These closely related questions address the accessibility of information provided by an ITM to people with different kinds of disabilities affecting vision, including low vision and the condition that is commonly referred to as "color blindness." These are based on the Trace Guidelines' recommendation that, "If the user has difficulty seeing the device, let them change the way it looks." The Trace Guidelines explain:
Text is malleable depending upon the constraints of the visual interface. For example, fonts can be enlarged, changed between serif and sans serif, made white on black or any other color combination. A visual interface may be constrained in the maximum size of the text, the colors that are available, and the clarity (resolution) possible.
Question 3, which relates to use of color, affects usability by users who cannot distinguish colors as well as those who need high contrast or low contrast color combinations. Question 4 relates to the size of the text, which would only affect users with low vision and not those whose only disability is a difficulty distinguishing colors.
In about one-quarter (19 of 82) of the surveys, components indicated that the ITMs potentially exclude some users because they display insufficiently contrasting foreground and background colors and because users are not able to change the color selections. See Table 3. Additionally, in 39% (39% or 32 of 82) of the surveys, components indicated that the ITMs do not provide text large enough for users with low vision to discern and do not allow users to enlarge the display. See Table 4. Almost all of the ITM surveys showed a strong correlation between components' responses to Questions 3 and 4. See Table 5.
5. Can users select speech input?
Providing speech input for a device provides accessibility in a number of different ways; if speech input is not provided, it may affect many people with disabilities. Those who are blind or who have low vision may be excluded if an ITM requires a form of input that relies on vision; relying on speech input is one way to provide usability for such users. Another group of users who may be excluded when speech input is not available are those who lack fine motor skills, have limited reach or strength, or who lack sufficient neuromuscular coordination (e.g., because of tremors) to operate physical input controls. These users may also benefit greatly from allowing speech input. Finally, speech input may assist users with certain learning disabilities and cognitive impairments.
In 80% (66 of 82) of the surveys, components indicated that the ITMs do not permit speech input. Fortunately, the unavailability of speech input generally does not independently exclude any group of users if other accessibility features are provided. Relatively low availability of speech input on federal ITMs may reflect the current state of technology. Because voice recognition technology is swiftly becoming more accurate and affordable, its use in federal ITMs is likely to increase. See Table 6.
6. If speech input is used, is an alternative method available for inputting information, such as typing on a keyboard or scanning printed material, so that someone who cannot speak can use the technology?
While speech input may assist some, others need alternatives such as the ability to use a keyboard or keypad to input information. Question 6 targets issues affecting those users who have difficulty speaking, such as those with speech disabilities and some users who are deaf or hard of hearing.
Because, as noted with respect to Question 5, relatively few federal ITMs permit speech input, users who have difficulty speaking only rarely face the barriers addressed in Question 6. In 16% (13 of 82) of the surveys, components indicated that the ITMs rely exclusively on speech input. See Table 7.
7. For all sound cues and audible information, such as "beeps," are there simultaneous corresponding visual cues and information?
8. Is there a headphone jack to enable the user to use an assistive listening system to access audible information?
These 2 questions focus on an ITM's usability for persons who are deaf or hard of hearing. Where a user cannot hear, the ITM should provide the user with simultaneous visual cues and information for all sound cues and audible information. As the Trace Guidelines state:
If the user cannot hear the sounds from the device, show the sounds visually.
Any sounds that a device makes can be shown visually, for example by making a display or indicator light flash when a sound is made. Spoken text and sounds can be shown in "caption" form, enabling someone who cannot hear at all to have access to the same information as people who can easily hear.
Users who are hard of hearing may also benefit from visual information and cues, but may not be completely excluded from a lack of visual information and cues (such as where he or she is able to adequately amplify the volume). A "no" answer to Question 7 indicates that the ITM likely excludes users who are deaf and may adversely affect those who are hard of hearing.
In 18% (15 of 82) of the surveys, components indicated that the ITMs do not provide visual cues and information corresponding to sound cues and audible information. See Table 8.
Users who are hard of hearing can often benefit from audible cues and information if they are provided with a tailored means of listening. One important way to accomplish this task is by providing a standard headphone jack, which permits users to use standard headphones or make use of the T-switch(7) technology that is built in to many hearing aids.(8) In addition to amplification, providing a standard headphone jack can assist people who are hard of hearing by giving them an alternative that minimizes distracting background noises.
