALIFORNIA LIBARY, PLAINTIFF NEW ENGLAND LIBARY, PLAINTIFF NEW YORK LIBRARY, PLAINTIFF ASSOC.OF COMMUNITY, PLAINTIFF FRIENDS OF THE PHILA, PLAINTIFF PA. ALLIANCE FOR DEM, PLAINTIFF ELIZABETH HRENDA, PLAINTIFF C. DONALD WEINBERG, Certificate of Service. (mld) [Entry date 04/19/02]

* * * * *

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

5/31/02 79 OPINION AND ORDER DATED MAY 31, 2002 THAT JUDGMENT IS ENTERED IN FAVOR OF THE PLAINTIFFS AND AGAINST THE DEFENDANTS, DECLARING THAT 1712(a)(2) AND 1721(b) OF THE CHILDREN'S INTERNET PROTECTION ACT, 20 U.S.C. 9134(f) AND 47 U.S.C. 254(h)(6), ARE FACIALLY INVALID UNDER THE FIRST AMENDMENT TO THE UNITED STATES CONSTITUTION; AND THE UNITED STATES, MICHAEL POWELL, IN HIS OFFICIAL CAPACITY AS CHAIRMAN OF THE FEDERAL COMMUNICATIONS COMMISSION, THE FEDERAL COMMUNICATION COMMISSION, BEVERLY SHEPPARD, IN HER OFFICIAL CAPACITY AS ACTING DIRECTOR OF THE INSTITUTE OF MUSEUM AND LIBRARY SERVICES, AND THE INSTITUTE OF MUSEUM AND LIBRARY SERVICES ARE

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

PERMANENTLY ENJOINED FROM WITHHOLDING FEDERAL FUNDS FROM ANY PUBLIC LIBRARY FOR FAILURE TO COMPLY WITH 1712(a)(2) AND 1712(b) OF THE CHILDREN'S INTERNET PROTECTION ACT, 20 U.S.C. 9134(f) AND 47 U.S.C. 254(h)(6). (SIGNED BY CHIEF JUDGE EDWARD R. BECKER, JUDGE JOHN P. FULLAM AND JUDGE HARVEY BARTLE, III ) 5/31/02 ENTERED AND COPIES MAILED. (mc) [Entry date 05/31/02]

6/20/02 80 Notice of appeal to the UNITED STATES SUPREME COURT by DEFENDANT UNITED STATES, DEFENDANT MICHAEL POWELL, DEFENDANT FED.COMMUNICATIONS, DEFENDANT BEVERLY SHEPPARD, DEFENDANT INSTITUTE OF MUSEUM. Copies to: JUDGE JOHN P. FULLAM, JUDGE HARVEY BARTLE, III, USCA CHIEF

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

JUDGE EDWARD R. BECKER, and JANET M. LARUE, MEERA DEO, LEE TIEN, DAVID L. SOBEL, CHARLES F. SIMMS, FRANK SCIBILIA, ANDREW L. LEE, STEFANIE S. KRAUS, CHRISTOPHER A. HANSEN, ANN ELIZABETH BEESON, STEFAN PRESSER, ANDREA GACKI, LISA M. BORNSTEIN, TIMOTHY ZICK, THEODORE C. HIRT, ANNETTA FOSTER GIVHAN, SCOTT A. COFFINA, KATHERINE A. FALLOW, JENNIFER S. MARTINEZ, PAUL M. SMITH, DANIEL MACH, THERESA A. CHMARA, WAYNE C. STANSFIELD, ROBERT A. NICHOLAS, Certificate of Service. (mld) [Entry date 06/20/02]

* * * * *

 

UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF PENNSYLVANIA (PHILADELPHIA)

Civil Docket No. 01-CV-1322

MULTNOMAH COUNTY PUBLIC LIBRARY, ET AL., PLAINTIFFS

v.

UNITED STATES , ET AL. DEFENDANTS

RELEVANT DOCKET ENTRIES

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

3/20/01 1 Complaint. filing fee $150 receipt #756978 (aam) [Entry date 03/20/01]

* * * * *

3/26/01 3 ORDER THAT INASMUCH AS SECTION 561 OF THE COMMUNICATIONS DECENCY ACT OF 1996 REQUIRES A DISTRICT COURT OF THREE JUDGES TO BE CONVENED IN ANY CIVIL ACTION CHALLENGING THE CONSTITUTIONALITY, ON ITS FACE, OF ANY PROVISION OF THAT STATUE, AND I HAVE BEEN REQUESTED TO CON

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

VENE SUCH A COURT BY THE DISTRICT JUDGE TO WHOM THE PLAINTIFFS' REQUEST WAS MADE, PURSUANT TO THE PROVISIONS OF 28 U.S.C. & 2284, I DESINGATE THE HONORABLE EDWARD R. BECKER, UNITED STATES CIRCUIT JUDGE, AND THE HONORABLE HARVEY BARTLE, III, UNITED STATES DISTRICT JUDGE, TO SIT WITH THE HONORABLE JOHN P. FULLEM, UNITED STATES DISTRICT JUDGE FOR THE EASTERN DISTRICT OF PENNSYLVANIA, AS MEMBERS OF THE COURT FOR THE HEARING AND DETERMINATION OF THE ABOVE-CAPTIONED MATTERS (SIGNED BY CHIEF JUDGE THIRD CIRCUIT EDWARD R. BECKER) 3/26/01 ENTERED AND COPIES MAILED. (jl) [Entry date 03/26/01]

* * * * *

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

5/15/01 8 ORDER THAT ACTION NOS. 01-1303 AND 01-1322 ARE CONSOLIDATED FOR ALL PURPOSES UNDER ACTION NO. 01-1303; AND DEFTS SHALL FILE A MOTION TO DISMISS BY 6/8/01; PLFFS SHALL FILE THEIR RESPONSES BY 6/29/01; DEFTS MAY FILE REPLY BRIEF BY 7/13/01; PLFFS SHALL PROVIDE DEFTS WITH THEIR PROPOSED DISCOVERY PLAN BY 5/29/01; DEFTS SHALL PROVIDE PLFFS A COUNTER-PROPOSAL BY 6/15/01; A DISCOVERY CONFERENCE WILL BE HELD ON 6/26/01 AT 10:00; AND ARGUMENT ON MOTION TO DISMISS WILL BE HELD 7/23/01 AT 11:00. (SIGNED BY JUDGE JOHN P. FULLAM) 5/15/01 ENTERED AND COPIES MAILED AND FAXED. (mld) [Entry date 05/15/01] [Edit date 05/15/01]

* * * * *

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

6/8/01 - MOTION by DEFENDANT UNITED STATES, DEFENDANT FEDERAL COMMUNICATION, DEFENDANT INSTITUTE OF MUSEUM TO DISMISS PLFFS' COMPLAINT, CERTIFICATE OF SERVICE. (See case 01-1303, paper #18) (mld) [Entry date 06/08/01] [Edit date 06/08/01]

6/8/01 - Memorandum of Law by DEFENDANT UNITED STATES, DEFENDANT FEDERAL COMMUNICATION, DEFENDANT INSTITUTE OF MUSEUM in support of their MOTION TO DISMISS PLFFS' COMPLAINT, Certificate of Service. (See case 01-1303, paper #19) (mld) [Entry date 06/08/01] [Edit date 06/08/01]

* * * * *

6/29/01 9 Response by PLAINTIFF MULTNOMAH COUNTY, PLAINTIFF CONNECTICUT LIBRARY, PLAINTIFF MAINE LIBRARY ASSOC., PLAINTIFF SANTA CRUZ PUBLIC, PLAINTIFF SOUTH CENTRAL LIBRAR, PLAINTIFF WESTCHESTER

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

LIBRARY, PLAINTIFF WISCONSIN LIBRARY, PLAINTIFF MARK BROWN, PLAINTIFF SHERRON DIXON, PLAINTIFF JAMES GERINGER, PLAINTIFF MARNIQUE TYNESHA OVERBY, PLAINTIFF EMMALYN ROOD, PLAINTIFF WILLIAM J. ROSENBAUM, PLAINTIFF CAROLYN C. WILLIAMS, PLAINTIFF QUIANA WILLIAMS, PLAINTIFF AFRAIDTOASK.COM, PLAINTIFF ALAN GUTTMACHER INST, PLAINTIFF ETHAN INTERACTIVE, PLAINTIFF NATURIST ACTION, PLAINTIFF WAYNE L. PARKER, PLAINTIFF PLANNED PARENTHOOD, PLAINTIFF PLANETOUT.COM, PLAINTIFF JEFFERY POLLOCK, PLAINTIFF SAFERSEX.ORG to DEFTS' MOTION TO DISMISS PLFFS' COMPLAINT, Certificate of Service. (mld) [Entry date 07/02/01]

* * * * *

7/13/01 - Reply Brief by DEFENDANT UNITED STATES, DEFENDANT FEDERAL COMMUNICATIO, DEFENDANT INSTITUTE OF MU

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

SEUM in support of their MOTION TO DISMISS PLFFS' COMPLAINT, Certificate of Service. (See case 01-1303, paper #23) (mld) [Entry date 07/13/01]

7/26/01 12 MEMORANDUM AND ORDER THAT DEFTS' MOTIONS TO DISMISS THE COMPLAINTS ARE DENIED. (SIGNED BY JUDGE JOHN P. FULLAM) 7/26/01 ENTERED AND COPIES MAILED AND FAXED. (mld) [Entry date 07/26/01]

* * * * *

9/10/01 15 Answer to Complaint by DEFENDANT UNITED STATES, DEFENDANT FEDERAL COMMUNICATION, DEFENDANT INSTITUTE OF MUSEUM, Certificate of Service. (mld) [Entry date 09/12/01]

* * * * *

11/15/01 17 ORDER THAT IN ACCORDANCE WITH THE UNDERSTANDINGS REACHED AT AT A TELEPHONE CONFERENCE HELD THIS DATE, THAT TRIAL OF THE

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

CASE IS SET TO COMMENCE ON 3/25/02; PLFFS' PRELIMINARY PRETRIAL MEMORANDUM WILL BE FILED BY 12/14/01; DEFTS' PRELIMINARY PRETRIAL MEMORANDUM WILL BE FILED BY 12/28/01. (SIGNED BY JUDGE JOHN P. FULLAM) 11/15/01 ENTERED AND COPIES MAILED AND FAXED BY CHAMBERS. (mld) [Entry date 11/15/01]

* * * * *

3/20/02 - Joint Trial Stipulations of all Parties. (FILED UNDER SEAL) (See case 01-1303, paper #47) (mld) [Entry date 03/21/02]

3/20/02 - Redacted Joint Trial Stipulations of all Parties, Certificate of Service. (See case 01-1303, paper #48) (mld) [Entry date 03/21/02]

* * * * *

3/26/02 - Minute entry re: Civil Non Jury Trial held 3/25/02 Day 1: Witnesses called and sworn: Motion of Non Party N2H2 for limited intervention: Motion GRANTED. (See paper #52 in 01-cv-1303). (jpd) [Entry date 03/26/02]

* * * * *

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

3/27/02 - Minute entry re: 3/26/02 civil non-jury trial day 2. Geoffrey Nunberg, resumes; witnesses sworn. (See case 01-1303, paper #54) (mld) [Entry date 03/27/02]

3/27/02 - Transcript for Trial held on 3/25/02. (See case 01-1303, paper #55) (mld) [Entry date 03/27/02]

3/29/02 - Minute entry: Non-Jury Trial resumes 3/28/02. (See 01-1303 paper #56) (td) [Entry date 03/29/02]

3/29/02 - Transcript for Trial held 3/26/02. (see 01-1303 paper #57) (td) [Entry date 03/29/02]

3/29/02 - Minute entry re: Non Jury Trial Day 3 held 3/28/02: Witnesses called and sworn. (See Paper #58 in 01-cv-1303) (jpd) [Entry date 03/29/02]

3/29/02 - Minute entry re: Non Jury Trial Day 4 held 3/29/02: Witnesses called and sworn. (See paper #59 in 01-1303) (jpd) [Entry date 03/29/02]

3/29/02 - Minute entry re: 3/29/02 non-jury trial, day 4. Remarks by the Court; witnesses called and sworn in DEFTS' case. (See case 01-1303, paper #60) (mld) [Entry date 04/01/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

4/1/02 - Minute entry re: 4/1/02 non-jury trial, day 5. Remarks by the Court; witnesses called and sworn in DEFTS' case. (See case 01-1303, paper #61) (mld) [Entry date 04/01/02]

4/2/02 - Transcript for Trial held 3/28/02. (See case 01-1303, paper #62) (mld) [Entry date 04/03/02]

4/2/02 - Transcript for Trial held 3/29/02. (See case 01-1303, paper #63) (mld) [Entry date 04/03/02]

4/2/02 - Transcript for Trial held 4/1/02. (See case 01-1303, paper #64) (mld) [Entry date 04/03/02]

4/2/02 - Minute entry re: 4/2/02 non-jury trial, day 6. Remarks by the Court; PLFFS' case resumes; witnesses called and sworn in PLFFS' case. (See case 01-1303, paper #65) (mld) [Entry date 04/03/02]

4/3/02 - Minute entry re: 4/3/02 non-jury trial, day 7. Remarks by the Court; witnesses called and sworn in DEFTS' case; exhibits moved into

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

evidence by Atty BHATTACHARYYA on behalf of DEFTS; DEFTS rest; PLFFS rest. (See case 01-1303, paper #66) (mld) [Entry date 04/03/02]

4/3/02 - Transcript for Trial held 4/2/02. (See case 01-1303, paper #67) (mld) [Entry date 04/04/02]

4/4/02 - Minute entry re: 4/4/02 non-jury trial, day 8. Remarks by the Court; closing remarks by Attys SMITH and HANSEN on behalf of PLFFS; closing remarks by Atty BHATTACHARYYA on behalf of DEFTS; rebuttal by Atty SMITH. By the Court: Counsel for both sides to submit to Court Findings of Facts and Briefs by 4/11/02. (See case 01-1303, paper #68) (mld) [Entry date 04/05/02] [Edit date 04/05/02]

4/5/02 - Transcript for Trial held 4/3/02. (See case 01-1303, paper #69) (mld) [Entry date 04/05/02]

4/9/02 - Transcript for Trial held 4/4/02. (See case 01-1303, paper #70) (mld) [Entry date 04/09/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

4/11/02 - Joint Post-Trial Brief by PLAINTIFF MULTNOMAH COUNTY, PLAINTIFF CONNECTICUT LIBRARY, PLAINTIFF MAINE LIBRARY ASSOC., PLAINTIFF SANTA CRUZ PUBLIC, PLAINTIFF SOUTH CENTRAL LIBRAR, PLAINTIFF WESTCHESTER LIBRARY, PLAINTIFF WISCONSIN LIBRARY, PLAINTIFF MARK BROWN, PLAINTIFF SHERRON DIXON, PLAINTIFF JAMES GERINGER, PLAINTIFF MARNIQUE TYNESHA OVERBY, PLAINTIFF EMMALYN ROOD, PLAINTIFF WILLIAM J. ROSENBAUM, PLAINTIFF CAROLYN C. WILLIAMS, PLAINTIFF QUIANA WILLIAMS, PLAINTIFF AFRAIDTOASK.COM, PLAINTIFF ALAN GUTTMACHER INST, PLAINTIFF ETHAN INTERACTIVE, PLAINTIFF NATURIST ACTION, PLAINTIFF WAYNE L. PARKER, PLAINTIFF PLANNED PARENTHOOD, PLAINTIFF PLANETOUT.COM, PLAINTIFF JEFFERY POLLOCK, PLAINTIFF SAFERSEX.ORG, Certificate of Service. (See case 01-1303, paper #71) (mld) [Entry date 04/12/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

4/11/02 - Joint Proposed Findings of Facts by PLAINTIFF MULTNOMAH COUNTY, PLAINTIFF CONNECTICUT LIBRARY, PLAINTIFF MAINE LIBRARY ASSOC., PLAINTIFF SANTA CRUZ PUBLIC, PLAINTIFF SOUTH CENTRAL LIBRAR, PLAINTIFF WESTCHESTER LIBRARY, PLAINTIFF WISCONSIN LIBRARY, PLAINTIFF MARK BROWN, PLAINTIFF SHERRON DIXON, PLAINTIFF JAMES GERINGER, PLAINTIFF MARNIQUE TYNESHA OVERBY, PLAINTIFF EMMALYN ROOD, PLAINTIFF WILLIAM J. ROSENBAUM, PLAINTIFF CAROLYN C. WILLIAMS, PLAINTIFF QUIANA WILLIAMS, PLAINTIFF AFRAID TOASK.COM, PLAINTIFF ALAN GUTTMACHER INST, PLAINTIFF ETHAN INTERACTIVE, PLAINTIFF NATURIST ACTION, PLAINTIFF WAYNE L. PARKER, PLAINTIFF PLANNED PARENTHOOD, PLAINTIFF PLANETOUT.COM, PLAINTIFF JEFFERY POLLOCK,

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

PLAINTIFF SAFERSEX.ORG, Certificate of Service. (See case 01-1303, paper #72) (mld) [Entry date 04/12/02]

4/12/02 - Post-Trial Brief by DEFENDANT UNITED STATES, DEFENDANT FEDERAL COMMUNICATIO, DEFENDANT INSTITUTE OF MUSEUM, Certificate of Service. (See case 01-1303, paper #73) (mld) [Entry date 04/12/02]

4/12/02 - Proposed Findings of Facts by DEFENDANT UNITED STATES, DEFENDANT FEDERAL COMMUNICATIO, DEFENDANT INSTITUTE OF MUSEUM, Certificate of Service. (See case 01-1303, paper #74) (mld) [Entry date 04/12/02]

* * * * *

5/31/02 22 OPINION AND ORDER DATED MAY 31, 2002 THAT JUDGMENT IS ENTERED IN FAVOR THE PLFFS AND AGAINST THE DEFTS, DECLARING THAT 1712(a)(2) AND 1721(b) OF THE CHILDREN'S INTERNET PROTECTION ACT, 20 U.S.C. 9134(f) AND 47 U.S.C. 254(h)(6), ARE

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

FACIALLY INVALID UNDER THE FIRST AMENDMENT TO THE UNITED STATES CONSTITUTION; AND THE UNITED STATES, MICHAEL POWELL, IN HIS OFFICIAL CAPACITY AS CHAIRMAN OF THE FEDERAL COMMUNICATIONS COMMISSION, THE FEDERAL COMMUNICATIONS COMMISSION, BEVERLY SHEPPARD, IN HER OFFICIAL CAPACITY AS ACTING DIRECTOR OF THE INSTITUTE OF MUSEUM AND LIBRARY SERVICES, AND THE INSTITUTE OF MUSEUM AND LIBRARY SERVICES ARE PERMANENTLY ENJOINED FROM WITHHOLDING FEDERAL FUNDS FROM ANY PUBLIC LIBRARY FOR FAILURE TO COMPLY WITH 1712(a)(2) AND 1721(b) OF THE CHILDREN'S INTERNET PROTECTION ACT, 20 U.S.C. 9134(f) AND 47 U.S.C 254(h)(6). (SIGNED BY CHIEF JUDGE EDWARD R. BECKER, JUDGE JOHN P. FULLAM, JUDGE HARVEY BARTLE, III) 5/31/02 ENTERED AND COPIES MAILED. (mc) [Entry date 05/31/02]

________________________________________________

DOCKET

DATE NUMBER DOCKET ENTRY

6/20/02 - Notice of appeal to the UNITED STATES SUPREME COURT by DEFENDANT UNITED STATES, DEFENDANT FEDERAL COMMUNICATION, DEFENDANT INSTITUTE OF MUSEUM. Copies to: JUDGE JOHN P. FULLAM, JUDGE HARVEY BARTLE, III, USCA CHIEF JUDGE EDWARD R. BECKER, and JANET M. LARUE, ANDREA GACKI, LISA M. BORNSTEIN, TIMOTHY ZICK, THEODORE C. HIRT, ANNETTA FOSTER GIVHAN, SCOTT A. COFFINA, KEVIN S. BANKSTON, LEE TIEN, DAVID L. SOBEL, FRANK SCIBILIA, ANDREW L. LEE, STEFANIE S. KRAUS, CHARLES F. SIMMS, MEERA DEO, CHRISTOPHER A. HANSEN, STEFAN PRESSER, ANN ELIZABETH BEESON, Certificate of Service. (See case 01-1303, paper #80) (mld) [Entry date 06/20/02] [Edit date 06/20/02]

* * * * *

 

UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF PENNSYLVANIA

Civil Action No. 01-1303, 01-1322

AMERICAN LIBRARY ASSOCIATION, ET AL , PLAINTIFFS

v.

UNITED STATES, ET AL, DEFENDANTS

TRANSCRIPT OF TRIAL
BEFORE THE HONORABLE EDWARD R. BECKER
BEFORE THE HONORABLE JOHN P. FULLAM
BEFORE THE HONORABLE HARVEY R. BARTLE
UNITED STATES DISTRICT JUDGES

TRIAL TRANSCRIPT EXCERPTS

[TESTIMONY OF C. MORGAN]
[March 25, 2002]
[pp. 9-50]

* * * * *

Q How large is the Fort Vancouver Regional Library System?

A We're located in Southwest Washington, directly across the Columbia River from Portland, Oregon. We cover 4,200 square miles, which spans the Cascade Mountains. We serve 12 cities, three counties, the unincorporated areas of three counties, and we serve 372,000 people.

Q Okay. And how many library cards are issued by the system?

A 238,000 registered borrowers.

Q What resources does the system provide to those patrons?

A We provide access to over 700,000 materials of books, periodicals, videos, pamphlets, documents. We provide reference service, both at-at all of our branches and on the bookmobiles, and a backup reference service from our district headquarters.

We provide internet service, word processing, CD ROM base story stations for young children, programming for all ages, electronic databases that feature over 2,000 magazines, full text, as well as a number of encyclopedias and specialized reference books.

We provide, I think I said programming for all ages, and we provide a various number of outreach services.

Q Ms. Morgan, I think you're going to need to slow down a little bit.

A All right. Sorry, yes.

Q Now, does the library system have a mission statement?

JUDGE BECKER: I think she's talking just fine.

MR. SMITH: I was worried about the court reporter, not Your Honor.

JUDGE BECKER: She talks with celerity and clarity.

JUDGE BARTLE: This is a sound recording system.

MR. SMITH: Okay. Thank you.

BY MR. SMITH:

Q Does the library system have a mission statement?

A Yes, we do.

Q And can you briefly summarize what the mission is of the library?

A Yes. The mission of the Fort Vancouver Regional Library District has two major themes. First is the library will provide a diverse and unbiased source of information for all the individuals it serves, and secondly that it will protect the rights of each individual to make his or her own choice about what materials, information, or services to access.

Q And I believe that you have in front of you three of our exhibits, Ms. Morgan. Can you refer to plaintiffs' Exhibit 61 which is our volume 3 of the plaintiffs' exhibits?

And is the mission statement on page two of this composite Exhibit 61?

A Yes, it is. Yes, it is.

Q Now, does the library system have a policy on selection or collection of materials in the library's collection?

A Yes, we do.

Q And is that on page 83 of the same exhibit, referring to the numbers that are CM000083?

A Yes, it is.

Q And can you briefly-

JUDGE BECKER: What page is it?

MR. SMITH: 83.

JUDGE BECKER: 83, I'm sorry.

BY MR. SMITH:

Q Can you summarize for us the basic policies of collection developments of the system?

A Yes. The overall part of the policy mirrors the mission statement in that it says that Fort Vancouver Regional Library will select and maintain the collection that allows each individual to make choices about which materials to select, and that it is an inclusive policy.

Materials will be included based on the anticipated and express needs of the patrons that we serve, and materials will not be excluded solely on the basis of either content or viewpoint.

Q Now, you referred to the anticipated and express needs of the patrons. Can you tell us how you anticipate the needs of your patrons?

A Yes. We have our selectors, most of them, actually work on the public service floor, so they become very familiar with the types of questions that patrons are asking and the types of materials that they are checking out.

We have patron comment sheets in each of our libraries, oftentimes comments about the collection appear on there. We have a pamphlet that asks people to let us know what they would like to see in their library in which they can do a broader coverage in relationship to that.

We have youth statistics that are both based on types of materials. Our collection is automated and so we can categorize materials and then tell how often they circulate. Plus individual titles in our automated system also tell us how often they circulate in a year.

So all that information is available to those who are selecting the materials.

Q Now, under your library system's policy, are materials ever excluded because of their controversial content?

A No. In fact, the selection policy specifically addresses that in which it indicates that materials of a controversial nature or that which might be objectionable to some members of our public will be included based on what role they would play in making-in having a diverse collection, with a variety of subjects and diverse viewpoints.

Q Can you give us a couple of examples of some of the more controversial print materials that you have in the collection?

A Well, this book is no longer in the collection because it wore out, but Fort Vancouver Regional Library purchased Madonna Sex because we had had more previous request for that than any other book in our history, and that certainly was controversial material.

But, the kind of controversial materials we have are much broader than that. We have people who object to young children's books because they are perceived to challenge parental authority, children's books because they deal with themes related to gay families. We have-

JUDGE BECKER: Well, you've had objections to Harry Potter haven't you?

THE WITNESS: Actually, we have not had any objections to Harry Potter. It is one of the most challenged books, but not-we've not had any formal objections to Harry Potter in our system. But we have to Henny Penny because the family were apparently not familiar with folklore and the fact that it would teach children not to trust others.

JUDGE BECKER: What would your library do if a donor had a-prospective donor had a collection of hard core-what we might all agree was hard core pornography and offered to donate that to the Fort Vancouver Regional Library?

THE WITNESS: Well, with due respect, I ask-as a result of numerous discussions with people that use our library, I don't think there is a common definition of what is hard core pornography.

