- From: giorgio brajnik <brajnik@uniud.it>
- Date: Tue, 06 Dec 2011 11:18:04 +0100
- To: Christos Kouroupetroglou <chris.kourou@gmail.com>, public-wai-rd@w3.org, Markel Vigo <markel.vigo@manchester.ac.uk>, maia.naftali@gmail.com, nadiaf@di.fc.ul.pt, Jorge Manuel Moreira Fernandes <jorge.fernandes@umic.pt>, dsloan@computing.dundee.ac.uk, silvia.mirri@unibo.it, Detlev Fischer <fischer@dias.de>, an@ftb-volmarstein.de, luzrello@gmail.com
Dear Christos, thanks a lot for this question, which I forward to all panelists and rdwg. This is my own answer: I think that corpuses do not need to be static; in fact we could build mini web apps, that are very interactive with sliders, drag&drop, canvases, etc and use them as well as test cases. The real problem is enabling testing tools to dig into either javascript source code (but IMO static analysis of code is going to be very difficult, unless one focuses on one framework at the time, like jQuery or Dojo) or simulate some end-user interactions (ie provide a sequence of events) and do somethingon the resulting DOM. Cheers, Giorgio On 12/05/2011 06:40 PM, Christos Kouroupetroglou wrote: > Hello Giorgio, > > Time was running out... so I'm sending this through email. > Following up Simon's suggestion/question about a testbed corpus for > accessbility testing, I was thinking that an increasing number of > sites today include som kind of interaction elements through > Javascript/AJAX. > Corpuses for IR tests tend to be static, but in the case of > accessibility... how representative and useful results from tests run > in such a corpus will be? > Should this corpus include pages with interaction elements too? > > Regards, > Christos Kouroupetroglou
Received on Tuesday, 6 December 2011 10:18:35 UTC