Installation of a standard headphone jack also provides a private means of listening for people who use audio output, such as those who are blind or who have low vision. For instance, a blind person who uses an ATM may wish to maintain his or her privacy when the ATM audibly "displays" the amount of cash he or she is withdrawing, or account balance information.
A "no" answer to Question 8 indicates that the ITM surveyed contains a barrier to users who are hard of hearing and to those who use audible output and would have a lesser degree of privacy if headphone jacks were not provided. In 71% (58 of 82) of the surveys, components indicated that the ITMs are not equipped with standard headphone jacks. See Table 9.
9. Can users simultaneously change the visual display settings and the sound settings?
This question relates to users who have low vision and who are hard of hearing. A "no" answer to this question likely indicates that such users may not be able to interact with the ITM. In 68% (56 of 82) of the surveys, components indicated that the ITMs do not permit users to simultaneously change the ITMs' visual display settings and sound settings. See Table 10.
10. Can the user read displayed output with a tactile display such as Braille?
Braille tactile displays provide some blind people with the ability to use an ITM.(9) A "no" answer to this question may indicate a barrier to access by some blind users, but can be offset (for blind users who are not also deaf or hard of hearing) by the provision of adequate audible information. In over 85% (70 of 82) of the surveys, components indicated that the ITMs do not provide output with tactile display, such as Braille. See Table 11.
11. Does the technology allow the user to use scanning input?
Scanning input allows different user options to be highlighted in sequence. A user can choose from different options by selecting that option when its button or menu item is highlighted. As explained by the Trace Guidelines, providing scanning input may significantly improve accessibility for persons with physical disabilities:
If the user can see, but can only use one or two switches for input, let them step around the buttons using scanning
If someone can only use one or two switches (for example if they are paralyzed from the neck down), it is possible to control the interface by having each item highlighted (or said aloud) one by one. When the one that the user wants is highlighted, they can select it using a single switch. With a double switch they can use one switch to advance the highlight, and the other to select. The latter has more flexibility and control, but not everyone can use two switches which is why single switch is available. Note: It is possible to scan using auditory feedback, but it would be more likely that a user would use speech output and a list to interact with the device.
A "no" answer to Question 11 suggests that the device may present barriers to access by a person who has very limited mobility or dexterity.(10) In 82% (67 of 82) of the surveys, components indicated that the ITMs do not provide an option for scanning input. See Table 12.
12. Is the technology manufactured such that it allows a person using a wheelchair to approach the technology, including all controls, dispensers, receptacles, and other operable equipment, with either a forward or parallel approach?
13. Is the technology manufactured so that, if the equipment is properly placed, the highest operable part of controls, dispensers, receptacles, and other operable parts fall within at least one of the following reach ranges?
If a forward approach is required, the maximum high forward reach is 48 inches.
If a side approach is allowed, and the reach is not over an obstruction, the maximum high side reach is 54 inches; if it is over an obstruction which is no more than 24 inches wide and 34 inches high, the maximum high side reach is 46 inches.
14. If electrical and communication system receptacles are provided, are they mounted no less than 15 inches above the floor?
15. Are all controls and operating mechanisms operable with one hand and operable without tight grasping, pinching, or twisting of the wrist?
16. Is the force required to operate or activate the controls no greater than 5 lbf?
All of these questions are based on requirements for "Controls and Operating Mechanisms," as stated in the Uniform Federal Accessibility Standards (UFAS), which set design and construction standards for federal and federally-funded facilities under the Architectural Barriers Act, 42 U.S.C. §§ 4151 et seq. Each of these requirements affects the usability of ITMs by people with some other types of mobility impairments, such as those who use wheelchairs.
Federal agencies' ITMs generally meet the physical accessibility requirements of these Uniform Federal Accessibility Standards (UFAS). In 34% (28 of 82) of the surveys, components gave a "no" response to 1 or more of these 5 questions. See Table 13.
17. Are instructions and all information for use accessible to and independently usable by persons with vision impairments?
This question is based on section 4.34.5 of the Americans with Disabilities Act Standards for Accessible Design, 28 C.F.R. pt. 36, Appendix A (ADA Standards). These design standards apply to public accommodations and commercial facilities under title III of the Americans with Disabilities Act, 42 U.S.C. § 12182, et seq., and do not directly apply to federal facilities. However, section 4.34.5 is part of the ADA Standards which relates directly to the design and construction of automated teller machines (ATMs), a common type of ITM. Furthermore, the Access Board has proposed amending the standards that apply to federal facilities -- the Architectural Barriers Act Guidelines -- to include this requirement.