If somebody gave us something that we became aware had been declared legally obscene by a Court, if that's what you mean by hard core pornography, then we definitely would not add it. Otherwise we would have-we would look at any gift as we would anything that we select, about how it meets the criteria in our selection policy and what the anticipated needs or express needs of our patrons are.

We would not make the decision based on the content, but based on the item as its whole, and its place it might play in the collection.

JUDGE BECKER: Let's assume that the gift depicted kiddy porn, adults or children engaged in-children engaged in-

THE WITNESS: Child pornography-

JUDGE BECKER: -sexual acts or let's assume it was simply a collection of Hustler Magazine.

THE WITNESS: Well, for the first-in the first example you gave, child pornography is illegal and, of course, we would not add it to the collection.

The second example you gave is an interesting one. The library does not subscribe to Hustler, and magazines are in a different category. We don't not subscribe to Hustler because of its content. We don't subscribe to it because, in my knowledge, we've never been requested to subscribe to it. Magazines are-

JUDGE FULLAM: Nobody had the nerve to ask you to possibly.

THE WITNESS: Well, we have a very nervy population, so I'm not sure that that's the case. But we could discuss that, I guess.

The magazines are selected by the branch librarians, not centrally, and we have very limited space for magazines. We are- we're very challenged space-wise in our library district with little funds to be able to do anything about that.

Consequently, to explain this perhaps more succinctly, the only part of our collection that I have not-ever gotten a complaint from a branch library about not having enough money for collection has been the periodicals. They could use many more than what they can actually house, and consequently, we do have Playboy, we do have Cosmopolitan. It seems that the combination of no request and limited space has been the major indicator in that.

So if somebody tried to give us a bunch of Hustlers, I will not second-guess what the branch librarian would say, but I suspect that they would say they don't have room for it.

BY MR. SMITH:

Q Can you give us some other examples of some sexually explicit materials that are in the print collection in the system?

A Yes. We have sex education manuals. We have books on how to enjoy or improve your sex life, like The Joy of Sex and The Joy of Gay Sex. We have literature, poetry, novels with sexual themes.

We have art books with sexual themes. Mable Thorpe's books for an example. And as I mentioned, we have Playboy and Cosmopolitan.

Q And do you, as a matter of policy, limit access to these materials in any way?

A No.

Q Does the library system have a policy, a formal policy on minors' access to materials?

A We do, and it's very long. What the board, library board decided to do was to document the process they went through for about eight months after the controversy erupted about Madonna's Sex book simultaneously with a discussion of the gay sex manuals in our library.

And the board thoroughly examined all aspects of minors' access, held numerous public hearings, heard comments from all over the library district, held hearings in various parts of the library district and came to the conclusion that the policy would be continued that each individual has free and open choice to our collection, that parents or guardians are responsible for how their children use the library.

But the library staff would provide as much help that we can to parents and for them to be able to make those decisions in terms of offering help in selecting materials, doing bibliographies, in any way that we could to help parents carry out that responsibility.

Q And does that policy appear in Exhibit 61 at page 43?

A Yes, it does.

Q Was that policy developed prior to the availability of internet in your library?

A Yes, it was.

Q Do you have inter-library loan services available in the library?

A Yes, we do.

Q Can you just briefly tell us what those are?

A Inter-library loan arrangements are arrangement agreements between libraries that we will lend our materials to other libraries for use of their patrons. It's based on the fact that not every library can have everything, and also that even if every library did have everything, things-or could, things go out of print, they get lost, they wear out.

And so, you may no longer be able to purchase something even if you wished to do so. But some other library may well have it.

Q And what libraries are available as potential places to borrow from?

A Libraries all over the world.

Q How does a patron make a request for those materials?

A We have a form that can be used either to request that we buy or get an inter-library loan so that staff can make the decision about which is the best way to approach that.

Plus, I mentioned our backup reference service that we have. If a reference librarian is working on a question that has been referred and the only way to answer that question is from a book borrowed from another library, that librarian will initiate an inter-library loan request on behalf of the patron.

Q Are there limits on the content of the materials that you will borrow for a patron through inter-library loans?

A No. And our information to patrons about inter-library loan is that we will attempt to borrow anything the library does not own.

Q Would the library system borrow Hustler Magazine for someone?

A We would-the circumstances with magazines are that most libraries do not lend whole issues of periodicals. But they do photocopy individual articles from magazines.

So we would ask that patron if they had an individual article that they would like to see copied, and if so, then we would initiate, because there are libraries that own Hustler, so in all likelihood, we would be able to fill that request.

Q Does the system have any policy on the confidentiality of patron records?

A Yes, it does.

Q And does that appear on page 28 of Exhibit 61?

A Yes, it does.

JUDGE FULLAM: We'll take your word for it.

BY MR. SMITH:

Q Okay. And can you describe briefly what the policy is?

A And you'll notice it's a two-part, having been confirmed later after the first part. The first part was adopted in 1981, or '75 and then '81. And Fort Vancouver Regional Library follows Washington State laws requirements that we will not release any information that connects a library user with use of the library.

The later confirmation came as we were preparing to automate. We had to get an election to automate, and our patrons expressed great concern about automation having a potential threat to their both privacy and confidentiality because we didn't have library cards before we automated.

Q When you say automation, what are you referring to?

A I mean putting our catalog into computer format. And we didn't have library cards before, so it was a huge jump for us. And they were worried that would collect a lot of information from them, therefore violating their privacy, or reveal it to people other than what we needed for library business, and we assured them we would not.

So we reconfirmed the policy.

Q And do you currently provide internet access to patrons at the library?

A Yes, we do.

Q And when did that begin?

A 1995.

Q And how many terminals are available for patrons to use?

A Sixty.

Q How many are in the main branch?

A Twelve-excuse me, 14.

Q Now, why did the library system decide to provide internet access?

A Well, our mission says that we will serve as a diverse and unbiased source of information, and space and budget limitations affected how well we can do that. We have very, very small branches outside of our major branch, very rural and very isolated.

The internet was viewed by our library board and staff as being finally the way that we could meet this pledge and provide at basically no more budget-not much more budget costs by just purchasing the whole service throughout the library district. So that's why we decided to do that.

Oh, there's another reason, too. Although we used our backup reference service to respond and, we feel, fill most of the information requests of our patrons throughout our district, there were time delays for the referral. And also we took away some choice from the individuals who were asking the questions because we would search for the best answer and send it to them, we were making decisions for them based on our reference interviews.

Now more people had a choice to be able to do their own searches for information.

Q And does the system receive e-rate and IMLS funding from the Federal Government for those connections?

A Yes, we do.

Q And does that money play a significant role in making this possible?

A The IMLS moneys are the most significant to us. We get over a hundred-we have gotten since-for three years, 1999 through 2000, over $135,000. And what-

JUDGE BARTLE: Is that each year or total?

THE WITNESS: Total.

JUDGE BARTLE: How much again?

THE WITNESS: 135,000.

JUDGE BECKER: And that's a three-year period?

THE WITNESS: Yes, it is.

Now, whereas that may not seem to be an extraordinarily large amount of money, when you operate totally on a property tax and you have a need to upgrade your technology to be able to serve 400,200 square miles, you need an amount of capital outlay to put right in right now. And that has been difficult for us.

And consequently, the ability to apply for these funds is a way that we've been able to keep out technology up to date and being able to continue to provide this very important service to our patrons.

JUDGE BECKER: Your board is a public board, I take it.

THE WITNESS: Yes, it is.

JUDGE BECKER: How is a board selected?

THE WITNESS: We have seven board members. They are appointed by the elected county commissioners. There's one each in the two more rural counties, and the rest of them come from Clark County, the most populous county.

JUDGE BARTLE: What is your total library budget for each year?

THE WITNESS: Fourteen million, approximately.

JUDGE BARTLE: That's for each year?

THE WITNESS: Yes, it is.

JUDGE BECKER: So you're getting $40,000 a year out of 14 million?

THE WITNESS: Yes, as I said.

Also we have the e-rates, we get e-rates, and we have about $6,500 for internet service providers, and another 53,000 for telecommunications of which about 25,000 is for frame relay, which is the basic backbone of our providing our electronic services.

JUDGE BECKER: Now is that a fund or is that the discount, isn't it, the e-rate?

THE WITNESS: Yes, it is.

JUDGE BECKER: So that what you're identifying is what you save as a result of the e-rate discount?

THE WITNESS: Yes. Thank you, yes.

JUDGE BARTLE: So what's the total savings and amount of money actually received from you? Is that the 130 you're talking?

THE WITNESS: All right. Well, there are two different kinds of funds involved. And the first amount of fund that I discussed was the LSEA grants. And those don't come automatically, you apply for them through the state library.

So we have applied for three-for a total of four grants over three years, 1999 to 2001. And those grants specifically for helping us to provide the internet service.

JUDGE BARTLE: And how much was that?

THE WITNESS: They totaled-oh, it was 135,000 plus.

JUDGE BARTLE: So that's the 135,000-

THE WITNESS: Right.

JUDGE BARTLE: -over three years?

THE WITNESS: That is correct.

JUDGE BARTLE: All right. Now, what else did you receive in the way of discounts or subsidies?

THE WITNESS: All right. That's the e-rate funds that we spoke of.

JUDGE BARTLE: Right.

THE WITNESS: And we received $6,500 for, discount again, for our internet service provider fees.

JUDGE BARTLE: That's for a single year or for over three years?

THE WITNESS: Yes-no, that's a single, each year.

JUDGE BARTLE: Okay. So that's roughly 19,500.

THE WITNESS: That's right.

JUDGE BARTLE: Over that same three-year period, and that's in terms of discounts?

THE WITNESS: That's right.

JUDGE BARTLE: So the total then for the three-year period is roughly $155,000.

THE WITNESS: That is correct. As we understand how the e-rate will be applied by the Federal Communications Commission, no part of what they call their telecommunications budget will be covered. And that, of course, since this hasn't happened yet, it hasn't really- in fact, we're not certain-but of that, 53,000 in addition to everything I've told you before that we get from e-rate, about half of that goes for frame relay service, the primary purpose of which is to provide internet service.

JUDGE BARTLE: So what the plaintiff is saying is that if this library-regional library doesn't participate under the statutes, they would lose, in effect, $155,000 over a three-year period, or would.

MR. SMITH: True in this example, yes, Your Honor.

JUDGE BARTLE: Yes, this example.

MR. SMITH: I think the evidence is going to show that there's quite a variation in the amounts from library to library.

JUDGE BARTLE: Oh, I understand that. But in this case.

THE WITNESS: That is correct.

BY MR. SMITH:

Q Now, you have an internet use policy that you developed at the library?

A Yes, we do.

Q And is that on page 13 of Exhibit 61?

A Yes, it is. We call it a policy on access to electronic information because it covers access to our electronic resources as well as to the worldwide web.

Q And can you just briefly summarize the policy that you developed when you began providing internet services?

A Yes.

JUDGE BARTLE: Now are you talking about the CM number now or are you-

MR. SMITH: Yes, sir. Yes, Your Honor.

JUDGE FULLAM: So it's CM16.

MR. SMITH: I've been talking about the CM number throughout.

JUDGE BARTLE: Right. So you come back, now CM16?

MR. SMITH: 13, I believe.

JUDGE BARTLE: 13, all right.

MR. SMITH: I apologize if I misspoke.

JUDGE BARTLE: That's all right.

THE WITNESS: The policy starts out with a statement that the advent of electronic resources does not change the library board's policy in relationship to free, open, and equal access to information, as is mirrored throughout all of our policy.

It states that filtering will be mandated on no one, but that choices will be offered about how people search the internet. It states that parents and guardians continue to be responsible for their children's use of this library service. It indicates that the library will search for technological means to assist parents in being able to exercise that responsibility.

It states also that the patron, the internet searcher is responsible for not using the library's resources, electronic resources, as with all other resources to access illegal material or to use the service in any way that is illegal, like hacking or copyright violations, et cetera.

BY MR. SMITH:

Q Okay. And do you, in fact, make available internet blocking software or filtering software as an option for patrons?

A Yes, we do.

Q Okay. Let me just ask you quickly to look at Exhibit 62. Is this the document that you make available to patrons to discuss that option?

A 62?

Q Exhibit 62 would be-not page 62, the next exhibit, which would be the next folder, Ms. Morgan.

A Oh, I'm sorry. Oh, Exhibit 62.

Q Yes.

A This is a description of-we get many, many questions. Obviously, this is much too much to expect a library patron to read as they're anxiously trying to register for the internet, so that our staff informs the patron of what the policy is and how they can sign up for the policy.

Q Okay. And can you explain to us mechanically how this choice of filtering works in the library?

A Yes. It's based on a library card and the PIN number, which is the means of additionally protecting the confidentiality of that library card number. And a patron who wishes to use the internet registers one time and they enter their library card and their PIN number.

They select at that point whether they want to search with a filtered or unfiltered desk top. They can password that choice. If they password it, they only know the password. There is no way that staff can get access to that so that no one can change that default.

In addition, we say, as we do with all of our resources, that a parent that gets their child's library card, can also come into the library one time, make that choice for their child, password it, not tell their child what the password is, which means it cannot be changed.

Then, when you are ready to search the internet after you've reserved a terminal, you log-in with your library card and the password, the PIN, and it knows, the computer then brings up the desk top filtered or unfiltered.

Now, there is an additional choice that parents can exercise. They can choose no internet access also, which means should a child try to log-in in using their bar code and their PIN, they would be blocked, they could not use the internet at all.

Q And the third option, of course, is no filtering at all, is that right?

A Filtered or unfiltered, yes.

Q Okay. And how did you arrive as a system at the decision to make blocking software available as an option?

A Well, we began internet service in 1995 and we started with saying that our current policies in terms of access would apply while the board and staff studied what the issues would be in relationship to the internet. We didn't want to jump to judgment

They then held a series of several public hearings, elicited comments via e-mail, fax, in person, in writing from the public concerning how they would handle the internet. As with every single one of Fort Vancouver Regional Library District's public processes, there is a great diversity of opinion, no clear majority, as well as, always a few more people that will speak out in terms of no restrictions at all.

But our board then went back to the missions statement where it indicated that individuals should have a choice and they directed us to figure out how to do that. We started with having filtered search engines available that are generally available on the web as we searched for software that would give us the opportunity. And we actually had the software we use developed just for us.

We had purchased internet management software for time limits because that's our biggest problem, the time limits. And we asked that company to develop a process by which we could offer the options that I just mentioned. So they developed it for us specifically.

JUDGE BECKER: By time limits, you mean limiting the library patron's internet access to a half hour or what -

THE WITNESS: An hour a day is what we do.

BY MR. SMITH:

Q And why is that your biggest problem?

A Well, because of space limitations, we have not enough terminals, and the patrons fight with staff and with each other about whether it's whose turn on the internet or whether they have exceeded their time limits.

The time limits work automatically and shut down, but people have very creative reasons why that shouldn't have happened to them.

Q Can you tell us what categories of web sites are blocked if the patron chooses to use the filtered option?

A We use Web Sense, and Web Sense offers the option of over 50 categories. We have chosen 14 that reflect what we heard during the hearings about people who said I would like my children not to have access to, or I myself would not like to have access to.

So there are a variety of adult sexual, nudity, violence, gambling, illegal activity, hacking, hate groups, that's sort of the nature of what are involved in those categories.

Q And can you tell us how many people have chosen each of the three options?

A Yes. This has been pretty consistent throughout -from when we started it. Eighty percent have chosen unfiltered, 19.9 percent have chosen filtered, and .1 percent currently have chosen no internet access.

JUDGE BECKER: Does this include parents choosing for their children?

THE WITNESS: We assume. This is a self - people register and we don't ask who they are or who they're registering for as long as they have the library card.

JUDGE BECKER: Do the children get their own PIN?

THE WITNESS: The PIN is a default of the last four digits of their phone number, although a patron can change their PIN themselves in any online catalog.

BY MR. SMITH:

Q Now, is it possible for a parent to make a choice that the minor can't change?

A Yes. The parent needs only to password it and not to tell the child, and even a lot of persuasion to staff wouldn't work because we don't-we don't see that password, it's hidden.

JUDGE FULLAM: Does the child get a library card, or could a child get a library card without his parents knowing?

THE WITNESS: Yes. Fort Vancouver Regional Library does not require a parental signature to get a library card.

JUDGE FULLAM: So this 80 percent who chose unfiltered, do you have any notion of how many are children and how many are not?

THE WITNESS: We don't. Age is optional in our patron record because one of the things we heard when we automated about privacy was you aren't going to make us give you our age, are you? We have it there as an option so that we can apply for grants that say how many kids you choose, and I explode the figures based on the percentage, but we have no way to know.

BY MR. SMITH:

Q Okay. Now, why would somebody choose the option of having no internet access at all?

A Well, of course, we haven't interviewed, or I haven't interviewed everyone, but I have at least one anecdotal example.

In one of our branches, a family came in and said that-they thanked us for having that option, even though they knew it probably was not heavily used, because they have internet in their home and their family rules are that the children can only use the internet in their home.

Q Do you have any rules that would prevent people from using the internet to access e-mail or games, things of that sort?

A No.

Q Okay. Can you tell us what efforts you make to protect the privacy of patrons using the internet?

A Well, again, because we've heard continuously from our patrons about the importance of that, we use a combination of-well, right now we're using almost entirely what are called recessed monitors. What that means is the monitor is below the level of the table. So it's pretty hard to see what somebody is searching with that setup.

We also have used terminal placement and privacy screens.

Q What's a privacy screen?

A A privacy screen fits over the monitor and makes -the only place you can see the screen is directly in front of it. You can't see from the sides.

Q Okay. Do you also have rules on protecting the privacy of lists of sites that people have gone to?

A Yes. We don't keep them. What happens after you've completed your search session, everything shuts down whether you're ready to have it shut down or not. And the cache and the history is erased, so there's no more record and no possibility that anything will remain when the next person comes, which protects the person from being inadvertently exposed to something the person has seen before.

The sign-up-the fact that you have used the internet connected to-or the library card record of having used the internet remains till the end of the business day because that's the way we track the time. Our patrons will go from branch to branch otherwise.

And we-at that point, the log is automatically erased and it starts anew. All of the timing is reset for the next day.

Q Does the system suggest web sites the patrons might want to access?

A We have a home page, and on that home page

our reference librarians and our librarians and our children's and young adult coordinators have selected sites that we know from our experience are in the most demand. This is a place where we can also point to local resources, chambers of commerce, local organizations in a way that we can't with any of our other resources.

We have a teen page, a kids' page, and a variety of subject pages.

JUDGE BECKER: Do you recommend a particular search engine, or do you just-

THE WITNESS: We have a portion where you do internet search and we have a whole range of search engines. And we have carefully selected ones-we have some that are filtered, we have some Christian search engines, we have a Jewish search engine. We have family friendly search engines.

We've tried to reflect, again, the needs expressed by our patrons. Some of them go to only certain sites, some of them filter from their site. There is no default search engine. You choose when you're searching from our home page.

BY MR. SMITH:

Q Now, the internet policy that you've been describing, do you see that as consistent with the collection development policy on the print collection that you described before?

A Yes, it is. When we select things for the collection, they are selected, processed, cataloged, and after they're checked out they're returned to the shelf. So they remain in the library and we apply the criteria and the diversity, unbiased, et cetera that we talked about before.

When we select the internet, it's more like selecting a service in that we select it for what it provides, not the individual site. There's no way that we can select the individual site. But we do not-as with the print collection, we do not exclude anything on the basis of its controversial nature or its viewpoint or its content, and on the internet we don't also.

Q Now, you mentioned before that there were rules against patrons accessing illegal speech on the internet. How do you go about enforcing that rule?

A The policy specifically states that it is the patron's responsibility, similar to not violating copyright or using any of our materials in the regular collection in an illegal manner, that we provide the information, the patron is responsible for behavior.

The presumption is that they will follow that rule. It would be next to impossible for our staff to monitor that, given the recessed monitors. And we're a very, very busy library. Even our small libraries are. We have barely enough staff to provide public service.

So, in addition to that, particularly in relationship to obscenity, our staff don't have the knowledge to serve as if they were judge or jury in making those decisions.

Q And if it came to your attention that a patron was disobeying the law accessing something that was clearly illegal, what would the response be?

A If it was-if we knew it was clearly illegal and it came to our attention, then they would be breaking our rules of conduct and we would ask them to stop. If they refused to stop, we would have to call the police because in addition to breaking the rule about illegal materials, the patron would be not cooperating with staff, and that, as I said, would have to result in some action.

Q And have you had occasions where patrons have complained to the librarians about other patrons viewing sexually explicit web sites?

A We have. It's not a whole lot, but yes, we have.

Q And have you also had some issues with behavior of patrons connected or viewing sexually explicit sites?

A We have. Again, in the context of the total number of behavior problems that we have in the library, it is minuscule.

Q Now, I'm going to ask you to get Exhibit 63 out.

A Okay.

JUDGE BECKER: You mean in terms of other behavior problems, loud noise and food and that kind of thing?

THE WITNESS: Well, loud noise, and actually we have more sort of sexually related incidents in the library that are totally unrelated to the internet than for the internet.

We're a public place. We're a public building. And when you have that and it's a come-one, come-all, can't exclude except on the basis of behavior, it happens.

JUDGE BECKER: What is the largest city in the Fort Vancouver Regional Library District?

THE WITNESS: It is the City of Vancouver, and it's the fourth largest city in Vancouver.

JUDGE BECKER: You mean Oregon? You're Washington, not Oregon.

THE WITNESS: No, in Washington.

JUDGE BECKER: In Washington.

THE WITNESS: Yes, yes, right, in Washington.

JUDGE BECKER: What's the population of Vancouver?

THE WITNESS: Oh, it's a couple hundred thousand.

JUDGE BECKER: Is the rest of the district rural, urban-suburban, rural?

THE WITNESS: There's some suburban, but most of it is very rural.

To kind of give you an idea, most people of heard of Mount Saint Helens, the southern flanks of Mount Saint Helens are in our library district. The Columbia River scenic gorge, the entire Oregon side-excuse me, Washington side of the scenic area is in Fort Vancouver Regional Library District. Mount Adams also is in the-the southern flanks are in Fort Vancouver Regional Library District.

JUDGE BECKER: And it goes Pacific Ocean on the west or-

THE WITNESS: No. Vancouver is the furthest west of the cities. And it goes across the crest of the Cascade.

BY MR. SMITH:

Q Exhibit 63, this is a chart you prepared?

A Yes, it is.

Q And I wonder-first of all, it says Incident Reports, 1995 through October 18th, 2001. Is that correct?

A No. Let me explain a little bit about this chart. I prepared it on the plane before I was being deposed just to give myself a context. At that time that was the date.

I revised it to include up through 2000, and so I forgot about the headings.

Q So it actually covers through the end of 2001?

A It does. It does.

JUDGE FULLAM: To the end of 2001?

THE WITNESS: Yes, it does.

BY MR. SMITH:

Q Okay. And if you could just explain to us each of the horizontal lines and what those figures represent, I take it the columns represent the years. But what is the total report, what do you mean by reports?

A We ask our staff to fill out what we call an incident report anytime anything unusual happens in the library. We do this both so that administration will be alerted, and also we use it when patterns develop to indicate that perhaps we need to address a problem that is more than just isolated.

So what we have here are a category of issues that are related to the subjects we're discussing, and the percentage they represent of the total reports received that year.

Q Okay. And the next line down then, it says Materials Displays, what are those figures?

A That is when a patron complains to staff about the regular collection materials, the books, periodicals, et cetera. We also have display areas that are available for use by the public, and sometimes patrons complain about the content of those displays.

So that's what falls-this does not include formal complaints about library materials that come in written into our appeal process.

Q Okay. Now, the next line says Internet Materials, number percent. What is that about?

A That's about-the question you asked me just previous to introducing this, and that is when people complain about what other people are searching on the internet.

Q Okay. So on average how many complaints like that do you get a year?

A Ten a year. Out of a total of-average of 399 a year of incident reports totally.

Q Okay. And the third row is Other Internet. What is that?

A Well, most of this has to do with time limits, or this fellow here came and took my reservation time. "Your system doesn't work at all well and cut me off before my time and you should give me more time."

Q And the last two rows have to do with sexually related behavior?

A Right. And I broadly defined this to include the leaving of printouts in a public place and that's what this means.

Q Okay. So the internet-

JUDGE BECKER: Would you explain that a little more. I'm not quite sure I understand the-

THE WITNESS: Well, somebody may print something off the internet that has sexual content, and leave it someplace in the library where somebody else can find it.

JUDGE BECKER: I see.

THE WITNESS: So the patron who finds it-

THE COURT: So, if they take it themselves there would be an incident?

THE WITNESS: No, not, the issue is-I was kind of using the-we kind of used the guidelines of what somebody might feel harassed, and finding you find objectionable feels harassing and the person should take their own printouts home, not leave them some place.

JUDGE BECKER: So, what happens if somebody leaves them, somebody is offended and then they go to the librarian and they-

THE WITNESS: Yes, that's correct.

JUDGE BECKER: And that becomes an incident-

THE WITNESS: That becomes an incident-

JUDGE BECKER: -report?

THE WITNESS: -which we average two a year.

BY MR. SMITH:

Q It would show up in the box second to the bottom row here-

A Yes, that's right.

Q -because it internet related.

A Yes.

Q What other kinds of behaviors sexually related behaviors have you had that are internet related?

A Well, we have had one case of a patron masturbating, and we have had a couple of cases of inappropriate remarks made to someone while someone is searching the internet. Maybe to the person next to them or something like that.

Q And the bottom row, what is that?

A Oh, that's-those are things with some sexual content that are not internet related.

This has to do again with inappropriate comments with showing other materials to someone who doesn't want to see them that has some sexual content. To asking staff for dates, to well, we did have a stalker on one of our staff That's the number in-I think-we think we had a case like that in 1998, I'm not sure but we've had a number of incidence like that. Well, not a number, but we've had a few incidents like that.

