This week in Accessibility: Gomez v. GNC
Multi-colored pills
Gomez v. GNC should serve as a warning to any company facing an accessibility lawsuit regarding the importance of their accessibility experts being accessibility experts. This case started as many digital accessibility lawsuits do:
1. A person with Disability can’t use the website to complete business transactions (in this case, the PwD was blind and trying to use JAWS to find a store, buy something online, etc.)
2. Plaintiff files class action lawsuit asserting various ADA violations committed against a group of plaintiffs in the same situation they are in.
When they go to trial, class action lawsuits are usually referred to by attorneys as the “battle of the experts.” The plaintiff’s expert says one thing, and the defendant usually says the complete opposite or, at the very least, contradicts the plaintiff’s expert wherever they can. This is where Gomez v. GNC went off the rails.
In a classic example of the Dunning-Kruger effect, GNC’s IT executive attempted to testify as an accessibility expert in this case. His sole basis for claiming this expertise was talking to some folks and running the site through two free online tools that returned zero errors. I find the latter statement suspicious, having NEVER had a free tool run return zero errors on anything I ever ran. Automated accessibility tools, especially free ones, are notorious for false positives. There are three reasons why you can never, EVER rely on automated accessibility tests as the only component of your accessibility practice:
On a good day, automated tests only cover 30 % of the WCAG accessibility success criteria.
Automated tests can’t run on mobile devices. Mobile web (which is hopefully responsive) must constantly be tested independently and must always be tested manually with screen readers.
People with vision loss don’t analyze code; they use screen readers. The only way you know with 100 % certainty what they are experiencing with your website is to have a human test it with a screen reader.
Fortunately, the judge saw through the defendant’s claims and decided that he was not an expert under a legal review known as the Daubert test, stating in the written ruling that GNC’s IT exec was not even minimally qualified to provide an expert opinion on web accessibility.
The judge granted the plaintiffs’ motion for summary judgment and struck the GNC expert’s testimony. That means she said, “I don’t need to hear anymore. Plaintiff wins.” such is quite rare. However, it happens occasionally, more often, when the defendant requests the motion for summary judgment and rarely when the plaintiff requests it. And I can’t tell you when the plaintiff has been granted a motion for summary judgment in an accessibility class action lawsuit. For the judge to have done this, she had to determine that if every fact were decided in favor of the defendant (GNC), Gomez (the plaintiff) would still win. It is tough to screw up so badly that this happens, but GNC seems to have found a way, and that road was paved by using an expert who wasn’t an expert.
Becoming an accessibility expert requires a LOT more than chatting with a few friends who may (or may not) know something about the topic and having someone else run a couple of free tools. I wrote an article a few months back detailing a comprehensive list of actions an aspiring accessibility expert should consider undertaking. In short, they include attending meetups and camps, webinars, formal training, and conferences, learning to use assistive technology, getting IAAP certified, and volunteering for disability-related causes. Being an accessibility expert requires an in-depth understanding of assistive technology, AT-friendly coding practices, and how people with disabilities use AT.
GNC could have looked at People with Disabilities as a business opportunity, and instead, they ignored them. The number one requirement of being a good accessibility manager is giving a damn. Hopefully, GNC finds someone willing to take the job, and hopefully, they don’t report to the executive who thought they knew everything they needed to know.
A fantastic article as always, thanks for sharing. I did want to make one observation, however and in making it, I confess that I do not know if it is applicable to this particular case in question. In my research, I have come across a few situations in which accessibility tools do in fact return 0 errors. While this is extremely rare, one such cause appears to be certain accessibility overlays which do something to block browser-based accessibility tools from running correctly. I presented on the topic of accessibility overlays during a past Axe-con and one of the sites I referenced in my presentation did exactly this when in fact the site contained numerous code-level accessibility issues that should have been flagged. Again, this may not be relevant to this particular case at all, but I felt it worth mentioning as unfortunately, it is possible to thwart accessibility tools in such a way that they return 0 errors. This is particularly unfortunate because it creates an even greater false sense of security. Again, thanks so much for yet another awesome article, I look forward to each and every one. :)