A "no" answer to this question would indicate that the ITM presents barriers to access to people with vision impairments. In 73% (60 of 82) of the surveys, components indicated that the ITMs did not provide instructions in a format that is accessible to and independently usable by persons with vision impairments. See Table 14.
18. Is the technology manufactured in such a way that it can be made detectable to persons with visual impairments who use canes to detect objects in their path?
Note: Objects projecting from walls with their leading edges between 27 in. and 80 in. above the finished floor should protrude no more than 4 in. into walks, halls, corridors, passageways, or aisles. Objects mounted with their leading edges at or below 27 in. above the finished floor may protrude any amount. Free-standing objects mounted on posts or pylons may overhang 12 in. maximum from 27 in. to 80 in. above the ground or finished floor.
People who are blind and many people with significant low vision use canes to detect objects in their path of travel. Question 18 is based on section 4.4 of UFAS, which is designed to ensure that all objects on which a person could injure himself or herself are detectable by the proper use of a cane.
A "no" answer to this question may indicate that the ITM poses a barrier to access by a person who is blind or a person who has low vision. If the ITM is not "cane-detectable," a blind person could walk into it and injure himself or herself.
In 24% (20 of 82) of the surveys, components indicated that the ITMs are not manufactured so that they can be made detectable to people who use canes to detect objects in their path.(11) See Table 15.
B. Summary of Impact on Disability Categories
The following chart (D) summarizes the survey questions and the disability categories that are affected by responses to those questions:
|People Who Are Hard of Hearing||Q's 1, 7, and 8|
|People Who Are Deaf||Q's 6 and 7|
|People Who Have Low Vision and Who Are Hard of Hearing||Q's 1-5, 7-9, 17, and 18|
|People Who Are Blind||Q's 2, 5, 10, 17, and 18|
|People With Color Blindness||Q3|
|People With Low Vision||Q's 2-5, 17, and 18|
|People With Speech Disabilities||Q6|
|People Who Have Tremors or Limited Strength or Dexterity||Q's 5, 11, 15, and 16|
|People Who Use Wheelchairs||Q's 12-15|
|People Who Have Cognitive Impairments or Learning Disabilities||Q's 2 and 5|
C. Objective Survey of Accessibility by Disability Category
1. People Who are Hard of Hearing
Questions 1, 7, and 8 address issues that affect users who are hard of hearing. In 12% (10 of 82) of the surveys, components indicated that the ITMs contain the barriers addressed in all 3 questions. In 82% (67 of 82) of the surveys, components indicated that the ITMs contain at least one of the barriers addressed in these questions. These results suggest that people who are hard of hearing encounter barriers when using a large percentage of federal ITMs. See Table 16.
However, a more careful analysis of this data indicates how different users who are hard of hearing are affected. Certain users may be able to use ITMs through the sound amplification (Question 1) or by the use of visual cues and information (Question 7). In 12% (10 of 82) of the surveys, components indicated that the ITMs contain barriers to use by people who require either sound amplification or the use of visual cues and information. See Table 17.
For other users, sound amplification alone may be insufficient; these users may require either a headphone jack for an assistive listening system (Question 8) or visual cues and information for all audible information (Question 7). In 16% (13 of 82) of the surveys, components indicated that the ITMs entirely exclude users who require either the use of visual cues and information or a headphone jack in which to plug an assistive listening device. See Table 18.
2. People Who Are Deaf
Questions 6 and 7 address issues affecting people who are deaf. A "no" response to either Question 6 or Question 7 may exclude deaf users. In 28% (23 of 82) of the surveys, components indicated that the ITMs exclude users who are deaf using these criteria. See Table 19.
3. People Who Have Low Vision and Who Are Hard of Hearing
There are a number of questions that address issues affecting users who have low vision and who are hard of hearing. Almost all surveyed ITMs, 96% (79 of 82), do not satisfy at least one of these questions. A more careful analysis of these questions reflects how this diverse group of users is affected. See Table 20.