JUDGE FULLAM: Excuse me, but I-

THE WITNESS: Somebody kissing somebody who didn't want to be kissed. That happened once.

JUDGE FULLAM: -this report says you had a total of 399 incidents in 2001.

THE WITNESS: That's the average.

JUDGE FULLAM: And you got -

THE WITNESS: I'm sorry, yes, 399.

JUDGE FULLAM: -62 specific types, what are the other incidents?

THE WITNESS: The other incidents are dogs left in the parking lot automobiles vandalized.

JUDGE FULLAM: Unrelated to the internet.

THE WITNESS: Not related to the internet at all.

JUDGE FULLAM: Okay.

BY MR. SMITH:

Q Not related to sexual behavior-

A No, they-

Q -of any sort?

A -mostly have to do with managing 12 public buildings and three book mobiles.

Q Okay. Can you give us in the chronology, when in this five years span or six years span did you introduce the recess monitors?

A Late in 1999.

Q And what kind of impact did that have on the volume of complaints from patrons about other patrons we surfing activities

A Well, simultaneously, as we added terminals which we added quite a few in 99 we got more and more internet users, but as you will notice the internet material complaints have not gone up.

In fact, they actually reduced from in 1999 it was like in September that we started putting them in, and it was down again to seven in 2000. It was up again to ten in 2001 but our number of internet users had increased from-significantly.

Q Now, do you consider the overall policies that you described that are in place at the Fort Vancouver system to be a success?

A Yes, we average about two, 2200 search internet session a week, open 540 hours. For this number of incidents that we have had plus generally very positive comments that we've received from patrons about the wondrous things that they have found on the internet. The thanks we get for offering choices and how to search. Yes, we feel it has been a success.

Q And has the system made a decision yet about whether it would come into compliance with the Children's Internet Protection Act if that question is ultimately posed to it?

A No, we have not made that decision. We have been following the FCC's guidelines for taking actions and considering the issue.

Q Now, do you have concerns about the Children's Internet Protection Act?

A I do.

Q And can you tell the Court what concerns you have about the requirements of that statute?

A Fort Vancouver Regional Library has a long history of serving-of dealing with it's-the juxta- position of serving the community being supported by the community and being an institution that is a public forum and needs to protect individual liberties.

We carry that out by study of the legal principles, and by a public process. We have had multiple public processes, and each time that we reach an approach to handling this juxtaposition this tension, we have reached it with consultation with the community.

As I mentioned we had to go out and buy our own software to reach the conclusion that we found from the community with electronic access. This Children's Internet Protection Act would impose upon us a congressional choice which is contrary to that.

Furthermore, since filters are not-there is no filter that can block out only constitutionally protected speech-

Q Unprotected speech.

A -I'm sorry, yes. Constitutionally-

JUDGE BARTLE: You would find a filter that could-

THE WITNESS: -yes, you're right about that.

JUDGE BARTLE: -eliminate all constitutions?

THE WITNESS: Slip of the tongue I apologize.

Yes, you cannot find a filter that wil only filter out unconstitutional speech and so for the first time in Forty Vancouver Regional Library's history we'd be in a position of restricting access to some constitutionally protected materials by us.

We would be mandated-we would be doing that by installing the filters and making them mandatory and it would be on our entire population adults children and staff.

BY MR. SMITH:

Q Okay. Now, the statute would allow you to disable the filter for adults who came to you and said I have a legitimate reason to want access on a unfiltered basis, can you tell us why that doesn't alleviate your concern?

A Yes, I asked my staff what they would expect from me should we do this, and they indicated as they have in all cases that of course I would provide them with guidelines that would make it clear about when they were disabled.

I have been unable to find a definition of bona fide research and even the definition of legal is a little difficulty when you're dealing with what's legally obscene and not.

Consequently I'm not sure how we could provide equitable provision of the disabling. In addition to that, you can disable only for adults. I recently hired a women who had been working for years in a school media center in one of our communities who as a part of her interview told us that the students are frequently blocked from-

MR. ZICK: Objection to that, Your Honor, move to strike. It's hearsay.

JUDGE BECKER: It certainly is.

THE WITNESS: Okay.

So we have concern that constitutionally protected speech would be blocked, and that we would have no way to determine the basis by which we would disable.

We've actually just-the law does not require disabling, it only says may so we would have to decide whether we felt we could provide adequate guidelines in order to do that.

BY MR. SMITH:

Q And is it your view that patrons would exercise the option of asking for disabling?

MR. ZICK: Objection, calls for speculation.

JUDGE BECKER: What was the question?

MR. SMITH: I asked her if it was her view that patrons would exercise the option based on her experience with patrons over 37 years. Your Honor, it seems to me a question that she could answer.

JUDGE BECKER: Overruled, you may answer.

THE WITNESS: I must say however, that it's hard to speculate because we've never put our patrons in a position-

THE COURT: Okay. We'll sustain the objection. She sustained it.

THE WITNESS: Yes, well, let me explain why. I think my explanation is important.

That is that we have never out our patrons where they have to ask to have access. If they ask us to buy a book we buy it almost always buy it. If they ask us for an electronic research we'll give it consideration.

Consequently they have not been put in that position before, so I'm not sure how they would act.

BY MR. SMITH:

Q Would it be logistically possible with the system that you have in place to set up a system where patrons could ask for disabling on an anonymous basis?

A I think it would be totally unfunctional because as I mentioned before our patrons are incredibly creative about how they can get around that hour and if they could. You would have to ask for disabling based on when you came up against something that you thought you should have access to and that would mean that you would have to be identified because your library card is already there.

In addition to that one of the-

JUDGE BECKER: If you keep these other things confidential why wouldn't you keep that confidential?

THE WITNESS: But his question was whether we could have it be anonymous.

BY MR. SMITH:

Q In other words-

A We couldn't have it be anonymous. Yes, of course we would keep it confidential, although certainly having to get up in the middle of the search session and go find a library staff member-

JUDGE BECKER: So, the problem would be they have limited time -

THE WITNESS: Yes.

JUDGE BECKER: -and it would take them maybe ten 15 minutes to out of their-

THE WITNESS: Right.

THE COURT: -time.

THE WITNESS: Right, one of the reasons it was a minor reason, but one of the reasons why we thought the advanced registration for desk top was an important aspect is that everything we did that slowed down the logging in took away time from the patron and we got complaints about that. In addition to that this would-could be fairly staff intensive depending upon how many times people might ask for it.

JUDGE FULLAM: May I ask, is the limitation on the number of terminals based on the amount of space in the building or the cost of the additional stations or what?

THE WITNESS: The amount of space in the building, we've always been able to successfully get an LSCA grant when we've needed more computers.

MR. SMITH: Your Honor, I have no further questions, I would move the admission of Exhibit 61, 62 and 63?

JUDGE BECKER: Any objection?

MR. ZICK: No objection, Your Honor.

JUDGE BECKER: They're received.

(Plaintiffs' Exhibits 61, 62 and 63 are received in evidence.)

MR. SMITH: No further questions.

* * * * *

[TESTIMONY OF G. COOPER]
[March 25, 2002]
[pp. 85-140]

* * * * *

A We have 17 branches and a central library in downtown Portland.

Q How many people have library cards to use the services of the library?

A The last time we purged our records we were at about 82 percent of the people who live in that county having library cards in Multnomah County.

JUDGE BECKER: Wow, that's a lot.

THE WITNESS: Yes.

JUDGE BECKER: Is that a record, I mean-

THE WITNESS: In Boston, I'm told, it's 109 percent of their population, but I don't think they've purged their records very recently.

JUDGE BECKER: That doesn't tell you anything about the elections up there, does it?

JUDGE FULLAM: Do you have any notion of how many people actually use the library? I mean-

THE WITNESS: We do. We purge our records about every 18 months. So if you don't use the library within about a year and a half, we no longer have a card on our books for you.

We have some other measures as well. We're one of the busiest libraries in the country. Last year we checked out about 14 million books and other library materials, that's 21 for every man, woman and child who lives in Multnomah County in one year.

JUDGE FULLAM: And how many of those books were actually read?

THE WITNESS: I'm sure all of them.

JUDGE BARTLE: You don't have to answer that.

BY MR. HANSEN:

Q Now, how long have you been director in Multnomah County?

A I've been there for 12 years, always as director.

Q And how have things changed since you've begun your job?

A Things have changed enormously in that 12 years. We-our central library was built in 1913 and we moved out of it for four years, three years for a complete renovation of that building. We reopened it about five years ago.

We have renovated or replaced every one of the branches, and we opened our first two new libraries in 30 years just last year as a matter of fact. The services we provided changed. When I came 12 years ago, we didn't do nearly as much with very very young children as we do now. Now we have children as young as six to eight months coming to book babies with their parents to learn how to use books with very young children.

Our work with schools has changed tremendously in that time, and, of course, technology has made a big impact on what we do and how we serve people.

Q The renovation in the main library and the opening of the new branches, how did you finance those?

A In Oregon we like to vote on everything often and the voters have approved the money in '93 and in '96 for the capital improvements that I described.

In addition, we vote on our operating budget every three to five years, about 55 percent of our budget is determined by that vote.

JUDGE BECKER: We who? Who was the voters?

THE WITNESS: The voters in Multnomah County-

JUDGE BECKER: Really?

THE WITNESS: -and I'm one of them.

JUDGE BECKER: But they vote on the library budget?

THE WITNESS: They vote for a special levy, to allow the library to operate. And that election happens every three years, or by very recent state law, every five years.

JUDGE BECKER: Did you ever hear of that happening in any place else?

THE WITNESS: There are other libraries that get some of their funding from public vote, but in Oregon we vote a lot.

JUDGE BECKER: Okay.

BY MR. HANSEN:

Q What is the governing structure of the library?

A We are a county department. As such, I work for the board of county commissioners. I report directly to the county chair. She is elected from the county as a whole. The other four members of that board are elected from districts, they are the legislative group that-policy making group for the library and she is the-it's a she at this moment, the county chair is the top administrator and she's my direct boss.

There's also a library board. In the case of Multnomah County the library board is advisory, and their role is to be advisory to me and to the board of county commissioners on any matters relating to the library. Those 15 are appointed by the chair and confirmed by her colleagues on the board of county commissioners.

Q The board of county commissioners-the members of the board of county commissioners are selected in what way?

A They're elected, four of them, from districts and one, the chair, at large.

Q Is the Multnomah County Public Library a plaintiff in this case?

A Yes, it is.

Q Who made the decision that you would be a plaintiff in this case?

A We were invited to consider being a plaintiff. After considering it on a staff level and much conversation at library board meetings at two or three of the monthly meetings, and then finally a decision made by the board of county commissioners that we would be plaintiffs in this matter.

Q Does the library receive e-rate funding?

A Yes, it does.

Q How much e-rate money does the library receive?

A This year we have about $70,000 for internet access. We've made application for about 100,000 for next year.

Q What is the total budget of the library?

A The library's budget because of the economy changes from day to day it's about 44 million.

Q What is the basic mission of the Multnomah County Public Library?

A Multnomah County Library helps the people who live in our county and others who use the library find books and other library materials and information that they need for education, recreation, entertainment.

Q Now, when you refer to recreation and entertainment, could you elaborate on that a little?

A Well, we think of ourselves as being an adjunct to formal education and informal education, but in truth a whole lot of what happens at the library is what people find for their hobbies, their own particular interest or what it is that they like to spend their leisure time at. Examples would include some of the videos we have for children and adults that are purely entertainment in nature, for example, and books that would fall into that category too.

Q Can you give a specific example of something that might be a purely recreational part of that collection?

A Faulty Towers, the videos starring John Cleese (ph) are very popular. There's a series called Disc World, which is science fiction, our fantasy series, we have it on cassette and C.D. in spoken form and in books as well. Oh, the novels of Danielle Steele and Nora Roberts, the kinds of materials, the mysteries that people read for their own entertainment are very very popular.

JUDGE BECKER: People don't read our opinions for entertainment.

THE WITNESS: We haven't had that experience.

BY MR. HANSEN:

Q Are the collections contained in each of the branch libraries the same?

A No, the collections at each branch library are somewhat different from one another depending on the population that's served at a particular library.

Q Can you give an example?

A Well, I'll tell you an example that specific to Portland. Unlike other libraries we have more material on roses than almost anybody. Portland is the city of roses. We've had a rose festival for about a hundred years. We have historic materials including a collection that dates from England in the century before the last century, and current materials as well on the subject of roses.

In our branch libraries the collections differ depending on the population that's served So for example, we have two libraries, one with a Russian Orthodox population nearby, a neighborhood nearby, and one with Russian Pentecostal nearby. And we have Russian language materials in those two libraries.

Q Does the library provide any services other than the lending of books and other materials?

A We provide many services in addition to coming in and checking out books, there are other library materials. I mentioned earlier that we do story hours for very very young children, also for toddlers and four-year olds. Preschool programs are very important in the services that we provide.

Of course, we provide information. We help people find the information they want and need in print and in electronic format. We have meeting rooms, we have exhibits, we have guides to help people start small businesses, for example. An important role that we play is in introducing people who are job seekers, perhaps new to our community or perhaps looking for work outside of our community, and those who are beginning small businesses or would like to find ways to improve them. We have a packet of material, for example, that is available on request that tells you about the city licensing requirements and some of the state requirements for small businesses to comply with.

Q Are children served by that public library?

A Children are a major part of the audience that we serve, yes.

Q And can children get a library card?

A Yes, children can get a library card. We don't give them in utero, bit I've given them the day after birth, so very early.

Q What, are there any restrictions on children getting library cards?

A The only restriction on children getting library cards is for a child under 14, their registration card should be signed by a parent. That registration card that the parent signs explains our policies with regard to access to books and other library materials, and to make sure that the parent has an opportunity to know that we rely on the parent to supervise the child's use of the library.

Q Are there any restrictions on materials that children can access of check out at the library?

A No.

Q Are there constraints in the library setting that prevent you from fully fulfilling your mission?

A There are many. One is that we serve a large and diverse population, and I think every individual has at least one unique library need that's not shared by others. We don't have the money and we don't have the space to meet all the needs that come to us.

Q How do you try and get around that limitation- those limitations?

A Well, there are a variety of ways. Operating a public library is very much a balancing act, trying to figure out the best way to serve our population and actually the best way to spend the tax dollars that come from that local population.

We use a number of different ways to do so, inter-library loan is one. Of course, technology gives us access to more materials and that's yet another way in which we can serve very diverse needs.

Q Did you hear Ms. Morgan describe what the inter-library loan program is?

A Yes.

Q And was that essentially an accurate description?

A Yes. There's a difference in ours in that you can actually make an inter-library loan request without talking to a library staff member. You can fill that out with your library card as the identifying information and do it from home, work or school as well from in a library.

Q Is there any material that-

JUDGE BECKER: You said do it from home, can you do it by e-mail or-

THE WITNESS: That's right, via the internet. We have about a third of our access is remote, in other words, people choose to use the library without walking in our doors.

BY MR. HANSEN:

Q Is there any restriction on materials that the library would attempt to get through inter-library loan based on the content of the materials?

A Not based on the content. There might be differences based on, can we identify that this is a real piece of material, do we have the correct bibliographic or other identifying information.

Q Does the library contain material dealing with the subject of sex?

A Yes, the library has material dealing with the subject of sex.

Q Can you give me an example?

A I'll give you several examples. We have material in the children's library on where babies come from. We have magazine articles, some are on line and some in print. I actually have some examples.

We had a question from a patron about a Cosmopolitan article that identified men's erogenous zones, his secret sexual moan zones, for example. We have material like the pictorial information or images that would be found in the Mable Thorpe's photographs, for example. We have sex manuals for heterosexual and homosexuals, and we have a number of other specific books to help people's-answer their questions in this area.

Q Does the library carry Playboy Magazine?

A The library does have a subscription to Playboy Magazine, yes.

Q In the-is there such a thing as a card catalog anymore?

A No, there is not. The information about what the library has is on an electronic format, it means you can know what is available in the whole system, not just in a particular library. And, in fact, you can know that from home or work or school via the internet.

Q And if we were to search that database today, would we find books on the subject matter of erotica?

A Yes, that would be one of the subject headings and it would certainly pull up material.

Q Does the library carry R rated movies?

A We have videos and some of them are R rated, yes.

Q Does the library provide patrons with access to the internet?

A Yes, Multnomah County Library does.

Q And why does it do that?

A It does it for a variety of reasons. For one thing, it's part of what our patrons expect. A lot of them know that that's a valuable source of information. We are one of the most wired communities in the nation, but even in our community just under half of the households do not have internet access at home, so we know that that's something that people need and expect us to have.

It also widens what we can provide them with. It has information that might be more current and certainly more extensive than we can provide in print.

Q Do you have any statistics on the amount of terminals you have and the amount of usage on the internet?

A We consider technology both our present way of providing service and the path to our future. More and more of what we do now is available electronically.

We have about 470 computers for-with internet access for the public. And we have about 450 for staff. Some of those are in staff work areas and some are on reference desks or other service desks.

JUDGE BECKER: Are you reducing your budget for hard copy because of internet access?

THE WITNESS: We are not, Your Honor. Although we are finding that in some cases electronic access to a magazine can replace having the hard copy of the magazine and we're very happy to make that switch.

JUDGE BECKER: Okay. You do it for magazines then, rather than books?

THE WITNESS: Right. Although there are some-the universe of what we can buy for books, both because of the cost of having it, of keeping it and of the space for it is different than what we can buy for the internet. Now about just under a third of our budget for books and other library materials is for something other than print on paper. That might be DVDs, it might be databases, it might be CDs.

BY MR. HANSEN:

Q Do you know how many people use the internet at your library?

A We don't have exact numbers, but I know that in the last fiscal year we had 120,000 unique users of internet access at Multnomah County Library.

Q Now, all of the internet-

JUDGE BECKER: Well, what do you mean by unique?

THE WITNESS: It means if you were there and used it twice, we'd only count you once.

JUDGE BECKER: Okay.

BY MR. HANSEN:

Q Are all the terminals at the public library set up the same way?

A The basic setup is the same, they're configured similarly, but different functions are enabled at different computers for different purposes.

Q And would you explain what possible functions are enabled?

A In the library we have what we call resource computers. Those provide access to the library's card catalog-or the library's catalog, now no longer in cards, to the databases to which we subscribe, the library's home pages and various helping aides to help people find the information they want and need. And they are not-and library users are no able to type in a URL or an address for an internet site at those computers.

Then we have a series of computers that have all the things that I've just identified and are used for those purposes, but in addition allow people to use the internet independently of the aides that we provide.

Q Do you also have some computers configured for children?

A Yes, about 20 percent of our publicly accessible computers are configured for children.

Q Would you describe how that works?

A If it's one of the children's computers, then there's a difference in the way that it is used, you have to be child or be an adult accompanying a child, a care giver or a parent in order to be able to use a computer in the children's library at our main library or in any of the computers that are identified as children's computers in the branches.

In addition, when that computer is on, it opens up to the kids' page. And the kids' page is configured to be able to help children find what they want and need in the internet.

JUDGE BARTLE: At what age do you consider a child to be for these purposes?

THE WITNESS: We actually have children and we have teens, and so it goes up to about the age of 18. The kids' computers and the kids' page generally serve 12 and under.

BY MR. HANSEN:

Q And can a child under the age of 12 use one of the non-kid computers?

A Yes.

Q Would you look at Plaintiffs' Exhibit 102, please.

What is Plaintiffs' Exhibit 102?

A It's a very recent version of our kids' page. I know that it's recent because the material on the right changes and those are upcoming events.

Q And would you explain what this is showing us?

A This shows what you see when you first log-in to a children's computer, this is the kids' page.

Q And what would happen if, for example, I-the blue-I'm sorry, the blue lines, do those represent links that I can click on?

A That's right.

Q What would happen, for example, if I clicked on the homework center line?

A Well, I believe that's the next page in that exhibit. And those also are topics on which you could click to get further to links that have been identified as providing information to help with homework in those specific areas.

Q Sites that have been identified by whom?

A By library staff.

Q Can you explain how that occurs?

A Librarians and library assistants have the responsibility to monitor particular areas and they are responsible for providing the links to those pages.

So, for example, one of the youth librarians in the branches is in charge of home schooling. We have an active home schooling community in Multnomah County and that information-and she's responsible for finding that information and making sure that the links are up to date and current.

Q What's the third page of Exhibit 102?

A That's the Outernet, which is our page for teens or young adults.

Q It's the equivalent of the children's page but for teenagers?

A That's correct.

Q And what are the next few pages of that exhibit?

JUDGE BECKER: Well, I note that on this so-called outernet you have a link to health, sex and your body?

THE WITNESS: Yes, that's correct.

BY MR. HANSEN:

Q What would happen if you clicked on that?

A You would find sites we had identified to provide that information appropriate for those who might be using it.

Q Okay. Now, what are the remaining few pages of that exhibit?

A They are more details of what it is we have linked to the children's page. The first one identifies some search engines that are available and tells you a little bit about each of those search engines, again, to provide a guide for children and for their parents in order to make effective use of the internet.

Q And why do you create all of these pages?

A The internet is a wonderful resource for us and for those who use the library, but I think we're still in the very beginning of the use of this resource and it's very complicated to find your way around the internet. And to get to what you want and need is really the challenge for us and for those who use the library.

We provide a number of pathways or guides to increase the chance that people will actually find what they want rather than find what they're not interested in or waste their time with what's not pertinent to their needs.

Q Okay. Do the pages represented in Exhibit 102 accurately represent the web site that is run by the Multnomah County Public Library?

A Yes.

MR. HANSEN: Your Honor, I offer Exhibit 102.

JUDGE BECKER: Any objection?

MS. GACKI: No objection.

JUDGE BECKER: Received.

(Plaintiffs' Exhibit 102 is received in evidence.)

JUDGE BECKER: Ms. Cooper, is it a fair statement that children are becoming computer literate or adept at accessing the internet at increasingly younger ages?

THE WITNESS: Yes, that's been our experience. For example, for preschoolers we have some games on the internet, on the children's computers, to help kids become ready to read that identify shapes, for example. We have a very popular Cat in the Hat phonic game.

JUDGE BECKER: But preschoolers know how to access the internet?

THE WITNESS: Yes. To access the internet might be going a little far. It's not unusual to see a three-year old with her parent typing something in or clicking a mouse in order to be able to identify something on a screen.

JUDGE BECKER: But how about-I mean the six year olds, seven year olds, eight year olds, they're now using the internet?

THE WITNESS: You know, when you say, using the internet, I don't think anybody even knows that's what they're doing. Some of the resources that are mounted here and linked to the home pages are accessed through the internet. Some of those may reside on, or it may have at sometime have resided on our database in electronic format within our building. So they may be using the internet and not even knowing that they're doing so, and in that case-

JUDGE BECKER: But my question is, three, four years ago was that happening, or is this happening more, the younger children, as time goes on?

THE WITNESS: It is definitely increasing.

BY MR. HANSEN:

Q What is the library's policy with respect to the use of the internet?

A Can you ask that question in a different way?

Q Yes. Does the library provide blocking software?

A We provide library users with the opportunity to choose to have a filtered search, and they make that choice each time they initiate a search that goes out to the internet.

Q What blocking software product do you use?

A We use Foolproof Software.

Q And how does that work?

A Foolproof actually has a vocabulary of words and combination of words, and when a URL is typed in, the software-and by the way, this software is networked throughout our whole system. It resides on a server and it's throughout our 17 branches and central library in the nearly 100,000 computers for staff and the public that are available in that way. And it will literally in nanoseconds determine whether there are specific words or combinations of words that it deems by the category that we have enabled, we've enabled only the pornography category to be chosen by library users to be filtered, and then let the library user know if that URL or address is to a site that that particular software has determined may be pornographic.

JUDGE BECKER: Two questions. What is a nanosecond and-

THE WITNESS: It's shorter than a second. I think it's-it's actually a tenth of a second.

JUDGE BECKER: And do you have anywhere on here the screen which shows-gives the patron the option to have a filter or not have a filter?

THE WITNESS: I don't think we do.

JUDGE BECKER: Could we access that over lunch and print it out?

THE WITNESS: You could. I did it last week and can describe it for you as well.

JUDGE BARTLE: Does the-

BY MR. HANSEN:

Q Could the-I'm sorry, Your Honor.

JUDGE BARTLE: Does the potential user know the criteria that's being used by the library to filter out, say, pornography?

THE WITNESS: No. We tell people that that's what the filter is set to filter.

JUDGE BARTLE: It's designed to filter out pornography, but the library user doesn't know how you define that.

THE WITNESS: That's exactly right.

JUDGE BARTLE: See, they don't know what the combinations of words are.

THE WITNESS: That's right.

JUDGE FULLAM: And I assume URL means Underwriters Laboratory, does it?

THE WITNESS: No.

JUDGE FULLAM: What does it mean?

THE WITNESS: It's the address of a site, and-

JUDGE BECKER: Uniform Resource Locator.

THE WITNESS: Yes, Uniform Resource Locator. It really means-think of it as the address to get to a particular location on the internet.

JUDGE BARTLE: Who develops the definition of pornography in your library? Do you do it? Does the board of commissioners approve it? How do you determine that?

THE WITNESS: Actually, we've asked the filtering company to use the definition that they've developed.

JUDGE BECKER: So you rely on the filtering company?

THE WITNESS: We asked them to do that.

JUDGE BARTLE: So you just tell them, we want to have a filtering system for pornography and you ask them what the criteria are, and if you're satisfied, when I say, you, is it you the director who decides that or does it have to be approved by the commissioners?

THE WITNESS: Actually, it's a different level than that altogether. We chose this particular product for the blocking software we installed after extensive staff review and review as well by an internet access committee board that participated together with staff in reviewing a variety of products that were available.

JUDGE BECKER: Right. And you might just give us-I'd be interested in seeing that screen, so maybe you could give one of my law clerks the web address and we'll try and print it out over the lunch hour.

