The Department of Justice is considering proposed revisions to the Americans with Disabilities Act (ADA) in order to expand the requirement for web accessibility. This move will make it “clear to businesses, educators, and other public accommodations, that their websites must be accessible.” This will extend the legal basis to websites that operate as places of public accommodation under title III of the ADA. The way I read it, this is a really big advance. But wait, there’s more.
What I’m curious about in the same proposal is the following statement: “The Department intends to consider various alternatives for ensuring full access to websites of public accommodations and will solicit public comment addressing these alternatives.” It reminds me of the text ghetto.
In 1999 when I started to make pasadena.edu accessible there was a pervasive and pernicious idea that there would be a text-only equivalent for each page. Who would update the text-only pages? No one. It would soon be a run-down neighborhood. I called this the text ghetto. My team ignored this idea and set out to make every page accessible. When I started to make csun.edu accessible in 2005 there was a copy of LIFT Text Encoder running. LIFT was a scheme users could invoke to render a web page as text-only. This was an advance I suppose: a semi-automated text ghetto. Incredibly, Pratik Patel @ppatel just yesterday told me that right now, in 2013, delta.com has a text ghetto. Bad ideas never die.
Text to Speech
iWebReader and BrowseAloud are two Text To Speech products that can be used to hear text read aloud from a web page. iWebReader uses the elegant Ivona Text to Speech engine and great voices. There are even Icelandic voices. BrowseAloud relies on a plain machine voice. Both have server-side elements. BrowseAloud has a client, but the client requires a page enabled with the server-side element. I did some brief exploring and iWebReader seems very advanced, even to the point of providing table construction information so it can read tabular data correctly. BrowseAloud failed to read a table on it’s own site. Both of these products are useful. Neither are full-scale accessibility solutions.
Invoking such a solution and calling a site accessible does not make it so. The iWebReader table tutorial tells us that some effort still must be made to code accessibly so the page – wait for it – can be successfully read by a machine. Remind you of anything? Most users will not benefit if the site is inaccessible and only this new improved speech ghetto is provided. Is this what is meant by “consider various alternatives?” How many other such schemes are out there? The good is often the enemy of the best.
On February 12 in an advertisement/article called “Universal Accessibility: A New Conversation About Web Accessibility” via @HuffPostUKTech, the author who is the Communications Manager for another text to speech product “Recite” said: “There are innovative, award winning, web products, like Recite, which read websites and online documents out loud, and include numerous features which fix the internet for anyone who currently struggles online.” I rest my case.