Some people who have low vision and who are hard of hearing may prefer all information to be conveyed visually to maximize their success of using ITMs. This group of users will require that the features of the ITM meet the needs of a person with low vision (Questions 3, 4, 17, and 18) and that all information conveyed through sounds is also conveyed visually (Question 7). In 79% (65 of 82) of the surveys, components indicated that the ITMs contain barriers for this community. See Table 21.
Other people who are hard of hearing and who have low vision may more readily use information that is conveyed through audible means, due to the severity of their vision impairment, the impracticality of using assistive equipment for visual media in a public setting, or in conjunction with a particular ITM's construction or location. For them, all information that is conveyed visually should also be available audibly. This group of users will require a favorable response to a different subset of questions posed in Table 20 than those users who would favor a visual media. They will require that the features of ITMs meet the needs of a person with low vision (Questions 5, 17, and 18) and that all information conveyed visually is also conveyed through sound (Question 2). Some may be able to rely principally on adjustable sound settings (Question 1). See Table 22. In 94% (77 of 82) of the surveys, components indicated that the ITMs are inaccessible in this regard. By contrast, other users may not be able to rely on adjustable sound settings and may require that ITMs are equipped with headphone jacks to allow people to use assistive listening devices (Question 8). See Table 23.
A third group of people includes those whose disabilities affecting vision and hearing are both sufficiently severe that changes in sound and visual settings, even when done simultaneously, cannot provide opportunities for independent access. For them, ITMs should allow people to use keyboard input (Question 6) and tactile displays (Question 10) such as Braille. In addition, instructions and user information should be provided in a format that is accessible to persons with vision impairments (Question 17) and ITMs should be manufactured to be detectable by persons who use canes to detect barriers in their path (Question 18). As a negative response to any of these questions may indicate that the ITMs have significant or insurmountable barriers to people with significant disabilities affecting vision and hearing. See Table 24. In 90% (74 of 82) of the surveys, components indicated that the ITMs are inaccessible in this regard.
4. People Who Are Blind
People who are blind require:
In 80% (65 of 82) of the surveys, components indicated that the ITMs are inaccessible in these respects. See Table 25.
This large number is compounded by the fact that positive responses to these three questions only provide the most rudimentary form of "access." In addition, permitting speech input (Question 5) and allowing users to use a tactile display (Question 10) would greatly facilitates usability by some blind users. In 95% (78 of 82) of the surveys, components indicated that the ITMs do not incorporate these additional features that would improve access to people who are blind. See Table 26.
5. People Who Have Difficulty Discerning Color
To be usable by persons who cannot distinguish certain colors, ITMs should provide sufficient contrast between foreground and background colors or allow users to select their preferred foreground and background colors (Question 3). In 23% (19 of 82) of the surveys, components indicated that the ITMs pose potential barriers to use for this community. See Table 3.
6. People with Low Vision
A number of questions affected the usability of ITMs by users with low vision. In 95% (78 of 82) of the surveys, components indicated that the ITMs include one or more barriers to access by users with low vision. See Table 27. A closer analysis of the questions, however, reveals that the actual number of ITM models excluding users with low vision may not be quite as high as this statistic would indicate, because different users with low vision may be able to use an ITM in different ways.
Depending on the severity of a person's vision impairment, among other factors, someone may need or prefer audible information instead of visual information (Question 2). For such users, changing the way that information is displayed visually (Questions 3 and 4) would not be all that helpful. To be usable by this community, ITMs would still have to provide instructions and information that is independently usable and accessible to people with visual impairments (Question 17) and would have to be constructed to be detectable by persons who use canes to detect barriers in their path (Question 18). In 79% (65 of 82) of the surveys, components indicated that the ITMs contain such barriers. See Table 28. These users may also benefit from speech input (Question 5), as well as the features addressed in Questions 2, 3, 4, 17, and 18. In 93% (76 of 82) of the surveys, components noted that the ITMs pose one or more of these potential barriers. See Table 29.
Another group of users with low vision may need or prefer enlarged visual displays or visual displays with sufficient contrast (Questions 3 and 4), instead of audible information (Question 2). Naturally, to be usable by this community, ITMs must contain information and instructions that are independently usable and accessible to users with low vision (Question 17) and should be detectable to persons who use canes (Question 18). In 78% (64 of 82) of the surveys, components indicated that the ITMs contain barriers to use by this community. See Table 30.