MR. HANSEN: We would be happy to do that, Your Honor.

BY MR. HANSEN:

Q You referred earlier to the fact that you enabled the category of pornography and it's triggered by words or a combination of words.

Do you, as the library, know what words or combination of words cause the software to label a page pornography?

A No, we do not.

Q Can you find that out from the blocking software company?

A I do not know.

JUDGE BARTLE: Well, then how do you know whether to accept it or not? I mean they can be screening out things that in the general view may not be pornographic.

THE WITNESS: And I think you're probably right, they well may.

BY MR. HANSEN:

Q Does that, in fact, happen?

A We have had experiences where that's happened. In our case, because it's a choice that the library user can make, they have the opportunity to back out of such a search and begin an unfiltered search.

JUDGE BARTLE: For example, the company can say anytime the word, breast, is mentioned or, testicles or uterus are mentioned, that's pornographic and people may be wanting to find information about breast cancer or testicular cancer, uterine cancer.

THE WITNESS: You've listed some examples and there are many more that happen as well. I can give you a couple of examples of the kinds of things that we are aware of. This filtering-this content blocking software will identify the word vibrator as one not to be found, but it doesn't include the word, vibrators, with an s on the end. So you can, thus, find information or material through that way.

Sexual dysfunction is not blocked, but through the sites that come up with that search on a filtered station, you have access to pictures of erect and flaccid penises to circumcised and uncircumcised. Foreskin is not blocked. You have access to herbal Viagra to ads about how to add four inches to penis lengths. Excessive masturbation is not blocked, and we've found that you can have access to partygirl.com and other similar sites even with the filter enabled.

MS. GACKI: Your Honors, I'd like to object at this time. Lack of foundation laid for this testimony.

JUDGE BECKER: Do you want to respond, Mr. Hansen?

MR. HANSEN: How is that you know that-I'll do a foundational question, Your Honor.

JUDGE BECKER: All right. Well, we'll reserve the objection.

BY MR. HANSEN:

Q Ms. Cooper, how is it that you know that these problems have existed with respect to the software that you use?

A I asked library staff to report to me in advance of my coming here for this testimony of specific examples of over-blocking and under-blocking of which they were aware.

We had earlier had some examples that were brought to our attention by library users and by other library staff.

Q Have there been instances of over-blocking that you haven't talked about?

A Where something was blocked that we think probably shouldn't have been, one that I found most interesting is that one of the frequent reasons people come into the library to use the internet is to access e-mail. In fact, we've just concluded a series of classes for seniors to-that was actually funded by the library foundation to help them be able to e-mail their grandchildren and family members in distant areas.

Many of them use free e-mail services like Yahoo or Hotmail. And because they are supported by ads and some of the ads that are part of that are for Find Your Perfect Match and other sites that would be blocked by our software access to that e-mail is soft-is blocked, so I think that would be one example.

Q Were there ever any occasions in which any of the sites you had chosen for your kids' page or your teen page were blocked by software you used?

JUDGE BECKER: Well, is this foundation? Well, it seems to me like this is continuing the direct examination.

MR. HANSEN: I'm sorry, Your Honor. I thought that-I thought I had asked the foundational question and it was sufficiently-

JUDGE BECKER: Well, we didn't give Ms. Gacki an opportunity to tell us whether or not she renewed her objection.

MS. GACKI: I would like to renew my objection, yes.

JUDGE BECKER: Ms. Gacki, have in mind we've got a three-judge court in a national case, and if she were able to-I mean you may be technically correct that this-

MS. GACKI: Perhaps I should let it go in this instance.

JUDGE BECKER: -is hearsay. Well, I mean you can bring in these other witnesses or they could be deposed. I mean have in mind that this is a constitutional adjudication and there seems to be, indeed, this material is material that both sides have recognized exists. The question is, what is the legal affect of it.

Well, do you want to -

MS. GACKI: I will withdraw my objection.

JUDGE BECKER: Go ahead, Mr. Hansen.

BY MR. HANSEN:

Q Ms. Cooper, the question pending is that there have been occasions when the blocking software you use has blocked pages that librarians have chosen for the kids' page?

A Yes, that's exactly right. And that has happened for especially the outernet page, but others about specific health related matters, puberty, for example, safe sex information, a variety of sites that we've identified as being appropriate to provide teens with information they may want and need have been blocked by the software.

Q If I'm a patron and I'm sitting at one of your terminals and I've chosen a blocked usage of the internet, and I find that I am blocked by something, what, if anything, can I do?

A You have the choice -

MS. GACKI: Objection. Foundation?

JUDGE BECKER: Overruled.

THE WITNESS: You have the choice of changing your choice. In other words, you can back out of the option to have a blocked or filtered search, and have an unfiltered or unblocked search.

BY MR. HANSEN:

Q I would like you to look at Plaintiffs' Exhibit 103 and 104.

Could you describe what those are, please?

A 103 is the parent's guide to Multnomah County Library which includes electronic resources. And 104 is all our policy on acceptable use of the internet.

JUDGE BECKER: Incidently, Ms. Cooper, this seems to assume, at least some of Ms. Morgan's testimony, seems to assume that there's always a parent. There isn't always a parent, is there?

THE WITNESS: No, and in fact that's part of what's covered in the Exhibit 103. We, and in a library card registration form that I referred to earlier, we really remind parents that we are a public place, and that when their children come there, whether they're with them or not, their behavior is not something that we'll try to identify or modify as regards what we'd need to know to have particular knowledge of that family's values and that child's maturity.

JUDGE FULLAM: What do you do in a case of a kid that has no parents? Do you take the word of a guardian or a care giver?

THE WITNESS: Yes, that's right, whoever it is that's empowered to act as the care giver for that child.

So for example, it might be a teacher in the case of a class that comes to visit the library who says, we're going to show our sixth graders how to access the internet, but I want all of them to use a content blocking software and they are all instructed to do so in that case.

BY MR. HANSEN:

Q What is the purpose of the library producing Plaintiffs' Exhibit 103 and 104?

JUDGE BARTLE: I think that's for you to decide. She doesn't know what the purpose of it-why-

BY MR. HANSEN:

Q What is the purpose of the library producing-

JUDGE BARTLE: Oh, you mean preparing it?

MR. HANSEN: That is precisely what I mean, Your Honor.

JUDGE BARTLE: All right.

BY MR. HANSEN:

Q Are Plaintiffs' Exhibits 103 and 10-who is the author of Plaintiffs' Exhibits 103 and 104?

A Library staff are.

Q And why did the library create these documents?

A Again, we found the internet to be a wonderful resource, but not a perfect one. And just as we provide health and guide for people and how to use our print collections, we provide that same kind of help and guidance in how to use our electronic resources.

Q And are these pamphlets available to patrons at the library?

A Yes, and in fact new cardholders are routinely given these as well as other information on materials about use at the library.

Q Do they accurately describe the library's policies with respect to internet usage?

A Yes, with the caveat that our policies continually are refined and changed. So for example, we now provide word processing on internet computers, and I don't think we did that at the time we did the last printing of the acceptable use of the internet. So there are some things that would be modifications and changes.

MR. HANSEN: Your Honors, I offer Plaintiffs' Exhibit 103 and 104 into evidence.

JUDGE BECKER: Any objection?

MS. GACKI: No objection.

BY MR. HANSEN:

Q Ms. Cooper, how did the library come to the policy choices it made with respect to the internet?

A Oh, a variety of ways. What's available in technology at any given time, that's part of the reason for now being able to provide the word processing. There are staff groups that make recommendations. We certainly hear from the public on a regular basis in a variety of ways. The library board has an internet access committee that has met for several different series of major decisions that we have made in this area.

Q Is confidentiality an important part of the library's mission?

A Yes, it is.

Q Would you explain how and why?

A In a variety of ways, somebody's library card is the only thing that allows that person or anyone else to know what they've checked out, even if the material is overdue, for example. We are very careful and the staff is trained to be very careful when somebody checks out a book to not comment on that book that's checked out even if it is a comment that might seem very innocuous. We want to make it clear to library users that even if they're checking out how to get a divorce from my abusive husband, we won't say, oh, I see you've got this book, to them. So confidentiality in that way is important.

We are similarly concerned about confidentiality of what questions people are asking and what responses they are given.

Q Can I take a book off of the shelves at the public library and read it in the library?

A Yes, you can do so.

Q Do you offer reference librarians?

A Yes, we do.

Q What is a reference librarian?

A Reference-public service staff, reference librarians and library assistants help people in their use of the library, particularly for finding information that they may want and need. And they are located at branch libraries and throughout our central library as well.

Q Are there methods by which I can ask questions of a Multnomah County Public librarian without going directly up to a desk and talking to that person face to face?

A There are. For some years we've offered telephone reference service and you can call from home or work or school and people do that about 80,000 times in any given year.

In addition, as of a couple of years ago you could ask a question over the internet. Your library card is the locator for that information and we get back to you with a response to that question.

Q Why do you offer those two forms of reference librarianship as well as the traditional form?

A Well, primarily for the convenience of library users. A part of what we do is spend the public's money and figure out the best way for them to get use of it, and one is for them to be able to get the information they want and need as rapidly as possible, and both of those two methods help that happen.

Q Is confidentiality relevant at all to that service?

A I think in both those examples it is one of the benefits. It's not the reason why we've done it, but it is one of the benefits.

Q And would you explain why that is?

A Because you don't have to look somebody in the eye and ask them a question, you might not be comfortable asking them. And you don't have to do so in the hearing of somebody who might also be ask the information desk other than the librarian, another member of the public or somebody.

Q Is that a problem known to librarians, not wanting to ask the question?

A Yes, it is.

Q Could you elaborate on that?

A Well, I'll give you a recent experience. A young man asked for health information, a man who looked to be in his early teens. The reference librarian went with him, this was at central library two, the health information and showed him that really it was hundreds of books, was there something in particular he was looking for. And he said that he wanted information on how to identify sexually transmitted diseases.

In this case the librarian and the young man found material that looked to be relevant in the library's catalog, and she actually went to the storage area in the basement, one of the books had recently been returned and hadn't yet been reshelved and brought it back to him so that he wouldn't have to encounter a page at the desk to ask for such information to be retrieved.

That's but one example, and we know as well that over the phone and through Ask Us Online, the internet connection, we do get questions that are unlikely to have-we're unlikely to have gotten that volume of questions on that subject over the desk in person.

Q Does the library take any steps to assist patrons in finding material on the internet that they want to find and avoiding materials that they don't want to find?

A That's our primary role, to help people find what they want and need rather than finding material that is not of interest to them or will waste their time, so we do that in a variety of ways, yes.

Q Can you describe come of those ways?

A Well, we've been talking about the kids' page, and we talked about the topics that are listed and the links that are provided.

In the adult material we do exactly the same thing. So we know what topics are frequently-frequent questions, it might be weather, it might be travel information, it might be cars, it might be job seeking, it might be history, a whole variety of topics on which we know-about which we know there's interest in our community and we'll provide people with appropriate links that we have evaluated to be able to find that information.

Q Do you provide any classes that would assist patrons?

A We provide a number of classes and we teach this material in two ways, by classes for children, for adults, I mentioned the cyber senior's classes a moment ago that we've just offered.

We offer classes in how to use the library's resources, as well as on how to access e-mail and search the internet in general in terms of quality of search engines and other questions.

But the majority of the teaching that we do is actually just in time teaching. In other words, not everybody will find the time to come to a class, and when you need to know something, you need somebody who can help you find it, and that's a lot of what-the work that our reference librarians do.

Q The problem of helping patrons find material that they want and avoid material they don't want, is that unique to the internet?

A No, it certainly is not. It's especially true in any library, but particularly a large one like ours with just under two million items both print and otherwise that can be checked out. So a lot of what we do is help provide people with guides to get to what they want and need.

Q Did you take any steps to deal with people who might be offended that they passed by another user's computer?

A Yes, we do.

Q What is that?

A We do have privacy screens on all of our computers.

Q Would you describe the kind of privacy screen that you use?

A We've had privacy screens, I think, since we began offering internet access in 1995. We've recently replaced the ones we had with a new style that actually is part of the monitor. So it is physically attached to that computer rather than just something that folds down or slides over.

Q What are Ginnie coupons?

A I've been at the library 12 years and I came with a very strong belief that we wanted to make sure we heard from our public and we heard on a regular basis, so we instituted a feedback mechanism that goes by the name of Ginnie coupons.

JUDGE BECKER: It's named after you.

JUDGE BARTLE: It's named for you.

THE WITNESS: Named for me. That's what the staff calls them. We don't think the public calls them that, but they identify our interest in hearing complaints, comments or compliments.

In addition, say that if we have a way to get back to people, we will respond to them and we do that either by e-mail or by regular mail if they've given us an address.

BY MR. HANSEN:

Q Have you received Ginnie coupons complaining about the content of materials either in the print or internet collections of the library?

A We received many Ginnie couples on virtually any subject you can think of, and those subjects are among those that we have coupons on.

Q Can you give me some statistics?

A In the last fiscal year we had about just over 2,200 Ginnie coupons that were logged by library staff.

JUDGE BARTLE: What is the fiscal year you're talking about?

THE WITNESS: It ended the 1st of July, so it would have been -

JUDGE BECKER: Of '01?

THE WITNESS: Yes, that's right.

JUDGE BARTLE: So roughly July of 2000 to July 2001?

THE WITNESS: That's correct.

JUDGE BARTLE: Sorry, go ahead.

BY MR. HANSEN:

Q Of those 2,200 Ginnie coupons how many had to deal with the content of the material in the library?

A 37.

Q Can you break that down further?

A I think I can. Of the 37, 11 were particularly related internet content, 11 were comments in response to the county and the library board's position to participate as a plaintiff in this action, and the remainder were print related. We may have-

JUDGE BECKER: What related, I'm sorry?

JUDGE BARTLE: Print, print related.

THE WITNESS: Print to a book or a magazine. We may have had others. There is an additional form the library has had for a long while for print materials and we would have maybe double that number total that would have come in on these forms and the other form I spoke of.

BY MR. HANSEN:

Q Does the library have rules relating to what is acceptable behavior in the library?

A Yes, we do.

Q Would you look, please, at Plaintiffs' Exhibit 106. What is Plaintiffs' Exhibit 106?

A Behavior rules governing the use of Multnomah County Library.

Q And is that an official policy of the library?

A Yes, it is.

Q Why do you have such rules?

A Because we need them. In this case, this is a printout from what's available on the internet. They're also posted in general form in all library locations to make sure people are aware of them. And it's the way that we deal with behaviors of a variety of kinds that are not acceptable in the library.

Q And what are the ranges of things you would do with unacceptable behaviors?

A They range from telling someone to stop that action, to banning people from the library for a period of time up to as much as a year.

Q And might there also be occasions when it resulted calling the police?

A Definitely.

Q And does Plaintiffs' Exhibit 106 accurately describe the behavioral policies of the library?

A Yes, they do.

MR. HANSEN: Your Honors, I would move Plaintiffs' Exhibit 106 into evidence.

JUDGE BECKER: Any objection?

MS. GACKI: No objection.

BY MR. HANSEN:

Q Does the library have a policy with respect to-

JUDGE BARTLE: Well, why don't we do this, I think we can save time. We will-why don't you wait until the end of your case and then move in all the exhibits at once and that way-

JUDGE FULLAM: I'm inclined to assume they're all admissible when you-

MR. HANSEN: Yes, Your Honor.

JUDGE FULLAM: -offer them unless they're objected to.

JUDGE BECKER: I think that's a good idea. At the conclusion -

MR. HANSEN: At the conclusion of each witness? Of each witness, that will be fine, Your Honors.

BY MR. HANSEN:

Q Does the library have policies with respect to child pornography?

A Yes.

Q Would you look at Plaintiffs' Exhibit 105.

(Pause in proceedings.)

Q Would you describe what that is, please?

A It is a memo from the deputy director of Multnomah County Library, Ruth Metz, to staff concerning- it's a clarification of the previous document we were looking at.

Q And what are staffs supposed to do if they find material that they suspect might be child pornography?

A They're instructed to call the police if that's the case.

Q Are you familiar with the Children's Internet Protection Act?

A Yes.

Q And is the Children's Internet Protection Act consistent with the mission of the Multnomah Public Library?

A No.

Q In what way is it not?

A The Children's Internet Protection Act, as I understand it, requires libraries who accept certain federal moneys to provide only blocked or filtered access to the internet for children and for adults.

We have found that providing information in print and electronically is complicated and difficult and this would be an additional complication in trying to fulfill that mission for the public.

Q Is it simple a practical problem, practical-it would be complicated to put in place?

A Well, it is that. I think it is more than that as well.

Q In what way?

A Material that is illegal because if child pornography or its been adjudicated as obscene is not allowed to be used or viewed in our library, but there is a whole variety of other material that is neither adjudicated to be obscene or child pornography that people do regularly ask us about, and I am interested in making sure that we continue to have the ability to provide that information for them.

JUDGE BECKER: At the bottom line, Ms. Cooper, I take it if there was the perfect internet filter, or the perfect blocking mechanism, you would have no problem with it?

THE WITNESS: As a matter of fact I have said exactly that fairly often. The real difficulty is helping people find what they want and need. It is a very imperfect-the internet is still a very gross tool to use, and if I had a way to make sure that you would find what you wanted and needed and not what you didn't need or didn't want, whether it's called a filter or something else, we would definitely buy it.

JUDGE BARTLE: I think Judge Becker was asking if you had a filtering system which could automatically filter out anything that a court has declared obscene or anything that is child pornography under the Constitution of the United States, you probably would have no problem with that?

JUDGE BECKER: And no more?

JUDGE BARTLE: And no more, just that?

THE WITNESS: Yes, I think you're right about that. I had not considered that and I misunderstood your question earlier, but yes, I think the answer to that is yes.

BY MR. HANSEN:

Q Would the implementation of the Children's Internet Protection Act also cause practical problems in the library?

A Yes, it would.

Q Specifically are you familiar with the way in which the internet-the Children's Internet Protection Act provides different content that is to be blocked to children from that that is to be blocked to adults?

A Yes, I know that there is a difference.

Q And would making that distinction present practical problems with the library?

A It would. At the time you have a library card, you don't have a particular card if you're a child and another one if you're an adult.

Q And what would you have to do in order to change that?

A I haven't even begun to think about what would have to be in place both with our current group of several hundred thousand cardholders and those who apply for cards as well.

Q Would you also have-

JUDGE BARTLE: That wouldn't be that difficult a problem, would it really, to have different cards for children, have one that's red and have one that's blue.

THE WITNESS: We don't have them now, so it would be-it would be quite something to put it in place especially with our current cardholder use.

Part of the reason we do it this way is because once you get a card, you've got the card forever.

JUDGE BARTLE: I see.

THE WITNESS: And it would be an additional workload or an additional decision to decide exactly how old the child is and when you stop being a child and give them a different card.

JUDGE BARTLE: It would be a problem with including your date of birth? You do it on your license at least in Pennsylvania, your driver's license has your date of birth now, couldn't that be done on a library card?

THE WITNESS: I suppose it could be.

BY MR. HANSEN:

Q Do I currently enter something from my library card when I sign on the computer?

A Yes, you enter both the card number and a personal identification number.

Q Would separating children's library cards from adult library cards be contrary to the current philosophy of the library?

A Yes, it would be.

Q Would you elaborate on that?

A Now. We believe that we can't know what's right for an individual child, that only that child and more importantly their family can know what that child's maturity level is and from the context of that family's values and ways of looking at it.

There are some 12 year olds whose parents would not want them to have access to birth control information. There are some whose parents would hope that the library would be there to provide that access.

There are some five year olds whose parents don't want them to know where babies come from, and there are some who would. So we can't presume to judge for everybody in our community and we don't try to, rather we work with parents to make sure that they understand how to do that and can do it effectively.

Q Are you familiar with the provision that the Children's Internet Protection Act that requires the blocking of staff computers?

A I know that that's required, yes.

Q And would that present any problems for the library?

A As I understand it now, it would, particularly for identifying some of the cites that I described earlier that are linked to our home pages.

Q I didn't understand that. Explain that.

A The kinds of sex information we provide for teens, for example, the kinds of health information we provide for adults as well.

Q What about reference librarians and their use of the internet?

A Yes. A number of the staff computers are on public information desks and often they are used to help somebody find the information they want and need and those searches would be easier to perform if they were not blocked.

Q You referred earlier to the fact that the software that the library utilizes has shown evidence of both over-blocking and under-blocking. Does that present problems for you with respect to the mandatory use of blocking software?

A Yes.

Q In what way?

A Because such software would then result in the over-blocking and under-blocking like the examples that I have identified.

Q Why do you use the soft wall at all given the problems you've just described?

A Because it's very important that people have that choice and that's why we put this practice in place. For example, while I have some concern that parents may think that we're providing an environment that is without-that will be without concern to them and their children.

I still think that we provide them with something positive when we say and you can use the library in this way, use the internet and instruct your children to use it in this way.

Q Are you familiar with other libraries that utilize blocking software on a mandatory basis?

A I know that there are some that do.

Q Does the librarian community discuss that at professional conferences and meetings?

A Yes.

Q And have any of the libraries that you're aware of that use mandatory blocking software express concern about that use?

A I do know that at least one library-

MS. GACKI: Objection, hearsay.

JUDGE BECKER: Well, the objection is they have expressed concern.

JUDGE FULLAM: Yes. The fact that it was expressed, not the truth of what it said the question was what it was offered for.

JUDGE BECKER: We'll allow it.

JUDGE FULLAM: State of mind. Go ahead.

JUDGE BECKER: I'm not sure it's relevant, but we'll allow it.

THE WITNESS: Can you repeat the question for me?

BY MR. HANSEN:

Q Yes. Have the other libraries that you've referred to express concern about the mandatory use of blocking software?

A I know of at least one library with mandatory blocking software that has expressed concern about the effect of that software?

Q Which library is that?

A Tacoma.

Q And who was it that expressed that concern?

A I spoke at the Public Library Association conference with the library director just briefly.

Q And specifically what was the nature of her concern?

A She offered to me her concern about the effect of blocking software and their ability provide library service.

Q Are you familiar with the provision of the Children's Internet Protection Act that provides that librarians can unblock sites based upon a bona fide research or other lawful purpose?

A I know that that provision exists.

Q And does that provision provide any practical problems for you if you were to be required to implement the act?

A I've talked with, you know, probably ten lawyers and librarians specifically about this and nobody comes up with the same definition of what this means or how it would actually work, so I believe it would cause some problems, yes.

Q If the standard were to require you to judge the purpose of the patron in going to the web site-

JUDGE BECKER: Well, why don't you give her the exact language from the act? Was it bona fide research?

BY MR. HANSEN:

Q Bona fide research or other lawful purpose.

A Right.

Q What are the options of how you would interpret that?

A Well, first of all, that's contrary to how we operate. In our library and in most libraries we really don't ask why people want to know what they want to know. That's really their business and not ours.

So we don't judge why somebody is asking for how you make jelly beans or how much beer is consumed in Oregon. We don't say is that work or personal, or make a judgment in any way.

So it would be a very different role for us and I'm not sure how we would make a determination about bona fide research.

Q To the extent you were called upon to judge the other half of that, the other lawful purpose -

A Right.

Q -how would you do that?

A Well, except for the categories that we've talked about, almost everything would be lawful.

JUDGE FULLAM: So you would determine it as meaning you could unblock any time you wanted to?

THE WITNESS: It might well mean that.

BY MR. HANSEN:

Q At what level within the library would you provide unblocking?

A Again, I don't think that's clear in what's included in the statute, but we would have an interest in providing it at the lowest level possible.

I'm especially concerned about how people would respond to a delay in getting information they need. But we're already working to make sure that people know that the library is the place where they can go to get help finding the information they want and need, and I hate to see you can get help but, oh, by the way, wait now while somebody decides what to do about this situation.

JUDGE BARTLE: In other words, you wouldn't want a policy where you had to go to the county commissioners to unblock?

THE WITNESS: No, or even to me.

BY MR. HANSEN:

Q If you could give the authority to unblock to all of the librarians on the library floor would that present any problems?

A Well, remember that it's not just librarians, it's library assistants as well, so it's not everybody who serves the public that way who is-have a library degree or is classified as a librarian.

In our case that would mean nearly a hundred people would have that authority and I would have some concerns about the consistency with which that decision would be made from time to time.

JUDGE BECKER: Well, do you have some ba\ranch libraries in remote locations?

THE WITNESS: Yes, we do.

JUDGE BECKER: How far out?

THE WITNESS: About 20 miles.

JUDGE FULLAM: What kind of communities are they?

THE WITNESS: Those are a suburban community as large as a hundred thousand. There are very rural communities, a couple of hundred in a couple of cases.

JUDGE FULLAM: What are the rural communities, for example?

THE WITNESS: Troutdale, Wood Village, Corbit.

JUDGE FULLAM: And there are libraries-

THE WITNESS: Yes.

JUDGE FULLAM: -in those places. And who are the staffed by?

THE WITNESS: They're staffed by supervisors or library assistants. There's a youth library in all of our libraries, but there's not a reference librarian in all of our libraries.

JUDGE BECKER: Are any of them staffed entirely by volunteers?

THE WITNESS: No, although we use volunteers in every single one of our operations.

JUDGE BARTLE: Would they have authority?

THE WITNESS: They don't usually work-staff a public service desk, so they wouldn't have this responsibility.

BY MR. HANSEN:

Q Do you know whether you could set up the library system in such a way that a librarian in one branch could provide-could unblock sites?

A I don't know. It may be possible.

Q And do you know whether if you set it up that way it would unblock for all of the other branches as well or just for that one branch or just for that one patron?

A I don't know.

JUDGE BARTLE: So you could have a patron wait until a more "liberal" was servicing the desk and wait until that time to come in and ask to have it unblocked as opposed to somebody else?

THE WITNESS: That's possible.

