What does it mean for a website to be “accessible?” The standard answer for several years, now embodied in the implementing regulations for Section 508 of the Rehabilitation Act, is that accessible means “conforming” to WCAG 2.0 success level AA. A recent decision from the Eastern District of New York, Andrews v. Blick Art Materials, LLC, 2017 WL 6542466 (E.D.N.Y. Dec. 21, 2017), has a good discussion of why this should be the case, at least with respect to the visually impaired.* However, the focus in Andrews v. Blick Art was whether the terms of a settlement were appropriate, not whether WCAG 2.0 success level AA is what the courts should order.
In the one case that has gone to trial, Gil v. Winn-Dixie Stores, Inc., the court ordered WCAG 2.0 compliance after finding that the defendant’s website was not accessible. The order is interesting because it orders in general terms that the website be accessible and then specifically that it comply with WCAG 2.0, but never says that WCAG 2.0 compliance is equivalent to accessibility and never says what success level is required. This is one of many issues raised by the defendant on appeal, for it argues that because WCAG 2.0 is based on general principles rather than specific mandates the defendant can never be sure whether it has complied.**
This brings us to Eason v. New York State Bd. of Elections, 2017 WL 6514837 (S.D.N.Y. Dec. 20, 2017). Eason is a Title II / Rehabilitation Act case that bypasses the “does the ADA apply” question because Title II entities must make their programs accessible, meaning there is no real doubt that their websites should be accessible. That leaves more complicated questions about just what accessibility means.
The Court’s first observation is that the plaintiff and defendant have very different theories about how to approach the case in the first place. The plaintiff seems to have wanted a broad finding that the website was not accessible with an order for the implementation of policies and procedures to make and keep it accessible. The defendant, on the other hand, had fixed the specific problems listed in the complaint and wanted the case dismissed on the theory that the case concerned only those problems and was therefore moot. The Court rejected both approaches.
It rejected the defense approach because, it found, the complaint was sufficiently open to include accessibility problems other than those listed in it. It also analogized websites to physical facilities, noting that inaccessibility at one place might keep a disabled person from even getting to another place to discover what was inaccessible there.
The Court did not reject the plaintiff’s theory of relief, but did reject the report of the plaintiff’s expert, which it found had a number of problems, the most important of which was a failure to focus on “meaningful access” as opposed to “equal access.” Equal access is, of course, impossible. A blind person cannot under any circumstances equally see an image. What is required, according to the Court, is meaningful or functional access to the services offered on the website.
This focus on meaningful access required a trial according. The opinion says the Court wants to see how screenreader software works on the website itself to determine whether there is meaningful access. That, it found, depends on things like how long it takes to access a service for a vision impaired person as compared to a person with ordinary vision. It isn’t hard to imagine any number of difficult questions raised by the “meaningful access” standard. If an image is important, how much description is required for a person using a screen reader to have meaningful access. Is “boy and dog” enough, or does it need to specify “boy and dog playing with stick” or “boy and dog playing with stick in the park?” Those are the kinds of questions the Court seems to believe could only be answered by looking at the entire process by which an individual with a screen reader uses a website. This is, by the way, something that good website accessibility consultants agree is true. No algorithmic or software analysis is sufficient. If you want to know whether a website is accessible, you have to test it with humans.
Notably absent is any mention of WCAG 2.0. The Court seems to have adopted an approach that asks the plaintiff to explain where the problems are, the Court looks at the problems to see if it agrees, and then the Court orders that problem fixed as opposed to accepting that problems exist and then ordering WCAG 2.0 compliance as the solution. From a legal perspective this makes perfect sense. The Court can only decide the dispute it has before it, and that dispute involves a particular blind individual who had or might have specific problems with a specific website. Requiring that the website meet some general level of accessibility for the good of all disabled persons is generally beyond the court’s authority in individual private lawsuits.
Although Andrews v. Blick touches on the question of whether WCAG 2.0 is an appropriate accessibility standard, none of these cases delves very far in to the difficult questions raised by the Winn-Dixie appeal. The first of these is how to bridge the gap between general principles and the hard reality of software code and hardware choices. One of the disputes in another recent case, Del-Orden v. Bonobos, Inc., 2017 WL 6547902 (S.D.N.Y. Dec. 20, 2017) concerned the ability of low-vision users to see an on-screen button while using the screen zoom function. On screens larger than 23.5 inches it was visible. On smaller screens it was not. Is WCAG 2.0 compliance required for every possible screen size or resolution, or can a business ignore outdated displays and standards? Similar problems exist with respect to the differing capabilities of screen reader software and mobile devices. WCAG 2.0 is supposed to be device independent, but that comes at the cost of not being precise, and a lack of precision is an invitation to litigation.
The ADA standards for access to physical places are “objective,” “precise” and “thorough.”† A business that satisfies them can be certain that its physical facility complies with the requirements of the ADA. WCAG 2.0, for all its commentary and technical resources, ultimately concerns the subjective experience of a particular user, with particular software and hardware, interacting with a website that, unlike most physical businesses, constantly changes. WCAG 2.0 Success Level AA is the best bet for any business website, but cases like Winn-Dixie and Eason are raising questions about what “accessible” means that courts may find difficult to answer when it comes to websites.‡
* This decision is worth reading by any person who wants to understand how websites can be inaccessible, for it includes a detailed description of the problems in the defendants’ website from a functional perspective.
** The appeal is Case No. 17-13467 in the 11th Federal Circuit Court of Appeals. On January 13 the Court of Appeals found the case required oral argument, but it has not yet set a date.
† Chapman v. Pier 1 Imports (U.S.) Inc., 631 F.3d 939, 945 (9th Cir. 2011).
‡ In this regard businesses should note that WCAG 2.0 will be updated this spring to WCAG 2.1, which includes a number of new Success Level AA criteria, including one that would probably fix the zoom problem identified in Del-Orden v. Bonobos. Like any change in accessibility criteria this one begs the question, “if the change is needed, does that mean the earlier standard did not insure accessibility?” Every accessibility technology can be improved and no accessibility standard guarantees complete access for all. When technology is rapidly evolving every effort to find the right compromise is bound to be temporary. Rapid change is not something that Congress, the Courts, or regulators are very good at.