7. People with Disabilities Affecting Speech
People with speech disabilities are affected by the issues addressed in Question 6. In 16% (13 of 82) of the surveys, components indicated that the ITMs pose barriers to users with speech disabilities because the ITMs require users to be able to use speech input. See Table 7.
8. People with Tremors or Limited Strength or Dexterity
Users with tremors, limited strength, or limited manual dexterity will be affected by the issues raised in Questions 5, 11, 15, and 16. In 93% of the surveys, components indicated that ITMs pose one or more barriers to access by this community. See Table 31.
A more careful analysis of this data, however, may suggest that a smaller group of these users would actually encounter barriers to using federal ITMs. Assuming that a person does not have other disabilities which affect his or her ability to receive information (e.g., an additional hearing or visual impairment), the difficulties encountered by this group of users relates to inputting information to ITMs, rather than receiving information output from ITMs. Questions 5, 11, 15, and 16 all relate to controls or means of inputting information to ITMs; however, Questions 5 and 11 pertain to two alternative means of providing information input. Question 5 relates to whether ITMs provide a means for speech input by users. Question 11 asks whether ITMs permit scanning input. Each of these alternatives may independently provide access for users within this category. Questions 15 and 16, however, relate to the accessibility of the controls of ITMs. As these controls may include card reader devices or mechanisms for activating an ITM, a positive response to these questions is very important.
Table 32 summarizes the accessibility of ITMs for users who are able to use speech input. In 85% (70 of 82) of the surveys, components indicated that the ITMs pose one or more barriers to this community.
Table 33 summarizes the accessibility of surveyed federal ITMs for users who are capable of using scanning input. In 83% (68 of 82) of the surveys, components indicated that the ITMs pose one or more barriers to this group of users.
9. People Who Use Wheelchairs
The issues affecting people who use wheelchairs are slightly different than those of users with other physical disabilities. See Questions 12, 13, and 14. Compared to other disability categories, this group of users is least affected by agencies' ITMs. In 22% of the surveys, components indicated that the ITMs pose potential barriers to people who use wheelchairs. See Table 34.
10. People With Learning Disabilities and Cognitive Impairments
Finally, users with learning disabilities and cognitive impairments may encounter barriers when using federal ITMs. A "no" response to Questions 2 or 5 may indicate that the ITMs contain barriers to some people with learning disabilities or cognitive impairments. In 89% (73 of 82) of the surveys, components indicated that the ITMs potentially exclude users with learning disabilities or cognitive impairments in these regards. See Table 35.
II. Subjective Evaluations of ITMs Accessibility
Components were asked to subjectively evaluate the accessibility of their ITMs to a wide-range of users with differing disabilities. Specifically, the Department directed components with ITMs to do the following:
After you have evaluated this ITM using the Checklist, have users with a wide variety of disabilities test it for accessibility. Describe the accessibility successes and problems they encountered during these exercises, including any suggestions for improvement.
Question 19, ITM Accessibility Checklist.
Subjective evaluations by components of federal ITMs were rather limited:
In 10 of the surveys, components reported that they believed the ITMs are generally accessible or were designed with accessibility in mind. In another 10 of the surveys, components reported that the ITMs are regularly used by users with disabilities without assistance. In 3 surveys, components reported that the ITMs pose some barriers, but the components did not elaborate on the nature of the problems.
Some agencies and their components made specific observations regarding issues that affect users with disabilities. Some entities indicated that they were making their programs (normally delivered through the ITMs) available through alternate means:(12)
Some agencies expressed interest in improving the accessibility of their ITMs or indicated that they were already addressing this issue.
People with mobility impairments. For people with mobility impairments, including those who use wheelchairs, agencies and their components made the following findings:
People who are deaf or hard of hearing. With regard to issues concerning people who are deaf or hard of hearing, agencies and components made the following findings:
People with vision impairments, including those who are blind. The group most affected by a lack of accessibility of ITMs are users who have visual impairments, including those who are blind. Agencies and their components made the following specific observations with respect to issues affecting this group:
None of the federal agencies has outlined an adequate strategy for eliminating barriers to accessibility for their ITMs, even though barriers were identified. The Department recommends agencies take the following steps:
1. Non-Agency-Owned ITMs. Each agency that has facilities or property containing ITMs that are owned or controlled by other entities (including private entities, other federal agencies, or others) should notify them of any barriers to access in their ITMs and recommend that such entities address accessibility issues on a specific time schedule.