JUDGE BARTLE: You could sort of-instead of having random selection, you could wait-

JUDGE BECKER: For a more libertarian.

THE WITNESS: It's like shopping for a librarian.

JUDGE BARTLE: Or a more libertarian, yes.

JUDGE BECKER: On the other hand, you could have let's say somebody in Portland or a couple of specialists who would be available and on call at all times for a call in by the Trout-was it Troutdale or whatever who could quickly answer the question, isn't that so?

THE WITNESS: We could, but they're already pretty busy, so I'm reluctant to identify that it would be easy for us to have that added assignment for our staff.

It's also not clear to me if we're talking about a site by site blocking or a session by session blocking or if an individual comes and say this site is blocked and I want to see it or says I want to do research on hmm-hmm, and it might be blocked. I don't think that's clear, either.

BY MR. HANSEN:

Q Are there reasons why patrons would not want to request unblocking?

A Yes, I think there are.

Q Could you explain what that is about?

A Well, I think for one reason it would be the time, you know, the difficulty of giving up your internet computer terminal or raising your hand or something to get a librarian to come over and deal with that.

And, in addition, I can understand circumstances similar to those we talked about earlier where somebody might be reluctant to call attention to their request for information in a particular area.

Q Given these problems that you discussed, what will the library do if the act takes effect?

A I don't know. We have not make a decision about what we would do.

MR. HANSEN: Your Honors, I have no further questions. I would move the admission of Plaintiffs' Exhibit 105.

JUDGE BECKER: Hearing no objection, we'll admit it.

(Plaintiffs' Exhibit 105 is admitted in evidence.)

JUDGE BECKER: Ms. Gacki, you can cross-examine.

(Pause in proceedings.)

JUDGE BARTLE: Ms. Gacki, could you give us a hint as to what volumes you've placed there so we can get them out.

MS. GACKI: 227-

JUDGE BECKER: 2 what again?

MS. GACKI: 227.

JUDGE BECKER: Go ahead, Ms. Gacki.

[CROSS-EXAMINATION]

BY MS. GACKI:

Q Ms. Cooper, the Multnomah Public Library keeps records of incident reports, correct?

A Yes, that's correct, it does.

Q And in some of these incident reports they concern patrons access of pornography on library internet terminals, correct?

A Some do, that's right.

JUDGE BECKER: Which one is this now? What exhibit are we talking about?

MS. GACKI: We're going to go to-I will get to those in just a second, two more questions and then we're at the exhibits.

JUDGE BECKER: Okay. Very well.

BY MS. GACKI:

Q In your letter you also retained some e-mail messages between employees, correct?

A That's correct.

Q And some of those e-mail messages between employees relate to patrons access of pornography on the internet, correct?

A Yes.

Q At this time I would refer the witness and the Court to the Exhibits 254, 256, 257, 258 and 261.

MS. BEESON: I'm sorry, could you list them again?

MS. GACKI: Sure. 254, 256, 257, 258 and 261.

JUDGE FULLAM: 261 is in the next book, okay.

JUDGE BECKER: Go ahead.

BY MS. GACKI:

Q Do you recognize these exhibits, Ms. Cooper?

A In a general way I do, yes.

Q These were the incident reports and e-mail messages that I had previously mentioned, correct?

A Yes.

Q And these incident reports are compiled by employees of the library in performance of their job duties, correct?

A Yes.

Q And they're reported to the library system, correct?

A Can you-

Q They are reported to people in your office, to people in the executive offices of the library system?

A In some cases yes, in most cases we just keep them on file.

Q But they're regularly kept according to-

A Yes, that's correct.

Q Your computers are equipped with privacy screens you've testified, correct?

A Yes.

Q And privacy screens allow the person who is looking directly at the screen to be able to see the material that's being displayed, correct?

A Correct.

Q And if patrons at your library choose to view pornography on your internet, they are supposed to use a terminal that is equipped with a privacy screen, correct?

A Yes.

Q And, in fact, it has been your library's practice to tell patrons only to view pornography at terminals equipped with privacy screens, correct?

A All of our computers are equipped with privacy screens.

Q In the past, not all of your terminals have been equipped with privacy screens, correct?

A I think we have always had privacy screens for all computers. Sometimes they fall off or get moved.

* * * * *

[TESTIMONY OF NUNBERG]
[March 25, 2002]
[pp. 256-314]

* * * * *

BY MR. SMITH:

Q In summary, Dr. Nunberg, what conclusions did you reach about the current operation of the commercially available software in terms of its efficacy?

A I concluded that none of the programs that are currently available can accomplish their stated task of filtering out-let me just use sexually explicit material in a kind of vague way here.

Q I've been doing that all day.

A Well, just without specifying-can filter out such material without also both over-blocking a great deal, that is to say tagging as sexually explicit material content that shouldn't properly be included in that category, and under-blocking, that is to say failing to tag a substantial amount of material that should properly be included in that category.

JUDGE BARTLE: Of course, doesn't it depend on what you mean by sexually explicit material?

THE WITNESS: Absolutely. But in a sense, it doesn't, in a sense, because however you define it-

JUDGE BARTLE: However you do it.

THE WITNESS: -the under-blocking and over-blocking are going to be problems. You can define it in different ways and make it narrow or wider, but however you define the category-and in this sense, sexually explicit material is no different from articles about baseball. Whatever the category is, this is a chronic problem with this kind of software.

JUDGE BECKER: I guess you could only-your problem can only be solved if you want to omit any article that has a certain word, that way, that would be simple?

THE WITNESS: Well, sure, if you want to say give me every article that, for instance, mentions-

JUDGE BECKER: That has the word sex in it.

THE WITNESS: -the word baseball or sex-

JUDGE BECKER: Or mentions the word sex.

THE WITNESS: -then or let's say exclude-

JUDGE BECKER: That's easy.

THE WITNESS: -every article that mentions the word sex, then I will exclude a very large amount, by no means all of the sexually explicit material-

JUDGE BECKER: I understand, but-

THE WITNESS: -on it, but I'll also exclude article in Scientific American, articles in Women's Day and so forth.

JUDGE BECKER: I understand. If I could refine that, let's assume that the particular document does not contain the word sex but contains a word that ends in S followed by another word that begins in E-X.

THE WITNESS: That could or could not be a problem for the software depending on how it was implemented. There are cases that have been tried, I don't think any of them were entered here, in which the filtering software blocks, for example, the documents in which the name Middlesex appears.

There is a case that was cited of-it's actually a class, this is over here at the University of Pennsylvania, who teaches Augustine and the Early Christian Fathers, and he put up a great deal of St. Augustine in Latin, and the site was-that text was blocked by one of the filters, not one of them here, because it contained the word cum, C-U-M, which presumably was also on somebody's list.

So the possibility of over-blocking with that technology is very real and-

JUDGE BARTLE: And of course our wonderful town of Intercourse in Lancaster County in this district would have these-

THE WITNESS: Right. And Beaver College, another Pennsylvania institution has had problems along this line as well.

JUDGE BECKER: They've changed the name. It's now Arcadia.

JUDGE BARTLE: They haven't the name of Intercourse.

JUDGE BECKER: No, they have not.

BY MR. SMITH:

Q Doctor, what conclusions did you reach, in summary, about the theoretical or inherent limits of software as a mechanism to block access to sexually explicit material on the internet?

A Given the methods that are available to software to deal with any kind of natural language, any kind of human language content, the difficulties posed by the size and dynamic nature of the worldwide web and the limitations of implementing these kinds of systems, the problem of substantial over-blocking and substantial under-blocking is going to be persistent for any state of the art, not just the way the stuff is implemented today.

JUDGE BECKER: Now, when you talk about the dynamic nature of the web, does that mean there are new sites every few days or, well, every day or-

THE WITNESS: Yes, the web is changing at an extraordinary rate, by one estimate, and that's actually about two years old, a million and a half sites per day are being added to the worldwide web.

BY MR. SMITH:

Q Is that pages or sites?

A I'm sorry, a million and a half pages, yes, are being added a day to the worldwide web. And somebody has to keep up with those sites, link in and out, particularly sites of this type tend to be not surprisingly to disappear very quickly and to reappear under new names and so on.

So keeping up with that, given the size of the problem, is just an inherent limitation of what you can do with any kind of software.

Q While we're on that subject, Doctor, what is the overall size of the material on the web?

A It's very hard to answer, partly because nobody is quite sure where the worldwide web ends, and partly because you don't know what to count. Should it be sites, should it be pages, should it be the number of servers or computers on the internet?

Partly because the tools for measuring the size of the web or themselves in precise. By a couple of recent estimates, there are two billion pages on what we can call the publicly indexable web.

JUDGE BECKER: What's that mean, indexable?

JUDGE BARTLE: An index.

JUDGE BECKER: Indexable?

THE WITNESS: Indexable. It's not my word and I wouldn't-that's the part of the web that can be found or located by the standard search engines like Google and Alta Vista, which use certain methods to find pages on the web, and that's what they could find.

And then another estimate has about 11 million web sites as a-that's from the OCLC.

BY MR. SMITH:

Q Okay. And can you just briefly explain the distinction between web pages and web sites?

A A web pages is basically-you have a browser, which is basically the software, the window through which you look at the web. A web page is basically anything that shows up in that browser. It might be text, it might be pictures, it might contain links or sound files or video files. But it's just what shows up in your window when you access a certain web address.

And the other was web site. A web site can be- well, you can use the word in two ways. It might be just all of the pages that are available at a particular server or computer that's linked to the internet. More often nowadays it's used to refer to the collection of pages or resources that are made available by a particular organization or company or person, whether they're located sometimes on more than one computer or whether they constitute only part of the content available on a particular computer.

JUDGE BECKER: Is a server a single computer, is that-

THE WITNESS: Yeah, a server is-

JUDGE BECKER: -or a network of computers?

THE WITNESS: A server is generally a-with certain caveats, a server is a single computer. So Stanford, you might speak of Stanford's web site, even though in fact Stanford has a number of different servers at which its-or IBM has a number of different servers at which its content is located.

You might also speak-I might say that I have a web site, even though what I have is basically a home page and a few articles that are part of the larger site at the center for the-

JUDGE BECKER: So if I want to look up Dr. Jeffrey Nunberg, I didn't have to look in here, I could have gone on the-

THE WITNESS: You could have gone on-

JUDGE BECKER: -I could have gone on the internet and I could have found your visage and your-

THE WITNESS: And my vita and my-

JUDGE BECKER: -vita and so forth.

THE WITNESS: -and qualified me remotely.

BY MR. SMITH:

Q Can you explain why it is that only some portion of the web is publicly indexable?

A There are a couple of reasons. Indexable is first of all a theoretical limitation. These web search engines like Google and Alta Vista locate content in the following way. They'll start with a certain number of pages and they'll take those pages and they'll put them on their server and they'll represent them in a certain form.

And then, if those pages contain links, they'll follow those links.

Q And what's a link?

A A link is a little bit of code in a web page that instructs a browser to go and retrieve another page that's at some other location. So it will follow, if it comes to a link, it will open that link-for instance, if it comes to my web page and there is a link to a paper, an article that I published in the American Prospect, the browser will-the robot that's used by the search engine will find my web page, see that link, click on that link so to speak and go to the American Prospect article.

And then it will take that and put it in its index. If there are links in that article, it will follow those and so on. It's like a chain letter.

JUDGE BECKER: Is that what the search engine does?

THE WITNESS: That's what the search engine does to compile the index of documents. The index is really the archive of documents that it stores on its computer.

BY MR. SMITH:

Q Right, but that's not actually the function of the search engine. What is the name of the process you just described?

A That's called spidering. And a spider is one of these little processes or robots, it's sometimes called, that goes out on the web and follows all the links and collects all the material and brings it back for storing on a single computer that's lodged, or set of computers in the case of Google, that's lodged at Google or one of the other search engines.

And then, when you give a request to the search engine for, say, Jeffrey Nunberg filters, it goes into its index, it see, oh, here's an article in The American Prospect that contains that contains the name Jeffrey Nunberg and filters, it returns the address of that article, and then you can click on that address and find the article itself.

Q Okay. And why is it that the spidering doesn't reach all of the web?

A Well, for several reasons. First of all, the spidering, even in the theoretically limited case, can only locate pages that are linked to from other pages. If, for instance, as is the case, I put my vita up at my web page-at my Stanford web pages, but don't put any links to it from my home page or any other pages, then the spider won't be able to get to it by linking.

It's also the case that many large sites, for instance a lot of news sites and other kinds of sites, will put instructions at the top level of their web site so that when the spider comes in to find content, the instructions will warn the spider off and say don't go any deeper in this web site. And although the search engines aren't obliged to conform to that, they, in fact, do.

So there's an enormous amount of content that just can't be located by any spidering process.

Q Now, this portion of the web that's not indexable, is that something that's accessible to people if they know where to find it?

A Oh, sure. You could find my-all I have to do is send you an e-mail with the address, the URL for my vita, you click on that or just enter it in your browser and it will find it. But there's no way to find it automatically by pursuing because there's no links to it.

Q And are there web sites that, as a deliberate policy, do not link to anything else but distribute their identity in other ways?

A There are plenty ways to, even if you want to have public access to your web site, there are plenty of ways to give access to what you could, as in the example I just suggested, send an e-mail to one individual. Or you could do a mass mailing to individuals.

You could-there are forms of links that don't show up to the spiders, but nonetheless can be followed. You might go to a web page and there might be what's called Java Script Program that basically works like a link but isn't the kind of thing that will show up to the spider. So there are lots of ways to even publicly distribute material on the web without establishing any explicit links to it.

JUDGE BECKER: What's that got to do with the issue that he's opining about with respect to the inability to block all-well, the under-blocking or the over-blocking?

MR. SMITH: It relates to the ability of the filters to avoid under-blocking because a web site that is not indexable, not reachable, doesn't ever show up on the-

JUDGE BECKER: But what I want, will you ask him that as a summary?

MR. SMITH: Sure. I thought you were asking me.

THE WITNESS: Should I just answer?

JUDGE BECKER: Treat my query to Mr. Smith as a question to you, Dr. Nunberg. Tinkers, baseball, tinkers to eavers (ph) to chance, okay.

JUDGE BARTLE: Do you see the link?

THE WITNESS: If you're going to do a filtering of web content in-however you do it, you've got to collect the pages you're going to put on a list of sites that satisfy whatever definition you're after. And in the course of collecting the pages, you're going to rely on search engines like Google and Alta Vista and so on. If they can't find a page, you are never going to get to look at that page to classify it one way or the other.

So if there's large amounts of content that aren't- and some people estimate that the amount of content, sometimes called the deep web that is not even in theory even accessible to the web-to the search engine, is two to ten times as much as is accessible.

All of that stuff can never be found by the search engines, so can't be harvested for purposes of categorization.

JUDGE BECKER: So it can't be blocked, right?

THE WITNESS: It can't be blocked, but it can't be found by these mechanical means that you have to use.

JUDGE BECKER: So if it can't be blocked-but if it can't be found, then what difference does it make?

THE WITNESS: No, it can't be found by the search engines. And I say, if I sent my vita, for instance, nobody will-even if one of these companies wanted to classify my vita, and I hope would not classify it as sexually explicit-

JUDGE FULLAM: Well, how could a library user get access to it if it can't be found?

THE WITNESS: Well, there are lots of ways. For example, e-mail. There are programs that go out, and I've had this happen to me, and I think many people have, that go out looking for new e-mail addresses and spamming everybody. That's sending messages to everybody that they collect with information about the new porn site.

This happens to me routinely since I put another web address up once on a-

JUDGE FULLAM: You lost me on the e-mail. Who sends the e-mail and to whom does it get sent?

THE WITNESS: Let's say I put up a site for a book I just published at earthlink, which in fact it happened to me, and I put an e-mail address at that site. There are programs that go around and look for e-mail addresses, and when they find an e-mail address, they just send you what's called spam. In this case an e-mail that says sexually explicit material or hot babes, whatever, for, you know, click on www.xxx.whatever. And this routinely happens.

I mean this is not-and it's a very easy way of publicizing your sexually explicit site, or your mortgage broker's site, or whatever the case may be, or your Viagra site without actually having to worry about anybody linking to your page.

JUDGE BECKER: So what you're saying is that a program that was blocking the mortgage broker or sexually explicit or what have you couldn't find this deep web material-

THE WITNESS: Right, there's no way to find it.

JUDGE BECKER: -and therefore it couldn't block it.

THE WITNESS: Right.

JUDGE BECKER: But it would nonetheless be accessible?

THE WITNESS: Easily accessible to anyone, anyone who had received one of these e-mails. And that's only one of the ways of publicizing these sites.

JUDGE BECKER: But this is an under-blocking then?

THE WITNESS: Yes, it would absolutely speak to under-blocking.

BY MR. SMITH:

Q Dr. Nunberg, are there any reliable estimates of the percentage of web sites in the publicly accessible part of the web that are sexually explicit?

A Reliable is a tricky word. There are two-there are a couple of estimates. There's one by OCLC which has done-

JUDGE BECKER: What is OCLC?

THE WITNESS: OCLC used to stand for Ohio Consortium, it's a-

JUDGE BECKER: Library Central.

THE WITNESS: -it's a group that does the cataloging. So what I think their name is just OCLC.

JUDGE BECKER: Can we stipulate what OCLC is?

MS. BHATTACHARYYA: Yes, Your Honor, if someone can tell us what it is.

MR. SMITH: My understanding is that it doesn't-

JUDGE BECKER: What is it, Ohio?

COUNSEL: Ohio College Library Center.

JUDGE FULLAM: Yes, it used to do all the central -

THE WITNESS: But I believe they no longer call-I think they just call themselves OCLC now. So I think they don't even use the -

MR. SMITH: So it now a required acronym, there is no words-

JUDGE FULLAM: It's what libraries go to to get information they don't have in their own libraries.

MR. SMITH: It is the organization that coordinates the inter-library loan.

THE WITNESS: They've been doing a survey for a number of years of the size of the internet, which involves sampling IT addresses, which are basically computer addresses, and seeing what's out there. They estimate-

JUDGE BECKER: Okay. So what did OCLC estimate?

THE WITNESS: OCLC estimated about one percent of sites dealt with-I can't remember the characterization they gave it, whether it was sexually explicitly content or pornographic content.

JUDGE BECKER: And how many would that be in number?

THE WITNESS: Well, if there are ten million sites, that would mean 100,000 sites-

JUDGE BECKER: That's a lot of sites.

THE WITNESS: -that dealt with sexually explicit content. There's another survey that was published in an article in Nature a couple of years ago that estimated that 1.5 percent of servers, that is to say the computers on the web, contained some amount of-and again, I don't remember whether the characterization-

JUDGE BECKER: You're talking about in the hard drive? Contained in their hard drive?

THE WITNESS: Well, contained, no, as pages, so if you have a server on the internet that has a number of different pages-a number of different kinds of content, that 1.5 percent of such servers of such computers, contain some amount of sexually explicit material. And it's hard to translate that into an estimate-

JUDGE BECKER: Where? Are you talking about in their hard drive? You're not talking about on web pages?

THE WITNESS: No, in the material that could be publicly accessed.

JUDGE BECKER: Okay. These are on web pages?

THE WITNESS: These are-a server is a computer that's on the web whose pages or part of whose content can be accessed remotely over the internet. So the part of the content that could be accessed over the internet contains sexually explicit material in 1.5 percent of the servers in their survey.

BY MR. SMITH:

Q Can you tell us how often it is that content tends to change on the web?

A There have been estimates that, if memory serves, that the average web site has a half-life of about 90 days. As I mentioned, the Nature article that appeared in 1999 estimated 1.5 million new pages per day.

Q What does it mean to say that the average web site has a half-life of 90 days?

A Most web sites just don't last that long. They appear, they disappear, people put up content. I think it's web page, I'm sorry, not web site.

The people put up content, they take it down, they revise it, they move it somewhere else. It's a very-

JUDGE BECKER: If it's not taken down, does it just remain?

THE WITNESS: So longer as the server is there or so long as the person continues, the owner of the account continues to pay their account and so long as the institution remains at the same institution, it will remain there.

BY MR. SMITH:

Q Now, the particular types of internet blocking software that you studied and educated yourself about, I think you said use lists of sites that they block?

A They have, they all operate with something called a control list, which is a list of sites that's been compiled by one of several methods that contains sites that are preclassified as belonging in certain categories, and in the case relevant here, as belonging in categories that they describe in various ways.

Some describe it as sexually explicit, some describe it as pornographic, some have several categories in this area, some have only one. But the control list is a precompiled list of sites to which categories have-or pages or addresses to which categories have been assigned.

Q Now, the actual entries on these control lists, what form do they take?

A Well, it might be helpful if I just could be a little professorial for a moment?

JUDGE FULLAM: Is there a microphone near there?

JUDGE BECKER: Yes, right there. Just be sure to speak into the microphone.

THE WITNESS: So for example, suppose there's a page containing the picture of Miss September at the Playboy site and I want to categorize it as being sexually explicit or something like that on my control list. That page will have a URL that might look like this (indicating).

This is continuing (indicating).

So this URL, which is a long string of characters, is the universal resource locator, this is the address of that particular page. And if I enter it into my browser, the browser will return that page.

And this URL, like most URLs, has several parts. It's got this part, http:// which tells you that-tells me the browser should use a certain, to locate the page, it should use a certain transfer protocol, in this case the http, transfer protocol.

It has this part, which is sometimes called the root URL or the root node, that's name of the computer on which, the server on which this page is located.

And it's got this part,/month/missseptember.html, this might be the file on that server in which the months are kept and this might be the name of that particular page.

So one way to block the page would be simply to say, when somebody enters this long URL, return a message like the one that you were seeing, you know, this page has been blocked for whatever reason.

Another way to block the page would be simply to enter-you say well, we're going to block everything at the Playboy site, so everything that has this top level URL or root URL will be blocked, and then this will be blocked by-

JUDGE BECKER: But what's the vehicle by which it's blocked?

THE WITNESS: The vehicle in the sense-

JUDGE BECKER: Yes, mechanically, how is it blocked?

THE WITNESS: If the URL is entered, the-well, there are several ways to implement these systems. It could be sitting on your -

JUDGE BECKER: I mean what does the blocking program do?

THE WITNESS: It could be sitting on your PC or it could be sitting, in which case, if you entered the-it could be installed on your PC, in which case if you entered this URL, it would look at the list it had and return, instead of going out to find this page, would return a message that said this page is not-

JUDGE BECKER: So that this URL would have to be on the blocking list?

THE WITNESS: It would have to be on the blocking list. Or, it could be the case that this particular URL isn't on the blocking list, but that this root URL playboy.com is on the blocking list, in which case it just says, well, it's in that.

Or there's a third possibility which I should mention. This, when you send a request to the internet, so to speak, to return this page, it looks at this www. playboy.com and it looks that up as a number of a particular server called the IP number, and that's a series of four, so that-I'm just guessing here, but 15 point 54 point 0 point 92 or something like that.

There would be some number associated with that computer. And for various reasons, and perhaps we'll come to it, might also-and most of the assistance also do list that number, so that this would be another way of blocking access to that particular computer on which this page was sitting.

So any three of these, and in fact all three, are used by filtering software companies whether the software is located on your PC or on a remote server or a proxy through which access, say to a library, is being monitored.

JUDGE BECKER: What do the numbers derive from?

THE WITNESS: This is the-this number-every server that's on the internet has an IP number (indicating), which is this series of four octets that can range from zero to 255.

JUDGE BECKER: So you would simply identify the server which has the punitively offending material.

THE WITNESS: Because if I entered instead of-I have no idea whether this is, in fact, the number of the server or not, but it could be. But if I entered just HTTP://15.64 et cetera, it would bring up-

JUDGE BECKER: HTTP means Hyper text transfer Protocol, right.

THE WITNESS: Absolutely, yes, and that's the- this is how you should go an access that number.

JUDGE BECKER: So if you did that you would eliminate every article in Playboy?

THE WITNESS: Right, and if I did that-and I could do this either by-

JUDGE BECKER: Whether it has explicit material-

THE WITNESS: Right-

JUDGE BECKER: -or sexual material

THE WITNESS: -anything that's at the Playboy site. If I pout in www.Stanford.edu-

JUDGE BECKER: All right. Fine.

THE WITNESS: -everything at Stanford be blocked.

JUDGE BECKER: So those numbers or the broader name is a blunter instrument but it blocks more.

THE WITNESS: It blocks more and it's a procedure I guess we're going to come to it, it's a procedure that all filtering software invariably makes use of in the course of blocking.

THE COURT: So if it got down to just blocking Ms. September it wouldn't block necessarily Ms. October?

THE WITNESS: Right, not then it would just block this particular page and Ms. October would be passed.

BY MR. SMITH:

Q Okay. Resume your seat, I think. One more sort of general issue I wanted you to discuss before we get to specifics.

MR. SMITH: And I think perhaps, Your Honor, I'll probably be able to do the generalities now, and we can probably start with a closed room for a few minutes in the morning if 5:00 is the cut off.

BY MR. SMITH:

Q You mentioned a lot in your report, the concepts of precision and recall and I wonder maybe even flipping another page you might tell us-

A Right.

Q -what those two terms mean?

(Pause in proceedings.)

THE WITNESS: Forgive me I need to write this down.

So these are two notions that are used in the evaluation of the classification or information retrieval systems uniformly and that obey certain very general laws that govern the operation of these systems.

If you're evaluating a system you want to know two things about it. First of all, of the total number of in this case pages for example that a system returns or classifies, how many are correctly classified?

So for example, suppose I have a system that is trying to find dog pictures. I have a set of pictures that contains a 100 pictures of animals and 200 of them are dog pictures, and I asked this system to go out and identify the dog pictures as opposed to other kinds of pictures it returns 100 pictures that it identifies as dog pictures and 80 of those are in fact pictures of dogs.

Then having returned 80 of the 100 pictures that were returned are correct, I would say it has a precision of 80 percent.

