RE: Testing dynamic pages

Hi Charles - are you looking at the ability to test dynamic pages generated
based on form data and behind form submits? Or is there something about the
interaction of the pages and independent of the generated content of any
individual page? For the first part, I think a number of tools support it in
various ways. The way the Watchfire tools work is to scan the site, collect
forms, and allow you to fill out the form, then rescan to pick up the
resultant pages. You can provide multiple sets of form data if you need to
test multiple scenarios, and can continue the process if there is a chain of
forms as in the WAI meeting registration process. If you're looking for the
second part, I'd be interested to know what sort of functionality you're
thinking about, but am not aware of any tools that do anything around that.
Michael

> -----Original Message-----
> From: Charles McCathieNevile [mailto:charles@sidar.org]
> Sent: Wednesday, December 31, 2003 8:13 PM
> To: w3c-wai-er-ig@w3.org
> Subject: Testing dynamic pages
> 
> 
> 
> [Bcc'ed to EuroAccessibility technical groups]
> 
> Hi folks,
> 
> one of the issues that is going to come up for the EuroAccessibility 
> Task Force on testing tools is how to check the performance of tools 
> for dynamic pages - for example, can a particular piece of sooftware 
> test the 4-step process of registering for a W3C meeting, or the 
> process of applying for a credit card online?
> 
> Thre are lots of open questions about the best methodology to apply 
> here, and particularly about how to classify the results, but I am 
> wondering if anyone has a tool that can actually walk through the 
> process, and how this is done.
> 
> The EuroAccessibility task force is likely to look seriously 
> at how to 
> do this in the second half of this year (at least  it is this year in 
> Europe now :-) and having some idea now about any tools that are 
> designed to handle this problem would be helpful.
> 
> Some preliminary thoughts:
> 
> javascript extensions to something that functions as a browser, or 
> tools that have access to pprocessing back-ends, seem to me likely to 
> have an easier time of this. Although in principle it is possible to 
> save each page through the process and run it through a local 
> test, it 
> might not be that simple in practice to get all the scripts, styles, 
> multimedia, etc.
> 
> Testing this manually (without tools) should be largely possible, but 
> it is a complex job, and it would be nice to be supported by software.
> 
> cheers
> 
> Chaals
> 
> --
> Charles McCathieNevile                          Fundación Sidar
> charles@sidar.org                                http://www.sidar.org
>

Received on Monday, 5 January 2004 10:51:19 UTC