- From: Kynn Bartlett <kynn-edapta@idyllmtn.com>
- Date: Sat, 23 Dec 2000 14:04:55 -0800
- To: love26@gorge.net (William Loughborough)
- Cc: <w3c-wai-gl@w3.org>
At 06:26 AM 12/22/2000 , William Loughborough wrote: >Both the indexing and testability features are central to what we're about. At least at the P3 level it should be a requirement that one be able to test if a site conforms before trying to access it, else we become enmired in having to post yet another whine to Webwatch or whatever. Can you define what you mean by "testable"? My _assumption_ is that you mean "testable automatically" by some sort of mechanized process, which determines whether or not a resource can be utilized by the user; this then gives the user the option of accessing the site, or...what? I would like to see an example use-case of this scenario defined. I will provide my own guess at what you mean: TESTABILITY USE CASE Amanda is a blind web user; she uses a screenreader in conjunction with a "traditional" visual browser in order to access the web. She has javascript disabled, as most javascripts either do nothing for her (swapping images on mouseover), or are too complex and interfere with the operation of her screenreader. Therefore, her requirements for web access include (a) text equivalents for visual components, and (b) access to content even though javascript is turned off. Amanda may have additional requirements, but for today we will keep this rather simple. Amanda desires to buy a plane ticket. She starts up her web browser and goes to a search engine, where she locates a major travel web site. As she clicks on the link, her browser queries the travel site, requesting a self-reported WAI conformance claim (which is stored via RDF in a file on a standard place on the server, e.g. /wcag.rdf in the root directory). Amanda's browser compares the RDF summary of accessibility with what it knows of her own capabilities; it determines that this site does not claim conformance with the portion of WCAG which mandates alternative text for images. As this is a composite for the site, it may or may not apply to every function of the site, so this possibly means she'll be unable to perform the task she wishes to perform. Amanda's browser, before loading the site, prompts her with a question: "This site does not claim sufficient accessibility for you to fully access all functions of the site. What do you wish to do? Continue, Abort, Test." Amanda can choose one of those functions -- "continue" lets her continue on to the site anyway, "abort" means she doesn't follow the link, and "test" means that some sort of automatic test is invoked to verify the self-reported accessibility claim. Kynn's note: If Amanda chooses "test", I am honestly not sure what should happen -- I'm not sure how William's suggestion of Amanda being able to test the site before accessing it will work, especially as many of the questions asked can't be _automatically_ verified. For example, there is no way to test whether or not a site will be usable without javascript enabled, short of attempting to use that site. In this case, Amanda decides to take her chances. She continues on into the site and finds she is able to navigate throughout the site; at the top level of navigation, the claimed problem with image accessibility does not seem to exist. (Perhaps it's lower in the site somehow.) So now she goes to choose a seat on the flight she wants. Checking an embedded RDF claim on the flight choosing page, Amanda's browser discovers that it claims javascript is necessary in order to proceed. The RDF also gives a reference to another page on the site which duplicates the functionality but does not require javascript. Amanda's browser, before displaying the page, prompts her: "The page ahead claims to require javascript, but you do not have javascript enabled. An alternate page is avaiable. What do you want to do? Continue, Abort, Enable Javascript, Go to Alternate Page, Test." Amanda chooses "go to alternate page" and happily chooses her seat and eventually concludes her registration. (Kynn's note: Again, I am not sure what would have happened if she had chosen "test.") How's that for a start? Is that what you're talking about, William? -- Kynn Bartlett <kynn@idyllmtn.com> http://kynn.com/ Sr. Engineering Project Leader, Reef-Edapta http://www.reef.com/ Chief Technologist, Idyll Mountain Internet http://www.idyllmtn.com/ Contributor, Special Edition Using XHTML http://kynn.com/+seuxhtml Unofficial Section 508 Checklist http://kynn.com/+section508
Received on Saturday, 23 December 2000 17:28:34 UTC