JUDGE BECKER: Except you said there were 200 dog pictures.

THE WITNESS: Right, well, now of the-

JUDGE BECKER: Oh, I see.

THE WITNESS: -it returned 100, right so, it said here are 100 dog pictures and I look and I say well, of these 100, 80 are in fact pictures of dogs so if it returned a page there's an 80 percent chance that it is a picture of a dog. But if you say another problem is that, there are 200 dog pictures it only returned 80 of those 200 dog pictures, so it has a recall that is to say, the total number of dog pictures in the domain of 40 percent.

BY MR. SMITH:

Q Which is calculated?

A Right, so this would be the percentage of the total -the total in the category I should say.

JUDGE FULLAM: The total that it should have gotten?

MR. SMITH: Right, that are returned.

There's inevitably a trade off whenever you're doing one of these systems between precision and recall. So in the case of my dog picture finder, I could say, well, you know I want to get more recall here so I'm going to loosen the criteria. So that now, anything that looks remotely like a dog picture is going to get returned, and my recall will start to climb up at that point because I'll be getting all these pictures that are only kind of marginally dog pictures.

JUDGE BECKER: And you may also get pictures of wolves so-

THE WITNESS: Exactly, wolves and foxes and lost ocelots and so on will come in.

If I want to increase the precision I say look I don't really want you to return anything that isn't absolutely a dog picture. The wolves and foxes will start to disappear but now some of the marginal dog pictures will go with it.

It's like the example I've given when I taught this is sort of like you're building a bird house and you're trying to make the opening the right size to let, you know so that all the sparrows can escape the hawks, and if you make it really small only sparrows can get in and the hawks can't but some sparrows are going to get eaten.

If you make it bigger all the sparrows can get ion but now some hawks will get in, and inevitably there's a trade off depending on how big you make that opening, how generously you define the category and that's just endemic for systems of this type.

BY MR. SMITH:

Q Okay. Now, does that trade off apply to kind of blocking software that's at issue that's in this case?

A Absolutely, it applies to every possible kind of software or system for that matter, or method that's involved in doing classification of any type.

* * * * *

BY MR. SMITH:

Q Now, Dr. Nunberg, in preparing your testimony, you studied the depositions of the various filtering companies?

A Yes.

Q And how many of those were there?

A There were three, one from N2H2, one from Cyber Patrol and one from Smart Filter.

* * * * *

A So again, let me stress that this is just a general procedure in classifying content on the web, whether it's a question of classifying sexual explicit content, material on baseball, appellate decisions or anything, you're going to follow the same procedures. You're first going to begin by harvesting from the web a large number of pages that are-or sites that are candidates for belonging in the category that you're interested in.

I'll mention these four general procedures generally, and then I'll specify. The second thing you're going to do is try an winnow that group into a smaller set that might be candidates for inclusion on your control list again by automatic means.

Third, you're going to have a review of what comes out of the winnowing process, possibly by hand of at least some of the sites that are turned out. And, fourth, having identified some sites as belonging to the suspect category, you're going to, let's say, categorize by generalization-by association.

JUDGE BECKER: But you're talking about the blocking company doing this on an individualized basis. This isn't a blunt key word kind of operation.

THE WITNESS: Well, there are four stages here, if I may, Your Honor, just go through these.

JUDGE BECKER: All right.

THE WITNESS: So the first thing you want to do is just find all of the pages that might contain sexually explicit content and the filtering companies as per their depositions use several methods for that.

The first is to go to these very large search engine indexes like Google, and to give them lists of-give them queries containing words that are liable to return numerous pages that belong to that category. These might be, as using Judge Bartle's example, you might just say, sex or hotbabes, or whatever, and take back all of the pages or some set of the pages that are returned by Google for that. So that would be using search engines.

The filtering companies might also go to particular pages that have directories of sexually explicit sites or what are called web rings that are kind of collections of pages that all deal with a certain topic, so there might be specific directories.

JUDGE BECKER: Does that mean they look at each one?

THE WITNESS: Well, no, for this purpose they're just trying to assume a candidate set-

JUDGE BECKER: Okay.

THE WITNESS: -and this is all going to be automatic.

JUDGE BECKER: This is the harvest.

THE WITNESS: They'll give the harvest-they'll go to Google and they'll say, give me all the pages that have to do with-that have the string anal sex in it, for example, everything that contains that string, Google will then look at.

Now, Google, I should mention, if you give the query, anal sex, to Google, it says there are 530,000 sites, the last I looked, that have that string of words in them. Google will actually only show you the first 2,000 of those sites, you can't get beyond that, the Google won't show it to you. But, nonetheless, that will give you at least 2,000 sites that contain that string and these other methods will give you others.

And then you may also have your own little spider like the one used by the search engine, but a much less robust version, that goes to some of these pages and clicks on the links in the pages returned by those to return more pages.

So with this you'll recover a large set of sites that you haven't looked at all, but that possibly might be in this category, and presumably lots of sites that-

JUDGE BECKER: This is a big enterprise to do this, a lot of person-power to do it, it takes a long time to do this?

THE WITNESS: My understand, again, from the depositions is that some of this is done mechanically. The filtering companies have staffs of in the dozens of people, I think that's a fair characterization. But they're not of the size-they're not in the order of the computational resources of staffs of companies like Google and Alta Vista which have hundreds of employees and-

JUDGE BECKER: Okay.

THE WITNESS: -hundreds-just representing the amount of content on the world web-the worldwide web takes hundreds of computers just to represent it, just to represent it all.

JUDGE BARTLE: But if you say there are 500,000 which have the words, anal sex, and they only tell you about 2,000 of them, what about the rest of them?

THE WITNESS: Those sites will be returned in the Google search. So you're only going to get to look-

JUDGE BARTLE: So that's not going to be much of a help for you. You eliminate 2,000.

THE WITNESS: Well, you might-now, I'm going to have to speak on the filtering-

JUDGE BARTLE: No, but you eliminate 2,000.

JUDGE FULLAM: Will the filter apply to the other 480,000?

THE WITNESS: The filter won't-if it uses only Google to locate pages, and there are these other methods that the filtering company uses, it-

JUDGE BARTLE: So you'll have to go to other methods, then.

THE WITNESS: -will not find any of the 480-or 528,000 pages that are not returned by Google that contain that string.

Now, it might also give numerous different strings and they might come back because they contain-those same pages might contain some other word or pair of words that will come up in the first 500 of the first 2,000 of Google, but no, in principle they won't be looked at.

So from this large set that's harvested by these methods, which will itself not contain many pages that are in fact contain these strings. The filtering companies will use their own software that will do some analysis on these texts that are returned. It may involve just key word searching as in Judge Bartle's example of just, give me all the pages that contain sex, or it maybe a bunch of 15 words or 30 words or 50 words.

JUDGE BECKER: Are we into winnowing?

THE WITNESS: We're into winnowing now.

JUDGE BECKER: Okay.

THE WITNESS: Because we want to take this large set that we've come back with which will contain many pages that are doubtless completely innocuous and restricted further.

So it may also use statistical methods.

JUDGE BECKER: Well, how would a key word search help to winnow?

THE WITNESS: Well if, for example, I've given the-I've got my original set of pages, and now I say, well, give me back only the pages that have the word cum and anal sex and whatever, some combination of six of these words or something like that, and just take that subset and those will be the candidates for further-

JUDGE BARTLE: Because theoretically it would return the word, Middlesex or something.

THE WITNESS: It might return the word, Middlesex, or certainly if you've searched on-

JUDGE BARTLE: Or if the site was hotstripper and it turned out it was a furniture refinishing company-

THE WITNESS: Right. In fact there's a one case that we have in our documents of a site-

JUDGE BARTLE: That's why I picked it up.

THE WITNESS: -a site called pen-well, It's the pen is mightier than the sword, is the name of the site, but the URL is just penismightier-

JUDGE BECKER: So it's penis-

THE WITNESS: Yeah, and it has the string penis in it. Since there's nothing else in this site, it's a site where discuss games and it's hard to find anything else in the site except some mild profanity that would trigger the classification for pornography. So-

JUDGE BECKER: All right. Or on hotstripper the keyword would show nothing by furnishing refinishing. It wouldn't have any other key-any-

THE WITNESS: Right. If that's the only stripper, perhaps-it depends how these things are-how this particular system is implemented.

JUDGE BECKER: Right.

THE WITNESS: But it could also use statistical techniques which in this case, for example, as I've said, there are 530,000 sites that contain a string, anal sex. And if you look at the first 2,000 of those, lots and lots of them are pornographic sites, not surprisingly.

If you restrict the search to unprotected anal sex, not surprisingly now the first sites that come up are-tend to be more safe sex sites of health sites than-since it's not a string. Some porn sites do use that string, not all, but that will tend to push things.

So I might say, well, anal sex, I want you to give a waiting of such and such unless it's proceeded by the word, unprotected, in which case, reduce the waiting by such and such depending-

JUDGE BECKER: Judge Fullam is ready to observe that up to now this is more like Computer Science 301 than rocket science, but anyway.

THE WITNESS: So these techniques are used to winnow the site. And all of these-these techniques are sometimes called-you sometimes see the terms, artificial intelligence or you hear talk of learning and so on and so forth. And I should just stress that these terms should be understood at best a metaphorical (inaudible).

You can build fairly sophisticated-we use them, certainly, statistical techniques that are capable of learning in the sense that they will assign certain waitings to certain words, and then when you give it correction it will adjust the waitings automatically and so on.

But these systems can't do, though the term artificial intelligence is often used in hype, these systems can't do what people ordinarily do. They can't say, oh, here's a word I wasn't thinking about and let's look at that word, or, oh, you know, when anal sex appears and the background is hot ink.

And so in the kind of generalization that people routinely make and the kind of gestalts that they routinely consult in making these classifications simply aren't available to this technology. Moreover I should say-

JUDGE BECKER: But what is the-is the artificial intelligence different from the statistical method?

THE WITNESS: Well, artificial intelligence is a fancy name given to statistical methods that can automatically correct the waitings they give to certain features when they're evaluating the site-

JUDGE BECKER: All right. We're not into a modality here.

THE WITNESS: It's-but the terms-because a term is used and implies a kind of human-like capacity, it's important to understand that that's not the case.

Moreover I should say that these systems-when you look at a page, you're not looking at all the helping words or the order of words or so on, you're just looking at what we call a-

JUDGE BECKER: I would just observe, Mr. Wolfsohn, that if any of this is proprietary, you better be prepared to tell us what's proprietary.

MR. WOLFSOHN: I completely agree, Your Honor, none of this is.

JUDGE BECKER: Okay.

MR. WOLFSOHN: And I appreciate counsel, plaintiffs' counsel's caution and prudence.

JUDGE BECKER: All right. Well, we'll-

THE WITNESS: Okay. So having winnowed the set down again by some set of automatic methods, now the set of pages returned can be submitted for hand review. Of the filtering-

JUDGE BARTLE: When you say, hand reviewed, you mean by a human being?

THE WITNESS: By a human being. The filtering company is -

JUDGE BARTLE: Hopefully using one's head, not just one's hands.

THE WITNESS: Head review, I should say. In the depositions given by the filtering companies in this case, two of the filtering companies claim that all pages were hand reviewed that came out of this process. N2H2 admitted that certain of the pages were not- not all the pages were hand reviewed.

In fact, if you look at just the size of the web and the number of pages you're going to be dealing with even given the use of these processes, it's difficult to believe that anyone can submit all-subject all of them to hand review.

JUDGE BARTLE: But even if you could, there's certainly a subjective element when you're talking about a review by human beings as to whether it's, you know, explicitly sexual material, isn't there?

THE WITNESS: Yeah. Well-

* * * * *

THE WITNESS: The depositions did say that review was done occasionally of the decisions made by individual reviewers, but the individual reviewer, as I understand the depositions in each case, when documents are reviewed by hand-by eye, by whatever, each reviewer is responsible to do a certain set of web pages and there may be controls and supervisors looking over how these people do this.

They have had varying degrees of training ranging in one case for three days, to perhaps several weeks training.

JUDGE FULLAM: Could you give an idea of approximately the number of pages that would be hand reviewed?

THE WITNESS: Do you mean how many pages by one person in a day or how many pages-

JUDGE FULLAM: Total.

THE WITNESS: Well, the filtering companies claim -two of the filtering companies who were deposed claimed that nothing appears on their control list except when it's been hand reviewed, subject to certain other reservations that I'll come to in the fourth step.

JUDGE FULLAM: I'm trying to find out how much they reviewed before they wound up with their control list. What number of pages are being reviewed?

THE WITNESS: It's difficult to say.

JUDGE BARTLE: It would depend on what the subject was, how much raw material there is out there, wouldn't it?

THE WITNESS: They're getting these pages back and then they're giving them by their own account two reviewers who are going over them and making the final judgment according to the categories that they have, which are different from one filtering company to the next.

JUDGE FULLAM: And you have no idea whether it's 25 or 25,000 or 25 million?

THE WITNESS: Well, the filtering companies claim anywhere between I think 200,000 and 600,000 pages or URLs on their control list.

JUDGE FULLAM: That answers my question. Thank you.

THE WITNESS: Some number in that range.

JUDGE FULLAM: Okay.

BY MR. SMITH:

Q Now-go ahead.

A So the other point to make about hand review is that when you look at what the companies-what actually appears on the control lists, it's just difficult to believe that these pages have been reviewed by anybody. I guess we'll come to that later.

Q Did we have access to the actual control lists?

A No. And the control lists, as I understand, are held to be proprietary by the filtering company, so we have no way of getting them.

Q So when you refer to looking at what shows up on the control lists, can you explain how you did that?

A If I, let's say, want to find out if Miss September's picture is on the N2H2 control list, there is a site at N2H2, so I could do one of two things. I could actually load the software in my machine and see what it does, or N2H2 makes available at its home page a page in which you can enter a URL, an address, and it will tell you how that page is classified by the N2H2 classified similarly of the other two filtering companies who were deposed.

Q Okay. Now, just for clarity-

JUDGE BECKER: Can I do that? Can anybody do that or do you have to have a key word or something, a password?

THE WITNESS: No. No, anybody can just go there.

BY MR. SMITH:

Q For clarity, is this a constantly updated process, this control list development?

A My understanding, again from the depositions, is that the companies are continuing to harvest pages and continuing to winnow pages. They do not, as in their depositions, they indicate they do not routinely go back and look at URLs again to see if content has changed, and content changes very rapidly, unless a request is specifically made by somebody for them to review a classification.

So the classification, once made in the default, remains as such.

Q And is the list periodically redistributed to their clients of the company?

A Again, as per the depositions, the list may be in several places. It may be at a server in the library that served all of its stations, or it may be at individual PCs and it is periodically updated.

But it's going to be updated-you have to understand, this process takes a while to compile these lists. Meanwhile, the search engines like Google are going out and very rapidly updating their indexes. So it may be weeks or even months before a new site, even if it does make it through this process, finds its way to the control list of one of the filtering companies.

That period of several months is, in this world of sexually explicit sites, a very long time.

JUDGE BARTLE: So in other words, if it's reviewed as of say September 1st, sexually explicit material, and that's-whatever is blocked is blocked, then of course there's changes occurring every day on the web sites and it's not reviewed again for a number of months-

THE WITNESS: In the default case, it will never be reviewed.

JUDGE BARTLE: -so in October and November, the world has changed in that period of time?

THE WITNESS: Right. To take the trivial we're talking about, when Miss October appears, let's say on October 1st-

JUDGE BARTLE: That won't be blocked because that hadn't even -

THE WITNESS: -it may not be until January or February if it's found at all, because it has to make its way through this process.

JUDGE BARTLE: -because that didn't even exist in September.

JUDGE BECKER: But, of course, if they block the over Playboy site, then they block-

JUDGE BARTLE: Exactly.

JUDGE BARTLE: -some good articles that people are entitled to get.

JUDGE BECKER: Right.

THE WITNESS: What happens in this fourth category, categorization by association-

BY MR. SMITH:

Q Well, why don't we ask you about that? What is that?

A In having identified certain pages, and let's say it's sexually explicit, now certain generalizations will be made. And one of them, for example, would be the case of inheritance blocking, what we called inheritance blocking.

This is the case where having identified, for example, that here's this site, missseptember.-here's this page that ends missseptember.html, it's sexually explicit, let's block everything at the root URL, that is to say www.playboy.com. Each of the filtering companies that were deposed in this matter, for example, does block everything at playboy.com, including interviews with Jimmy Carter, film reviews and the like.

JUDGE FULLAM: Well, he lusted after women, as I recall.

JUDGE BARTLE: Only in his heart.

MR. SMITH: You have to be a certain age to even remember that at this point, Your Honor.

JUDGE BARTLE: We're all at that age.

JUDGE BECKER: Well, I won't repeat that Mr. Smith comes from Washington and he would be expected to know that. I wouldn't do that-

THE WITNESS: So in this categorization by association, from the occurrence-and in the case of the depositions-I now have to refer to my own report, I can't remember which of the filtering companies it was that claimed at its deposition that the existence of a single sexually explicit page at a site was sufficient to block the entire site.

JUDGE BARTLE: Maybe that ought to be called by guilt by association.

THE WITNESS: Possibly, Your Honor. I'll let Your Honor decide.

BY MR. SMITH:

Q Okay. Have you completed your description of the four steps that they go through?

A Right, yes.

Q Okay. At this point, I would just like-

JUDGE BECKER: Can tell me again, categorization by association. I'm not quite-

THE WITNESS: That's the case where you say, okay, here with missseptember.h-or here was www. playboy.com/missseptember.html, rather than looking at every page in this site, we're just going to assume that given the presence of this picture in that site-

JUDGE BECKER: I see, okay.

THE WITNESS: -everything at the whole site is going to be blocked.

And there are others, they might also generalize it not just for the www.playboy.com, but to the server that contains, which would block-

JUDGE BECKER: All right.

JUDGE FULLAM: A recipe for over-blocking in other words?

THE WITNESS: Yes.

JUDGE BARTLE: So in other words, if they decide to block the Star Report, then I can say maybe everything emanating from the Government ought to be blocked?

JUDGE BECKER: The Kenneth Star Report.

THE WITNESS: Yes, I know. For example, that would be it, that would be a possibility.

JUDGE BECKER: Yeah, just as a more recent example than the Jimmy Carter example.

THE WITNESS: Well, that's an extreme, obviously it's extreme. But there are cases, for example, N2H2, in its deposition, admitted to blocking the entire site I mentioned, www.salon.com, it's an online magazine most of whose-all of the content of which is certainly, by any definition, protected speech. It's got a sex feature which contains slightly ribald stories but without profanity, none of which couldn't appear in Harpers or the Atlantic.

I mean they are mainstream, well-known journalists who write for it. The entire site is blocked because evidently, according to the depositions, there were complaints about some of the features that appeared in the sex column.

So there is routinely massive over-blocking that results from this generalization process.

BY MR. SMITH:

Q Number four on the list there, is that something -the software companies?

A They all do it, and its given the amount of content available and the amount that keeps coming on and the resources available to the filtering companies. There is no conceivable way to even try to undertake to block this content without this kind of over generalization.

JUDGE BARTLE: Well, then theoretically if you take that categorization when we were involved in the saga of recent years, there were certain articles in newspapers that one could argue were maybe sexually explicit material, and so then you could have blocked the whole New York Times and the Washington Post, I mean if you had taken that literally?

THE WITNESS: That would have been at the discretion of the filtering company.

JUDGE FULLAM: Do you think that they tend to draw a distinction between text and photographs?

THE WITNESS: One of the companies, N2H2, has specific text only property I think or category to which applies to some of the-I think to their pornography or sexually explicit category. The other two don't have any such category.

BY MR. SMITH:

Q Did the companies whose depositions you reviewed use image recognition software in any way?

A None of the three companies that depositions were taken in this case used image recognition software. I think all of them indicated that it was deficient now for these purposes. There is-

Q Let me interrupt you.

A Yes.

Q What exactly is image recognition?

A Image recognition software is software which rather than trying to identify let's say sexually explicit text, looks at images itself in the effort, in this case to identify sexually explicit images, although of course image recognition could be again just a case of-you could just be looking for pictures of dogs and, in fact, that's one of the test beds in terms of which the software has been developed.

At this stage of development extremely primitive. There is a company-there was at least one company in New York State that was marketing a product that claimed to be able to detect pornographic images, but if you understand the way the software works, you see it's extremely difficult to do this.

Right now what the software tries to do is to identify naked flesh, skin regions by looking at color temperature and reflectivity of areas on the screen. It does that with fair performance, which is to say perhaps 60, 70 percent precision in the best case.

That means that lots of things that aren't, in fact, naked flesh from sunsets-I was playing with one of these that we have at Stanford, and I started getting pictures of tapioca pudding, then it continued to classify this stuff as pornographic.

Even if it can identify naked flesh, it can't tell so it massively over-blocks so to speak. Even when it can identify naked flesh, it can't identify flesh that, for example, is-if it's in black and white it can't do anything with it. If it's taken under an odd light, a red colored light for instance, it can't do anything with it.

It can't tell the difference between a painting of Saint Sebastian-again, this is on the basis of my experiments with the Stanford-with the state of art system that my colleagues at Stanford had developed, it can't tell the difference between the paintings of Saint Sebastian and a pornographic image or for that matter, just a picture of a swimmer, Mark Spitz and a pornographic image. It can't tell men from women. It certainly couldn't begin to tell adults from children.

Q So at this point, image -

A At this point, it's just beyond-there's no possibility that could be used with any level of tolerance.

Q Okay. Now, returning again to the more text base software that-

A Right.

Q -you analyzed and whose depositions were taken, do you have an opinion about the ability of these four steps in this methodology to identify and categorize enough content on the web to avoid a great amount of under-blocking?

A No. There's inevitably going to be substantial amounts of under-blocking because each of the procedures here are subject to-is limited in what it can return.

But one thing, to take it from where we started, the publicly indexable web isn't the whole web. There's lots of content out there that even the search engines don't know about.

If the search engines do know about it, you're only going to get back a part of it when you give them your query, and when you try to follow links that follow other content, it turns out that like chain letters in a way they tend to converge rather than diverging. At a certain point you have reached a large number of sites.

So the harvesting process itself is limited to only a small proportion of the actual sexually explicit content that's available there.

The winnowing process is subject to the considerations of precision and recall that I mentioned earlier. If you set the-if you make the hole very wide, you're going to let lots of unacceptable stuff-lots of stuff through that shouldn't be blocked. If you set it narrowly, you're going to miss other material.

The review process is itself subject to various considerations that we've talked about, and then the guilt by association or categorization by association that is going to lead to over-blocking rather than under-blocking, so I-

Q Okay. Now, what about the argument that while it may not be perfect it's sufficient to identify those sites that a person is likely to encounter in every day use of the internet in the library or in their home?

A There are several points to make there. First of all, as I mentioned earlier, there's lot of sexually explicit content that it can be easily obtained.

I get invitations two or three times a day now since I put my-foolishly my address up on-my e-mail address up on a web site to visit sexually explicit content, which since I've been working this case, I check routinely. They are almost never at the search engines, the ones that-

Q Get you through the e-mail.

A -get me through the e-mail. So there's lot of ways to get the content.

THE WITNESS: I can't speak as an expert on the economics of pornography, Your Honor, but I mean when you go to these sites, there will be a bunch of sexually explicit pictures perhaps-

THE COURT: But you're not paying?

THE WITNESS: -you can click here for more. You won't pay to see the top level or maybe a few pages and then you-

BY MR. SMITH:

Q Are there some sites where you do have to give them a credit card number?

A There are plenty of sites that want you to give them a credit card number to get in. I haven't gone to that.

Q You're not an expert on this subject.

A I don't know what happens when you give them a credit card.

Q Are there sites that have advertising that they flash?

A There's some that have advertising, there's some that also serve as links to other sites, so there's lots of -can you repeat the last question?

Q Okay. You were talking about why it is that it's not sufficient to get to the sites that most people would encounter in their every day use of the-

A Right. The second point to bear in mind about pornographic sites in particular, adult sites but sexually explicit sites in particular, is that they have a very interesting and distinct pattern of use with them.

* * * * *

[TESTIMONY OF J. NUNBERG]
[March 26, 2002]
[pp. 5-47]

JUDGE BECKER: Is that because of the inability to count or because what you described yesterday as the dynamic nature of the web, all the new pages everyday?

THE WITNESS: It's both of them. It's very difficult just to know how big the web is, how much material there is out there. There's no reason to suppose that any of these filters block more than a small part of it, and the reason for that is that none of the search engines themselves index more than a small part of the web.

A recent article in science it was estimated that the best search engine index is only 16 percent of the indexable web, that is to say the web it could in theory find and that all of them taken together only find 50 percent of it.

BY MR. SMITH:

Q But if you came to the conclusion that the filter was blocking 90 or 95 percent of the sexually explicit sites that patrons actually tried to access, what would be the practical significance of that?

A Well, that's the sort of claim that you see in the advertising for filter companies, but I think you have to take it in context.

To hear that there are 100,000 sexually explicit sites on the web can be a dramatic and alarming figure, and it certainly is an awful lot. But reducing that even by 90 or 95 percent isn't equivalent to say reducing the number of dirty book stores in a jurisdiction by 90 or 95 percent where you might in that case assume that you had in some measure also reduced the access to this kind of material that people in the jurisdiction can have.

It's rather more like having 100 cable stations in a region and then reducing the number to ten. You reduce the amount of variety that people have but you probably wouldn't reduce t.v. viewership because any site is accessible to anyone. That's the first point. So, even if you've reduced it by 90 percent, there's still more than enough for people who want to find it to get it.

Moreover, I recall that yesterday we were talking about the special the unusual properties of these sites with regard to the kinds of surfing behavior we see where unlike a new site where somebody goes to one side gets the news, that's it for the day, people will go to one side and then another site, and then another site, they'll keep clicking until they can find something. If things are blocked, they'll continue to look for things until they can find things that aren't blocked because of the nature of the appetite for this kind of material.