2. Location of ITMs. Each agency that has ITMs should ensure that its ITMs are located on accessible routes and are otherwise accessible to people with disabilities such as those who use wheelchairs.
3. Inaccessible ITMs. If an agency's existing ITM is inaccessible or contains inaccessible features, the agency should ensure that whatever information or services the agency provides on the ITM are also available through an accessible and comparably convenient and useful alternate means of access (e.g., automated telephone service or through the Internet). The agency should provide appropriate signage with full instructions regarding use of the accessible alternative method of obtaining information or services.
4. Upgrading Existing ITMs. While section 508 does not generally require retrofitting existing EIT, each agency that replaces or updates an ITM's software or hardware should look for and take advantage of easy opportunities to improve the ITM's accessibility.
5. Instructions. Many times, an ITM contains accessible features, such as a volume control mechanism, but instructions on how to use these features are missing or inadequate. Each agency that has an ITM should survey the ITM and, if appropriate, contact the ITM vendor for a full list of accessible features. The agency should provide clear instructions in accessible formats.
1. This document is available on the Department of Justice's section 508 Web site (www.usdoj.gov/crt/508). People with disabilities may request copies in Braille, large print, or on computer disk by calling 1-800-514-0301 (voice) or 1-800-514-0383 (TTY).
2. Throughout this section, the numbers reflecting components' surveys -- and the percentages calculated by the Department of Justice for the purposes of analyzing the accessibility of federal ITMs to persons with disabilities -- are raw data that reflect only the number of ITM surveys provided to the Department. They do not reflect the number of ITMs captured by each individual survey (for instance, where multiple ITMs of the same model were evaluated on a single survey), nor are they weighted by frequency of use. The Department did not receive reliable data on which to perform the necessary calculations.
3. Use of the Trace Center's materials does not constitute an endorsement of the Trace Center or its work by the Department of Justice. Likewise, the Department of Justice's ITMs Accessibility Checklist has not been adopted, endorsed by, or in any way approved by the Trace Center, NIDRR, or the Department of Education.
4. The Access Board has proposed amendments to the ABA Guidelines. 64 Fed. Reg. 220 at 62248 (Nov. 16, 1999).
5. Although the ADA Standards do not apply to federal agencies, they provide useful guidance because they includes specific standards for automated teller machines, a common form of ITMs.
6. Accompanying this analysis are 3 sets of appendices, which include tables and descriptions of the data provided by the agencies. These ITM Appendices can be summarized as follows:
7. The T-switch, which is also known as the telecoil, the induction coil, and the induction pick-up coil, is a feature found on some hearing aids that can link the hearing aids to other sources of electromagnetic energy, such as audioloop systems and other types of assistive listening systems.
8. The Trace Guidelines recommend,
If the user has difficulty hearing the device, let them change the way it sounds
Sound contains properties that can be altered, such as volume (loudness) and pitch. Modifying these can help users who are unable to hear a device operating normally. In addition, it is possible to directly connect hearing aids to sound sources, providing a better listening system (e.g., a headphone jack connection or telephone hearing aid T-coil connection).
9. Approximately 10% of people who are blind use Braille.
10. Any conclusions drawn from components' responses to Question 11 should be tempered by the possible misunderstanding by evaluators of the use of the term "scanning input" in the question. Some evaluators may have mistakenly believed that inquiry went to whether the ITMs contained computer flatbed scanners or Optical Character Recognition (OCR) technology.
11. Agencies can address this issue simply by placing cane-detectable items (such as posts or permanent planters) on either side of the ITMs, making sure that in doing so they are not creating barriers for other people with disabilities (for instance, the cane-detectable items should not interfere with the ability of a person who uses a wheelchair to approach or use the ITM).
12. Under sections 501 and 504 of the Rehabilitation Act, agencies must ensure that their programs are accessible to persons with disabilities. When their ITMs have barriers, agencies should ensure that whatever programs or information the agency provides on the ITM is also available through an accessible and comparably convenient useful alternate means (e.g., through an automated telephone service or through the Internet). This will assist agencies in meeting their general nondiscrimination obligations under sections 501 and 504 of the Rehabilitation Act, 29 U.S.C. § 791, 794.