One other point I should make about that. One percent figure of one percent of porn sites, you have to understand that given the behavior of people who are looking for this kind of material that one percent of porn sites doesn't translate into one percent of the web surfing public who have an interest in this because a relatively small number of people with this appetite can support a far larger number of sexually explicit sites than the equivalent number of people with an interest in horses or baseball or news or something like that.

JUDGE BECKER: Would you explain that answer again, I'm not sure I understand. You're talking about the appetite of proclivities of these folks?

THE WITNESS: Well, because the tendency for people who are interested in this kind of material to keep looking for more sites. I mean, unlike a news site you get your news then you go home. In this case you get some pornography, you get some more pornography and so forth.

So a relatively small number of people can support a relatively large number of sites in that sense. And the one-if one percent of the sites on the web traffic in sexually explicit material that doesn't mean that one percent of the people who are browsing the web are interested in that material. It's just a caution against interpreting the proportion of sexually explicit sites into a proportion web browsing public.

BY MR. SMITH:

Q Let me then turn to the issue of precision or

over-blocking and ask you again with respect to the same four step process that the filtering companies use. Does that in your opinion allow them to achieve a high degree of precision in terms of the accuracy of the sorting that they do of sexually explicit sites?

A No, because at each one of these states a certain number of sites that ought not to be blocked are going to necessarily leak through the process. We already saw that in the harvesting process which uses very basic key word searches enormous numbers of sites are going to be brought in that shouldn't properly belong on the list.

Q In the winnowing process similarly which uses other kinds of either key word or statistical methods again many more sites will be brought in then should be tagged.

At the review process if there is hand review of all of many of the sites, here again I think we'll come to that soon, there are various ways in which people are going to over label sites. Like there will be a tendency to label sites more often than-sites that ought not to be properly labeled, and then of course the categorization by association is necessarily going to include a lot more content and we'll come to that as well.

One other thing I should say is that even if the filters perform 100 percent of identifying all and only the sites that were on their categories, there would nonetheless be over-blocking, and the reason for that is that the categories that the filtering companies are using are defined byu the filtering companies in varying ways. Sometimes sexually explicit adults, sometimes sex, sometimes pornography but are always defined in ways that are broader then the categories specified in CIPA.

Q Okay. Now, Dr. Nunberg, you identified in your report did you not various categories of over-blocking, sources of over-blocking and I want to take you through each of those.

Referring to the chart that we've put up here what is automatic or automated classification errors, what is that as a source of over-blocking?

A Well, that simply refers to errors that are introduced when a site is classified as belonging to one of these suspect categories, sex or pornography or whatever the case may be on the basis of an automatic classifier that's looking simply at key words of at some statistical counting, but that doesn't properly belong in that category.

* * * * *

BY MR. SMITH:

Q Did you with respect to these categories of sources of over-blocking make an effort to identify some examples that you found?

A Right in each of these cases I went to various filters to try to find examples of the kinds of over-blocking involved. I should say two things, first of all that in as mush as the filtering companies don't announce why they've blocked this site, or as to what process they've blocked the site these are only hypotheses as to why just certain sites were blocked.

Second of all, what I was doing here was merely trying to find sites of cases f over-blocking that were illustrative or diagnostic of what I took to be the deficiencies of these systems on the basis of an examination of how they worked.

Q Okay.

JUDGE BECKER: Dr. Nunberg. how may hours did you spend in this investigation preparatory to this testimony? I mean was this something that you did in a half an hour, an hour or so, did you spend days at it?

THE WITNESS: No, it was a couple hours. I would say I looked at-let me tell you how I did some of these. For example if I thought that a system might be using key word searching, then you would say well this example I gave yesterday of unprotected anal sex, if they're just pulling up all the pages that have anal sex what I'll do is I'll give a query unprotected anal sex because on that-to that query I'll get a lot of pages that are health pages or say sex pages, and I'll feed those to the filter and if it starts blocking those I'll say ah-ha, I'll bet that's what's going on.

JUDGE BECKER: All right. We look at this statute that we're concerned about, it talks about blocking visual depictions. Now, how does a blocking system block out simply visual depictions? In other words let's assume there are no words with the picture. How can a blocking system determine that that fits under the definition of the statute as something that's obscene.

THE WITNESS: It can't.

JUDGE BECKER: How do you do that?

THE WITNESS: It can't none of the systems-

JUDGE BECKER: It's one thing if you're looking for a word.

THE WITNESS: None of the systems being considered here and almost none of the systems commercially available even make an effort to identify images as such.

JUDGE BECKER: But that's what we're talking about here.

THE WITNESS: Right.

JUDGE BECKER: We're not talking about just an article that uses the words anal sex for example, we're using depiction and if you have something that has no words, and that's what the statute is directed toward how in the world can it be blocked?

THE WITNESS: The only way I take it that people hope to do it is to locate pages that have this sort of suspect text on it and then block all the images on that page or block the page because it contains the text and on the assumption that the images-

JUDGE BECKER: But if you don't have words then you're home free.

THE WITNESS: If you don't have words you're home free, if your words are in a language that the filter companies aren't searching on you're home free. If you're words are in english and they're represented as an image rather than a text, that is to say if you have so to speak a painting of the words on the page which is something that frequently done you're home free because that also doesn't show up-

JUDGE BECKER: Yes, you-

THE WITNESS: -to the search I do.

JUDGE BECKER: -could just say here's some great art.

THE WITNESS: Well, that, or you could just take rather then saying porn sex, whatever, you could actually paint so to speak the words, you can take a picture of the words porn sex and represent it not as characters but just as little pixels just as little dots that represent the-

BY MR. SMITH:

Q The search engine wouldn't be able to read that?

A No, and this is actually something that's frequently done, people do it, companies do it all the time for example when they want to conceal their e-mail addresses. You can put an e-mail address on the web but rather than putting it down as text in which case an engine will find it, you put it down as an image just as so to speak a photograph of the text, and then the search engine-

JUDGE BECKER: And another-

THE WITNESS: -doesn't find it.

JUDGE BECKER: Another issues that concerns me that you haven't talked about is you're assuming that a definition remains static and these words that are in these statutes are subject to Court interpretation and you may have one Court in Philadelphia interpreting the word one way, and a Court in California a different way so how would you-you not only have new pages added to the web everyday, and changes, but you may have Court's interpreting these words differently and that's the definition, so you sort of have a moving target don't you in terms of what the words mean?

THE WITNESS: Yes-

JUDGE BECKER: I mean it's not like saying some words have very definitive definitions and no body will dispute it, but when you come to obscenity child pornography and what's harmful to minors even with the definitions in the statutes they're to some extent moving targets, aren't they?

THE WITNESS: Yes, but even if they weren't the technology now is at a point where even identifying a category like pornography or explicit sexual content or representations of sex is well beyond the capacity of the technology.

To imagine that this technology now, or ever would be in a position to look even at a text and say this in the judgment of the average person appeals to all and largely in part to a puree at interest, I mean to make a judgment of that sort is just an absurd requirement to place on the technology. It's just beyond the capacity-

JUDGE BECKER: Or like Justice Stewart -

THE WITNESS: - of the technology.

JUDGE BECKER: -didn't he say, I don't know how to define it, but I know it when I see it.

THE WITNESS: And this technology can't see it.

JUDGE BECKER: Okay. Go ahead, Mr. Smith.

BY MR. SMITH:

Q Just for clarification, Dr. Nunberg the two hours you spent was on the looking for the examples that-

A Yes, two or three, I mean I would say two or three or four hours, several hours I spent looking for sites and just using these rough euristics (ph) to pull down sites and test them. I must have I would say I pulled down 250 to 300 sites in the course of looking for sites that would illustrate the various deficiencies of the search engines, of the filters.

Q I take it you spent considerably more time writing your 86 page expert report?

A Yes.

MR. SMITH: Now, we're going to be working with exhibits starting with Exhibit 73, Your Honors, which is all in volume four. Could you turn to Plaintiff's Exhibit 73, Dr. Nunberg and tell us, does that represent an example of what you believe is an automatic classification error in over-blocking?

A Right, this in fact was a site that someone had sent to me just as a joke someone had sent me the URL for this site and because I knew about it I ran it through the pages at the home-at the sites of the various filtering companies to see how they dealt with this site.

What it is in fact, it's called Kittyporn.

Q K-I-T-T-Y.

A Kitty, yes, as in pronounced K-I-T-T-Y and it's a spoof site protesting that 2000 times a day pictures of naked cats are posted on the internet and each day it is estimated as many as 20 pictures of young cats go up on the internet without their knowledge or consent with a computer and phone line even children can easily find and view these picturers a frightening new trend and so forth, and then there are pictures of kittens some of them bound in yarn and so on.

Q And what is the first page of Exhibit 73?

A The first page if Exhibit 73 is the print out which is time dated and stamped, of the response of the Smart Filter product when it was handed this URL.

So I went to the Smart Filter site, I gave it the URL for this particular page and it responded that the page was currently categorized as sex. This is as of October 1st, 2001 when I did this study.

JUDGE FULLAM: Excuse me, what? The first page of Exhibt 73 you say is dated and stamped, where is it?

THE WITNESS: Oh, I'm sorry, I printed it out at the top, if you look at the upper right, as of Monday October 1st at the time at which I printed this out. On the very upper right, the heading of the first page.

JUDGE FULLAM: All right. Okay.

BY MR. SMITH:

Q So all of this work you did with respect to examples is work that shows what was blocked as of about-

A As of that date-

Q -the 1st?

A -I have no way of knowing whether this is still blocked and so forth, but this page is blocked as sex from Smart Filter.

Now, my assumption is that no human reviewer would have looked at this page which is manifestly a spoof and classified it as sex. It could be of course that's it's blocked-so the question is why have they blocked this as sex. Did a human reviewer make an error or make a decision about it? That's just not likely.

Could it have been that it inherited it's classification from the root URL, the site to which it belongs? No, because when you look at just that corporation .com, that has another classification, not sex.

So my assumption is that an automatic process just went through saw these words porn and so forth and classified it as sex without anyone's going over it and checking it.

Q Okay. How about Exhibit 74, is that another example?

A Right, this is a site that was classified by N2H2 in it's URL checker, the checker it makes available on it's web page to find the classification of various sites as adults only and pornography.

What the site is in fact is the home page of the Center For Sex For Search at the College of Social and Behavioral Sciences, of California State University Northridge. This is an academic research group in sex. It's patently not there's nothing pornographic about this site, but again my assumption is that this had to have happen because an automatic classifier saw the word sex and whatever else might be on this sexuality or so forth classified it as pornography and no one reviewed it.

Why do I assume in this case it was a case of inheriting the classification from the root note, because the root note is the home page of the California State University of Northridge and that is not classified as pornography.

Q Okay. I believe -

JUDGE FULLAM: Just - I'm sure that the answers obvious to everyone but me, but if these sites were blocked how did you get the content?

THE WITNESS: I wasn't using the filter at the time.

JUDGE FULLAM: I see.

THE WITNESS: What I was doing rather was using an unfiltered browser to pick up the pages, then taking the URL and giving it to this page that's made available-

JUDGE FULLAM: Filter, I see okay.

THE WITNESS: - at N2H2 which then tells you what there classification of the site is?

BY MR. SMITH:

Q So in each case you've combined the answer about what is categorized. That's the first page of the exhibt with an actual image-

A Right, right-

Q - of the-

A -I should have said that.

Q - web site.

Now, I believe you touched a little on Exhibit 75 yesterday, this is the penismightier site, can you tell us why you concluded this was another example of automatic misclassification?

A Well, here again, this is a site that it's a kind of chat site or devoted to devotees of video games and electronic games in which they exchange e-mail, and so forth. There is some vaguely ribald language in this, but no content that could be remotely judged as sex or pornographic. There some people use the words ass and so on and it's classified by N2H2 as indicted on the second page of the exhibit as pornography and by Smart Filter as sex.

The only reason I could assume that I could imagine for that is that the URL for the site which is penis mightier, contains the string penis, P-E-N-I-S and that that was the reason for the misclassification, now the word ass appears on the second page, it could have been that but whatever the reason for the misclassification here again I have to assume that no human reviewer could possibly have looked at this site and decided that this was an instance of either sex as Smart Filter defines it, or pornography as N2H2 defines it, that rather this was a mistake just an error in machine classification which passed trough without anybody checking it.

Q Okay. And let me ask you them to turn over to Exhibit 80, and can you tell us what you determined about his example?

A This was a particularly disturbing example for me maybe because I'm the parent of a teenage girl.

JUDGE FULLAM: I'm sorry, which one are you looking at?

THE WITNESS: I'm - is this -

JUDGE FULLAM: Eight zero.

MR. SMITH: Eight zero should be opposite of sex.

THE WITNESS: Oh, I'm sorry, no I have 79, you're right. Yes. This seems to me a pretty patent example of machine misclassification.

It's the home page of the movie The Opposite of Sex, which was a major Sony Pictures film that came out a few years ago. I did check the rating, it was an R rated film, but it was an, in general, theatrical release. It was classified, this page, by N2H2 as pornography.

Again, it's inconceivable to me that anyone could have gone and looked at this page and decided that this was pornographic, and so that it must have been an example of machine misclassification.

I suppose it's possible somebody racing through at high speed would see a picture of this woman wearing a bathing suit and lying on a couch and say, well, that's pornography and so on. But my assumption here is that was machine misclassification.

Q Okay. The second category you've listed on the chart is over-zealousness. Can you tell us what that is?

A Over-zealousness is what happens when someone does look at a site or a page, and isn't sure how to classify it and errs on the side of caution, a phrase that came up several times in the depositions of the filter companies, says, well, it might be that some of our customers would be upset at seeing this material.

So rather than let it pass, even though it's really not in that category, we'll flag it as pornographic or sex or whatever as the case may be.

Q Okay. And turning to Exhibit 70-

JUDGE BECKER: It's not really over-zealousness, is it? It's over-assiduousness or something. You're not suggesting that there's a-

THE WITNESS: CYA would be the phrase that some people would use.

JUDGE BECKER: Okay. Well, that's an acronym we all understand.

MR. SMITH: We won't need to put that on the glossary?

JUDGE BECKER: You don't put that in the glossary.

JUDGE BARTLE: Christian Youth Association.

BY MR. SMITH:

Q Turning to Exhibit 77, is that an example of the second category then? Unlike everybody else, you don't have a binder, I'm afraid.

A Right.

(Pause in proceedings.)

JUDGE BECKER: We can loan him a binder if that will speed it up.

THE WITNESS: Here we go, I got it.

Right. This is the text of a little piece, a little commentary I suppose you would call it, that appeared in the site salon.com. In the column of the site, the feature of the site that's called sex, it's typical of the features that appear in that particular column. It's written by David Thompson, who is a very well-known film historian and novelist, and was a best-selling novelist at one point.

The story is-I suppose you could say it's ribald, there are no profane words and certainly no pictures associated with it and no photographs associated with it, in which the writer and his partner discuss whether Bush or Gore is sexier, whether it's sexier to speak in full sentences or sentence fragments.

It's clearly intended as a ribald piece, and again, the sort of thing that would easily appear in Harper's or the Atlantic or-I should say that this was categorized by N2H2 as pornography, and by Smart Filter as sex. In both cases, my assumption is that this was caused by somebody's over- zealousness, seeing something appear or-seeing something appear in a feature that's called sex and merely blocking everything in that feature.

I should also note that one hypothesis could have been that N2H2 blocked this because they block everything at salon.com, which is a general online magazine that includes articles by the likes of Orville Shell and Arthur Schlessinger and so on. That's all blocked.

But it's blocked under another category, it's not blocked under the pornography category. So my assumption is that in the case of the particular page, somebody at N2H2 went and looked at this page and said this is pornography and should be blocked.

BY MR. SMITH:

Q What is the N2H2 category under which the blocked the rest of salon.com?

A I think it's sex.

Q Is that like-

A I would have to look at my-

Q Okay. Now Exhibit 79, is that another example you would site of over-assiduousness?

A Right. This is the one that I said was particularly troubling. This is a very good site that's put up by Planned Parenthood, Ask the Experts Teen Wire. It's a sex advice site for teenagers. It's well-written, informative, and so forth, and exactly the sort of resource that the web has made very useful.

But one of the wonderful things about the web is that kids who may be timid about asking advice about these matters from parents or guardians or teachers, can now go on the web and get information about topics like this. This came up, in fact, because anal sex is mentioned on this page and it makes the point of the importance of using protection during anal sex, again, in terms that a teenager might understand.

It is blocked by Surf Control as adult sexually explicit, and it's blocked by N2H2 as sex, or it's categorized by N2H2 as sex, which is one of the categories that N2H2 on its home page advises should be activated in order to conform with the CIPA Act.

So an adolescent in a library who wanted access to this page in which-working in a library in which either Surf Control or N2H2 was activated, would not be able to get access to this page, nor, I should say to another example that was in my report that is not here, on another very good teen sex advice site that's sponsored by a group at Rutgers University.

Q Okay. Now what is the third category, inheritance? I think you've already talked about, but just for the record.

A Inheritance is the case we talked about in which rather than blocking Miss September, Miss October and so on, the filtering company just says, well, we're just going to block everything at this site, whether it's Playboy or salon.com or a sexually explicit site.

And this necessarily introduces very-a large number of over-blocking errors since people haven't checked all of the pages of these sites. And many, many times, in the case of Salon, there's nothing at this site that should be so categorized; in the case of Playboy, a large amount of content that shouldn't be categorized in this way.

Q If you could refer to Exhibit 71, is that an example of category three?

A Right. 71 is a page that happens to be a block by all three of the filters I looked at. It's classified by Surf Control as adult sexually explicit. It's classified by N2H2 as adults only and sex. And it's classified by Smart Filter as sex.

What it is is the text of a study that was published as a book called-it's the home page of this text, or the top level page-M as for mutual, A as for Ax, Male Sex Work and AIDS in Canada. It reports a study that, as described on the first page of the printout of the web page, was co-published by Health Canada, AIDS Vancouver, the HIV Social Behavioral and Epidemiological Studies Unit of the Faculty of Medicine at the University of Toronto.

The page was-so it's clearly a serious report and one that should be made accessible to users. It was placed, however, on the site of the Sex Workers Alliance of Vancouver. Now, the Sex Workers Alliance of Vancouver is, again, a serious site, dedicated to social and health issues involving sex workers.

It does, however, contain-while it contains itself no sexually explicit content, it contains in other pages links to pages that have sexually explicit content. My assumption here is that this particular text was blocked because this entire site, Sex Workers of Vancouver's site, was placed on the control list of all three of these companies who didn't bother to look through all of the pages to see, well, there's this piece of content that certainly shouldn't be blocked.

Q Just one more example, refer to Exhibit 72 and tell us what your analysis there was?

A This is a site that was blocked-this is a page, I should say, that was blocked by N2H2 as adults only pornography, and it does add text spoken only, and was classified by Surf Control as adult sexually explicit.

What the page itself actually contains is an open letter to Senator Hatch asking Senator Hatch to reconsider Section 1402 of the Juvenile Justice Bill which, according to the letter, is unlikely to withstand first amendment scrutiny. This is a clause that I take as the ancestor of the clause that made it into the CIPA provision mandating the use of filters.

Clearly, not by any stretch of the imagination, is there any content in this particular page that would be either sexually explicit or pornography. The reason it appears on the control list of these companies is that the particular site in which it appears is one that's dedicated to, in its own terms, erotic education, I think was the phrase, and contains a good deal of textual erotic material but no sexually explicit depictions.

Q Now the fourth category is listed as IP address over-blocking. Can you tell us what you mean by that?

A Right. IP address over-blocking is what happens when, in addition to blocking the root level URL of a site, that is to say the whole site, the filtering company also blocks the IP address. That is to say that four numeral sequence that's associated with the server or host at which the site is located.

This is done, as I suggested, because people could just enter the IP address rather than entering the URL. But it's very often the case that a single IP address or a single host will, in fact, be the host to a number of different sites, one of which might be a sexually explicit site but many of which might not be sexually explicit sites. And they will inherit this classification unless somebody explicitly goes in and changes the classification. They will inherit this classification and their sites will be blocked.

Again, it's kind of collateral damage of the fact that these same computers host one site that's sexually explicit.

Q And is this a relatively common problem with the filtering companies?

A In the submissions from one of the filtering companies, we received logs of the various requests that the company had received to unblock or re-categorize sites. In the seven week period covered by these logs, there were 563 requests that the filter company classified as requests-these were granted, the re-categorizations that were made because the filtering company said that they had blocked the IP address and this had been a collateral casualty of that process.

So if it's happening at a rate of 50 a week, say, for one filtering company-and this is people who have actually noticed that their site is categorized in an inappropriate way and have taken the trouble to write to the filtering company and ask that it be revised.

My assumption is that if that's happening 50 times a week with one company, it's happening an awful lot.

Q Okay. Now the fifth category is loophole sites. Can you tell us what you mean by that?

A Loophole sites is a phrase used by some of the filtering companies, and other phrases are used as well, to refer to a particular class of sites that are very important on the internet.

These are sites that for one reason or another return a URL different from the URL you've entered. So normally if I enter www.playboy.com, that will pass through-that will pass through the filter and it will look and see www.playboy.com and block it.

There are certain circumstances, though, in which I might get a site, go to a site that will go off and get another set for me and return the URL not of the original site but of its own URL. So that is to say at least will return a URL different from-

JUDGE BECKER: Why does that happen?

THE WITNESS: Well, this happens in several cases. The first case is what are called anonymizer (ph) sites. These are sites that enable users to get access to a URL without revealing their own location, what computer they're going from. And there are lots of reasons for wanting that kind of anonymous access.

People might want to have access, for instance, to a politically sensitive site, which particularly has been an issue for foreign nationals living in the United States who may want to have access to a site in their home country that they fear may be monitored.

It might be the case that you're an employee of a company and you want to have access to a site without having someone know that, it's critical of the company, without having your employer know that you've gone there. There are lots of reasons for doing this.

The point is that because such anonymizer sites return a URL that's distinct from the URL of the site you've requested, the filtering companies just block those sites categorically because the URLs that these sites return couldn't possibly be on their control list.

There are several other classes of these-

Q Loopholes?

A -loophole sites. A second is translation sites.

If I want, for instance if I'm interested in the text of an article about tropical diseases that was published in Latin American medical journals, let's say, and I've done a search either on the Latin name of a disease or maybe there's often an English language abstract, I might find an article written in Spanish and want to give it to one of these translation sites to translate for me.

Now, these translation software are something I've worked on, they're not that good, but they're often good enough to figure out, is this something I should pay more attention to or take the trouble to get translated in a serious way, or sometimes you're just interested in very simple information and you can get it from the translation.

But, again, what these sites do is, I'll give the URL of that Latin American medical journal to the translation site, it will go off, look at the site, produce a translation and return that site-that new site to me with its own URL, not with the original site. So, again, it changes the URL of the site I requested. It's also on the loophole list.

A third category of loophole sites are cache sites or archive sites. In particular, the Google archive. What Google does is to make a backup of every site that it finds in the course of its indexing and store that as what's called a Google cache site.

And these sites are given new URLs.

JUDGE BECKER: What was the word you used, cash?

THE WITNESS: C-A-C-H-E, like -

JUDGE BECKER: Oh, cache.

THE WITNESS: Yes.

JUDGE BECKER: C-A-C-H-E, all right.

THE WITNESS: Or archive, we can say archive. It's just a kind of Google archive.

There's another one of these called the Internet Archive, it's a nonprofit in San Francisco that has I think now 10 billion web pages stored in this form.

JUDGE BECKER: How many?

THE WITNESS: Ten billion web pages are stored in this form.

This is crucial resource for anybody who uses the web routinely for resource-for research because these pages disappear so rapidly that if you want to find something that's been taken down, the only place to get it is from the Google cache file.

It might be a question, for example, of an article in a magazine that appeared in March and has since been taken down or is no longer available at this site. It might be a question of the home page.

This happened to me just yesterday, I was trying to find the home page of a company to get information about some electronic equipment I had and the company had gone out of business, so I couldn't get anything from the company site. But Google had cached a version, an older version of that site and I was able to get information about the product that they had put out that had since been taken down.

These, again, are always considered loophole sites and blocked or-if that category said, a block by all of the filters because you couldn't-the filters just wouldn't work if you could use these sites, although you couldn't just block the sexually explicit sites because people could use these to presumably get to the sexually explicit sites.

One other smaller category, but very important one, is the Google image-

JUDGE BECKER: Would you clarify that? Judge Fullam and I-

JUDGE FULLAM: You have spoken at great length and very rapidly, but it doesn't make any sense whatsoever.

THE WITNESS: I'm sorry. I speak to fast. My students are always complaining.

The idea is this, that these are different kinds of sites-

JUDGE BECKER: Just give us the general notion of, the umbrella definition of loophole.

THE WITNESS: Okay. A loophole site is simply a site that returns not a page that had the URL that you requested, but another page that has the same content, or in the case of a translation site, a translated version of the content-

JUDGE FULLAM: Now, which is the loop-

THE WITNESS: -that has a different URL.

JUDGE FULLAM: Which is the loophole site, the one that you requested or the one that you got?

THE WITNESS: No, the loophole site will be any site that's set up to do this. Any translation site is a loophole site. The Google cache site in this sense is-or any Google cache page would be an example of a loophole site.

The anonymizer site, or I should add also the Google Image Search. Google has a huge repertory or repository of images that are available. There are 11,000 pictures of the Eiffel Tower, for example, at Google if you're interested in such things.

I went to find-I had actually an image of myself that had appeared on my home page and I was looking in the Google Image file to see if it was there, and in fact, Google had backed up that image. And when I tried to bring it-when I gave the URL for that image to N2H2, it came back and said this is pornography.

And the reason it said that my image was pornography is because the URL associated with it had Google Image in the beginning, and N2H2 was just set up to just block anything that has that-

JUDGE BECKER: I don't understand. Sexually looking guy, is that the -

THE WITNESS: Evidently.

BY MR. SMITH:

Q But the basic concept is it's sites where you can get access to another web site that might be blocked but with a different address attached to it so that it doesn't get blocked?

A That's it. They just change the address of the site for one reason or another and return it to you with a new address. And since that address won't be on the control list, the filtering companies deal with these just in a categorical way and they block them all.

Q Okay. Now, how significant is this particular category of over-blocking?

A This is a huge category. I mean if you think about the amount of content to which access is blocked or impeded, it dwarfs everything else put together. The Google cache pages represent hundreds of millions of pages of content, some of which is available in no other form.

Content in languages other than English on the worldwide web, this is research I've done, is now at about 30 percent and growing much more rapidly than English content. No language represents-other language represents more than a small fraction of that.

So if I want access to all of the material in Spanish and Italian and French and German and Russian that's on the web, and I want to find out actually what it means, the only way to do that is through the translation sites.

Q You can get the material, you just can't read it?

A No, I could get this Russian page, but it won't help me.

The anonymizer sites perform useful and sometimes critical functions, particularly in cases of, as I say, employees or foreign nationals and so on.

JUDGE BARTLE: Well, we have obviously a large Spanish population in this country where Spanish is the native language for these people. Do these blocking services take words which may have pornographic significance in English and get the Spanish equivalent and then search for articles in the Spanish language which should be blocked, or is it just English?

THE WITNESS: In the depositions-I can't speak to what all the companies do. In the depositions, one of the companies, Cyber Patrol-Surf Control, I guess, did indicate that they had out-sourced to other companies some identification of pages in other languages, classification of pages in other languages. And I don't know about the other companies.

But it's not going to cover the range of languages. Moreover-

JUDGE BECKER: We heard about the library, I believe it was in Wisconsin which has a lot of Norwegian documentation because a lot of Norwegian immigrants came to that part of our country I guess in the late 19th Century, early 20th Century. So I guess it wouldn't take-it wouldn't deal with whatever is in the Norwegian-

THE WITNESS: You would have to reproduce the entire procedure for Norwegian and Swedish and so on.

Moreover, certain languages, a lot of the sexually explicit content comes from sites in Asia, and the languages of Asia present particular problems because they're not alphabetic. And for various technical reasons, it's extremely difficult to do these procedures on languages like Japanese and Korean and Chinese which just don't use alphabetic script.

JUDGE FULLAM: Does this mean that with respect to these foreign languages, it results in under-blocking or over-blocking?

THE WITNESS: Well, depending, presumably-

JUDGE FULLAM: Speaking for myself.

THE WITNESS: -under-blocking.

JUDGE FULLAM: Under-blocking.

THE WITNESS: Presumably there's lots of content, let's say, in Chinese language web pages or Thailand.

JUDGE FULLAM: By and large, under-blocking is not a constitutional issue. Over-blocking is.

BY MR. SMITH:

Q Now, have you attempted to quantify the overall numerical precision rate of the various filter products that you looked at?

A No, I haven't. The-for a couple of reasons. In the first place, to do that properly, particularly if you don't have-I mean, the easiest way to do it is look at the control list, see what's on the control lists and classify that. But we can't get the control lists from the filtering companies, nor can we get direct access to the various procedures they use in compiling these lists, the search words and so forth.

Another way to do that would be to establish-to do a proper study with a stratified sample, but I estimate that that would require a huge number of web pages, 50 to 100,000 web pages to do it properly and I haven't done that.

More important, I think it's very-it's crucial to understand the precision and recall while those are the measures that we use in evaluating the performance of systems and appropriately relevant to certain goals, aren't going to be the crucial measures when what you're concerned with is the over-blocking of protected speech.

Q Okay. Now, if you were going to calculate a precision rate, what's the right way to do it? What is the formula that you would use?

A Well, let's say you had a number of sites that have been returned. In the example I used yesterday, you have a hundred pictures that a system has identified as pictures of dogs in your database and, let's say, I forget the number I used yesterday, but 80 of them are, in fact, pictures of dogs.

Then you would take the total number that gets returned, take the proportion that were pictures of dogs and take that percentage in this case, then 80 percent.

Q What about the option of starting with a group of sites which you had identified that sites that shouldn't be blocked, running them through the filter and then calculation the percentage of these innocuous sites that are blocked or not blocked?

A So I would, in this case, take a hundred sites, say they were just randomly assembled that were innocuous sites and saw how many of them, in fact, were blocked by the builders?

Q Right. If you did that, what would be the significance of a finding that say only one percent of those sites were blocked?

A Well, there are two problems with that. The first is that you have to understand that none of the filtering companies have in their index, have in their control list more-or have looked at more than a tiny proportion of all the sites on the web.

So if I am randomly grabbing innocuous sites, the odds are that no filtering companies looked at it just because nobody looked at more than a tiny proportion of the sites on the web.

So if I looked at a hundred randomly chosen sites and I found that, let's say, one of them were blocked or two of them were blocked, I might be alarmed because I would say that the odds that anybody's even looked at these sites would be very small. So probably they only looked at three of these sites and one of them is blocked, that is a high percentage.

Q Okay. And even if you factored that out, if you assume that they had looked at all 100 and categorized them in some way, would it be an indication of accuracy if you found in that set of 100 sites that only one percent of them were improperly blocked?

A All right. So let's say we have a filter and we-and it categorizes all of the sexually explicit sites correctly, and moreover it categorizes only one out of a hundred non-sexually explicit sites as sexually explicit, so that looks like a good figure.

But what you have to bear in mind whenever you're talking about a sample of this sort where one category is statistically far more predominant than another, is that the effects of that are multiplied by the size of the category.

That is to say, if there are one percent of the sites on the worldwide web that are sexually explicit sites and 99 percent are not sexually explicit, then categorizing one percent of those non-sexually explicit sites as sexually explicit actually brings your precision down to 50 percent. Let me give a statistical-

Q I think the numbers are going to help.

A Okay. So let's say I have 10,000 sites, just a sample of web sites, 10,000 sites, all of which have been looked at by filter, and let's say that it has-so in that 10,000 site sample I'm going to expect to find a hundred sexually explicit sites, that's to say one percent of the total, let's say it gets all of those rights, so it's got a hundred sites classified as sexually explicit and they're all properly classified as sexually explicit.

Now, let's say also that it misclassifies one percent of the sites that are not sexually explicit is sexually explicit. Well, that's 9,900 sites and if it misses one percent of those, it's going to 99 sites that it's misclassified as sexually explicit.

So the precision of this filter is going to be a hundred sites that it got right, plus 99 sites that it got wrong, or slightly better than 50 percent.

Q And what would tell you about the number-the percentage of errors on the control list if that were the facts?

A I'm not sure-

Q The percentage of accurately-

A Well, in that case, on the control list 50 percent-I mean the precision of-the actual precision of the control list even though is only misidentified one percent of thenon-sexually explicit sites given that disproportion the percentage of errors on the control list would be 50 percent.

Q Okay. Now, you've discussed at some length the number of flaws in the filtering technologies that you studied and I just want to ask in closing, is it your belief that these are only temporary problems that are going to go away with advances in technology?

A Well, certainly you can improve this technology.

JUDGE BECKER: Excuse me, but it would help if you would answer the question first, and then explain your answer rather -

THE WITNESS: Oh, I'm sorry.

JUDGE BECKER: -than go off on an explanation that doesn't tell us what the answer is.

THE WITNESS: No. No, I don't think these are simply temporary problems. You can improve the performance of these systems incrementally, but each new increment comes at a higher cost, and the reason for that is that in classification, as you try to make improvements you're dealing with harder and harder cases.

To take an example that I talked about yesterday, if you're looking at these dog pictures and you're trying to improve the performance, the better you get, the more you're having to deal with pictures of dogs that look like wolves or rule out pictures of wolves that look like dogs.

So it just gets harder and harder to make these improvements, and at a certain point you've reached the level where the technology-just given the limits of the technology just isn't capable of reproducing the kinds of subtle human judgments that are involved in this.

There are some tasks that people do that machines can do very well, because the tasks that people-the task could be reduced to a more or less mathematical for playing chess, for example.

It didn't take that long to build a machine that could beat a human at chess. It doesn't use the same- exactly the same strategies and processes, but it's something that could be reduced mathematically. If it's-a human capacity like-that involves subjective judgment as these do, then the technology just simply can't get there.

The other point to bear in mind that while small incremental improvements may be made in the performance of the filters, the number of the size of the control list is growing at an enormous rate.

So if you go from a 90 to 93 percent precision, that's good. But if in the course of that you've increased the size of the control list by ten times, then the amount of protected speech that's blocked by this has increased enormously in the process. I don't know if that-

Q In absolute terms?

A In absolute terms.

Q Okay. And just for the record, if you could just look at Exhibit 70-

MR. SMITH: Which I am not going to ask questions about, Your Honors, I just want him to identify it as an expert report.

BY MR. SMITH:

Q Is this the main text of your expert report that you submitted in this case?

JUDGE FULLAM: If you say so, we'll take your word.

THE WITNESS: Oh, 70 is the expert report? Yes.

JUDGE BECKER: Dr. Nunberg, how difficult would it be to design a filter that blocks only web sites that have been judicially determined to be obscene?

THE WITNESS: Well, if-that would be easy in the sense that if there are web sites that have been judicially determined, then that particular web site has been entered, the name of that web site is known, you could just say you can't have access to this site, that's easy provided that the web site maintains the same URL and the same address and so forth.

JUDGE FULLAM: And doesn't change the content?

THE WITNESS: And doesn't change the content.

JUDGE BARTLE: So all that someone would have to do would be to change it slightly, the it wouldn't be exactly what the Court had declared to be obscene?

THE WITNESS: Well, I'll leave that to you-

JUDGE BARTLE: I mean that's-

THE WITNESS: -Your Honors.

JUDGE BECKER: Just reviewing or recapitulating your testimony, what is the biggest problem, under-blocking or over-blocking?

THE WITNESS: I don't think that's a question you can answer in the abstract and I don't mean to be evasive, but I think it's a question of what your concern is. If your concern is getting maximum coverage and blocking anything that might conceivably be offensive, then you're going to stress under-blocking as a crucial problem.

If your concern is of not having protected speech blocked, then clearly over-blocking is going to be a problem. My sense is though that the filtering companies, in their advertising, tend to stress recall rather than precision. That is to say you see the filtering companies frequently advertising we get-we block 90 percent of the sexually explicit speech or sexually explicit sites on the web or 95 percent of the adult sites.

The claims are always of that sort. There are claims about recall, and the assumption there must be-

JUDGE BARTLE: What is the technical meaning of recall?

THE WITNESS: I'm sorry. Recall is the under-blocking problem.

JUDGE BECKER: Under-blocking.

THE WITNESS: What they're saying is we're getting as many of the adult sites as we can. They rarely say or they never say, as far as I know, we don't block any protected speech.

JUDGE BECKER: What you're suggesting is that the results depend upon the philosophy, as it were, of the blocking companies insofar as they designed their system to achieve a certain goal?

THE WITNESS: Yes, yes. That among other things. It could-also depends on the inherent difficulty of the problem. I mean if there is a lot of stuff to categorize, there are limited resources to categorize it with, and that's just going to let certain amounts of-

JUDGE BECKER: But if you're dealing with a reasonably, prudent, non-zealous blocking company, the best of all possible worlds, the best of all possible blocking company, your description of the intrinsic problems lead you to the conclusion that no matter what you do, there is always going to be under-blocking and over-blocking?

THE WITNESS: Right.

JUDGE BECKER: And which is the bigger of the two, problem of the two in terms of the built-in flaws in the system?

THE WITNESS: I don't think you can say.

JUDGE BECKER: You can't say that.

THE WITNESS: I mean I think it's very difficult to say. It's a question of how you measure it, what the consequences of it. When we build systems to do anything, we have to evaluate them relative to a particular intention or goal or purpose.

* * * * *

[TESTIMONY OF J. NUNBERG]
[March 26, 2002]
[pp. 51-52]

* * * * *

[CROSS-EXAMINATION]

BY MS. BHATTACHARYYA:

Q Dr. Nunberg, pornographic web sites are not that hard to find on the internet, are they?

A No.

JUDGE BARTLE: I'm sorry. I missed the question. What was-

BY MS. BHATTACHARYYA:

Q Pornographic web sites are not that hard to find on the internet, are they?

A No.

JUDGE FULLAM: How hard is that?

BY MS. BHATTACHARYYA:

Q In fact, you could find a large number of pornographic web sites fairly rapidly, couldn't you?

A Absolutely.

Q Based on what you know about the filtering companies, you assume, don't you, that the tools used by those companies do very well in finding large numbers of pornographic sites on the internet?

A Yes.

Q And in fact, if you were designing an automatic tool that was trying to extract from the web those sites that were likely to contain pornographic content, you would use all the techniques that are used by the filtering companies, wouldn't you?

A Yes.

Q And you would concede, wouldn't you, that the use of one of these filtering products is successful in restricting access to some large number of pornographic sites on the web?

A Yes.

* * * * *

[TESTIMONY OF J. NUNBERG]
[March 26, 2002]
[p. 62]

* * * * *

A The filtering companies rely as well, yes, that's right, the filtering companies rely as well on for example these web rings and directories and so on, but the web rings and directories proceed using the same techniques that the search engines do that is to say of following links and the limits that again are the limits that I was talking about yesterday.

Moreover, as I again mentioned in my expert report, there's extensive research on the mathematical properties of these link structures on the web which shows that they tend to converge rater then to diverge. That is to say that doubling the number of index pages and links you follow doesn't doubt the number o sites you find but it perhaps increases it by 40 percent and doubling it again increases it by 25 percent that there's a kind of law on diminishing returns on this, and again this is documented in a very extensive literature in the mathematic of web-of the structure of the web.

So again, it wold be perhaps misleading to call it an assumption since it's based on this large literature on the search engines and the structure of the web.

Q But you haven't reviewed any literature of that scope and coverage relating to the manner in which the filtering companies find things on the web have you?

A I'm not sure if I understand the question, but probably the answers no.

* * * * *

[TESTIMONY OF J. NUNBERG]
[March 26, 2002]
[p. 75]

* * * * *

BY MS. BHATTACHARYYA:

Q You discussed in your direct testimony the concepts of precision and recall, correct?

A Yes.

Q And the concept of precision which you defined yesterday as the percentage of documents that returned that are correctly classified, did I get that right?

A Yes.

Q That concept relates to the notion of over-blocking, that's fair?

A Yes.

Q And the concept of recall you defined as the percentage of total documents in the category that are returned, is that right?

A Yes.

Q And that concept recall relates to the notion of under- blocking, is that right?

A Yes.

Q You didn't attempt to determine as you testified the precision rate of any filtering product did you?

A No.

TESTIMONY OF J. NUNBERG]
[March 26, 2002]
[pp. 79-80]

* * * * *

BY MS. BHATTACHARYYA:

Q You didn't attempt to quantify as an absolute matter how much the filters may over-block did you?

A No.

Q And you didn't attempt to quantify as an absolute matter how much the filter might under-block did you?

A No, I didn't.

Q Dr. Nunberg, you're not qualified as an expert to judge how well a filter has to perform in order to be used in a library context, isn't that correct?

A Yes, that's correct.

Q You believe that the decision about how well filters need to perform in those contexts is a decision best left to the Courts, isn't that right?

A Yes, I do.

Q But you haven't provided this Court with data from which it can make a decision either relatively or absolutely about the weight-I'm sorry, I'll rephrase.

You haven't provided the Court with data with respect to the rates at which the filters either over-blocking or under-block, isn't that right?

A I haven't-to repeat what I said earlier, I haven't done any studies of precision or recall of these filters.

Q And you haven't provided the Court with any data with respect the absolute numbers of things that are either over-blocked or under-blocked, is that right?

A That's right.

(Pause in proceedings.)

Q I'm going to flip your chart again, can-

A Sure.

Q -I do that?

You testified on direct and you discuss in your report some 29 sites that you plan to demonstrate these five various flaws in the filtering products, is that right?

A Yes.

Q And in order to compile that list os 29 sites you used search engines to search on terms that you thought might give you less classified sites, is that right?

A That's how I found most but not all of the sites I used in this test.

* * * * *

[TESTIMONY OF J. NUNBERG]
[March 26, 2002]
[p. 93]

MR. SMITH: I have no redirect, Your Honor, but I want to move into evidence exhibits relating to this witness. Exhibits 70 through 75, 77, 79, 80 and 84.

JUDGE BARTLE: Any objection?

MS. BHATTACHARYYA: No objection, Your Honors.

JUDGE BECKER: They're received.

(Plaintiffs' Exhibits 70 through 75, 77, 79, 80 and 84 are admitted in evidence.)

JUDGE FULLAM: I have one question.

THE WITNESS: Excuse me, Your Honor.

JUDGE FULLAM: Am I correct in my understanding it's your testimony there's no way that visual depictions as such can be blocked?

THE WITNESS: Well, that's-if you want to simply say, don't show any visual depictions, show no logos, show no pictures of any sort, that's a relatively easy thing to do, you just say, a text only. And there are libraries, I understand, who have just taken this expedient of saying no one can have access to any visual material whatsoever by the web, and that would, of course, conform with the-

JUDGE FULLAM: No pictures.

JUDGE BECKER: But the minute you try to identify or distinguish among photographs, that's when the trouble begins?

THE WITNESS: Yes. It just can't be done.

* * * * *

[TESTIMONY OF E. ROOD]
[Mar. 26, 2002]
[pp. 142-151]

* * * * *

[DIRECT EXAMINATION]

BY MR. HANSEN:

Q Ms. Rood, how old are you?

A I'm 16 years old.

Q And where are you living at the moment?

A I currently live in Great Barrington, Massachusetts. I attend Simons Rock College.

Q What is Simons Rock College?

A Simons Rock is a liberal arts college. It bills itself at the college for younger scholars. It's a small school, we have 390 students and we're the only school in the nation that's designed for students who leave high school early. I came to Simons Rock after my sophomore year of high school.

Q And when you complete the program at Simons Rock, what kind of degree will you receive?

A If I stay at Simons Rock for four years, I'll have a B.A. degree.

Q What's your hometown?

A I'm from Portland, Oregon.

Q And with whom do you live when you're home?

A I live with my mother and my sister.

Q What is your mother's occupation?

A My mother is a librarian.

Q Are you a plaintiff in this case?

A Yes, I am.

Q And did you become a plaintiff in this case, and are you testifying here today with your mother's permission?

A Yes, we talked about it a lot before I became involved. She's really supportive.

Q Have you ever used public libraries?

A Yes, I have.

Q Can you describe why you would use a public library?

A For the same reason anyone would use a public library, for information.

Q Have you-specifically what library do you use when you're home with your mom and your sister?

A I use mostly Multnomah County Public Libraries. There's a main branch downtown, that's the central library. And it's the largest and the most extensive, so I mostly use that one when I'm trying to do research. But there are also a couple of smaller branches closer to my house that I'll use upon occasion.

Q And do you still use the Multnomah Library when you're home on summers or holidays?

A Yeah, definitely.

Q Have you used the internet?

A Ever?

Q Yes, ever?

A Yes, I have.

Q Do you have internet access at school?

A I do.

Q Does the school utilize any blocking software on its internet access?

A No.

Q Do you have internet access at your mother's home?

A I do.

Q And is any blocking software utilized on that computer?

A No.

Q Have you used internet access at the Multnomah Public Library?

A I have.

Q And would you describe how that came about?

A Well, I use the internet when I'm at the public library if I am-if I need to do research of one kind or another. And so I would go on the internet to look things up.

Q Was there a time when you went to the Multnomah Public Library to do a specific topic of research on the internet?

A Yeah. Most of-let's see, about two years ago, I'd say the summer before my ninth grade year in high school I was-it was '99, so I was 13, almost 14, that summer I went to the public library more than usual, more than I had before or have since because I was questioning my sexual orientation and I was doing research online to try and find out information about being gay and coming out, and it was, well, you could say it was a pretty sensitive topic and so I wanted to do it at the library rather than at home because I didn't want my family to know what I was doing.

Q Does the Multnomah County Public Library offer the option of blocking software for patrons?

A Yes, it does.

Q Did you utilize that option?

A No, most certainly not.

Q Why not?

A I had experience with blocking software before because my middle school did have blocking software installed on its computers in the library and the classrooms, and I knew that blocking software would probably inhibit my research at the library.

Q Can you give an example of that?

A Sure. One time when I was in eighth grade a friend of mine was trying to do a report on humpback whales and he was trying to access information on the internet and he was searching with the terms, humpback whales, and the filter wouldn't let him get anything with the words, hump, in it, and so he couldn't find very much information at all.

JUDGE FULLAM: Had to switch over to gray whales, I guess.

THE WITNESS: I guess so.

 

BY MR. HANSEN:

Q What kind of sites did you visit while you were doing research at the Multnomah Library about your sexual orientation?

A Mostly two different kinds of sites, I guess you can say. The one kind would be sort of very educational site, mostly made by grownups, such as like, you know, I think I might be gay, what do I do, and written by, like, guidance counsellors or people with authority, sort of basically educational informational resources for teenagers. And then also some of the sites I spent more time on were more personal web sites of individuals and telling their coming out stories and their personal experiences, and basically to counsel people in my same situation that I could read and become more familiar with and more comfortable with the idea of being gay.

Q Did you engage in any chat resources while you were using the library?

A Yeah. Yeah, I used a couple of sites like, planitout.com, I think, and gay.com, two of the major ones I used. They have chat rooms where you can talk to other people in real time on, you know, a variety of whatever you want to talk about. So I basically found people that I can talk to and explain my situation where I was in the closet, I wasn't out to anyone, I didn't really have anyone else I could talk to in real life, and it wasn't something that I was comfortable talking to anyone in my family about. And so I found people online that I could talk to and who could offer me advice and support.

Q Did you find this research helpful?

A Oh, incredibly helpful.

Q Could you explain why?

A That was a time in my life when I was really isolated and alone because I didn't really feel comfortable talking to anyone that I knew about this subject. It's a very sensitive topic, it was a big secret. And by being able to access information freely on the internet at the public library where no one knew who I was or what I was looking up. I was able to become so much more comfortable with myself and find out information that was really helpful to me in my basically personal growth as a human being.

Q If one of the sites you went to had been blocked by blocking software, would you have gone to a librarian to request that it be unblocked?

A Oh, definitely not.

Q Why not?

A It wasn't something I could share with someone. I couldn't go up to a grown-up, an adult working at the library and ask them to do that. You know, my mom is a librarian, she's in the library community in Portland, but if they saw my mom and say, hey, I saw Emma at the library looking up gay web sites, what does that mean? I mean it wasn't something that I could do.

Q Do you have any reason to believe that any of the kinds of sites you went to would have been blocked if you used blocking software?

A Yeah, probably.

Q And what makes you think that?

A Well, one time when I was at school, trying to use the internet at school that had the blocking software on it, I tried to look up-like I put lesbian teenager into a web site and I couldn't get any information. And then another time when I was at the public library, the Multnomah Library, I went to a computer that had already-when you try to use the internet at Multnomah, a screen pops up saying, would you like filtered or unfiltered, and usually I would choose unfiltered access. But at that particular instance I went to a computer that had already-somebody had already chosen and I tried to get some stuff on the search engine and things were coming up blocked and I realized they must have-I must have been using one that had been chosen to use filtered access, and so I just switched to a different computer because I couldn't access-I think I was trying to get to lesbian.org.

JUDGE FULLAM: Could I ask just for the record, how old are you?

THE WITNESS: I'm 16 and a half, I'll be 17 in September.

JUDGE FULLAM: I understood you to say 15, and I checked and elsewhere you're listed at 16. I wanted to make sure we were correct. Okay.

BY MR. HANSEN:

Q Are you generally familiar with the Children's Internet Protection Act?

A I'd say generally familiar, yeah.

Q What would be the effect on you if the Children's Internet Protection Act went into effect and the Multnomah County Library chose to use blocking software?

A I'd say it would pretty much have an entirely detrimental effect.

MS. BHATTACHARYYA: Objection, Your Honor. Speculative.

JUDGE BECKER: Overruled.

MR. HANSEN: You can answer it.

THE WITNESS: Okay. Basically I think it would be detrimental for one, because it would be pretty frustrating. I would be pretty frustrated that I couldn't have access to information that I needed and wanted. And because, you know, I'm sure I'm a minor, but I'm a freshman in college. I consider myself perfectly responsible enough to decide what information is valuable or what I need to see or don't need to see. And if by some chance I came across something that wasn't what I was looking for was objectionable, I could just close that window and go somewhere else. So I think it would just not be helpful in any way.

Q Do you think it would be detrimental to other teenagers who are questioning their sexuality?

A Yeah.

Q Why?

A I really think it would. One of the main problems that sexual minority, sexual minority as being gay, lesbian, bisexual, transgendered, questioning youth-

MS. BORNSTEIN: Objection.

JUDGE BECKER: Ms. Bornstein, I take it?

MS. BORNSTEIN: Yes. Object again for speculation. It's even more speculative.

JUDGE BECKER: Overruled.

THE WITNESS: Where was I? Oh, yes. One of the main problems for sexual minority is the sense of isolation that we feel because see how this is-it's really not socially acceptable to be gay in our society yet, and so it can be very hard to come out to people and find the support systems that we need to find friends and family who are accepting or basically anyone who is accepting.

So by being able to use the internet to find a community and resources and friendly people for youth who are in locations such as rural areas, or if they don't have the internet access at home, being able to use the library, the public library would be very helpful.

MR. HANSEN: Thank you, Ms. Rood. I have no more questions.

THE WITNESS: Thank you.

JUDGE BECKER: Cross-examine, Ms. Bornstein.

* * * * *