W3C home > Mailing lists > Public > wai-xtech@w3.org > August 2002

[XAG] auto-testing cp 1.1, 1.2, 2.1

From: Charles McCathieNevile <charles@w3.org>
Date: Fri, 16 Aug 2002 20:42:40 -0400 (EDT)
To: WAI Cross-group list <wai-xtech@w3.org>
Message-ID: <Pine.LNX.4.30.0208162036590.29412-100000@tux.w3.org>

comments are based on the June 17 2002 editor's draft -
http://www.w3.org/WAI/PF/XML/xag-20020617

Guideline 1 - how to test for whether media alternatives are permitted?

For checkpoint 1.1 and 1.2 there are only very vague diagnostics that I could
come up with. For elements who have an attribute that has a URI value, having
content is good. It means that there is a possibility to provide content
instead of anything that is included by reference, and to provide an explicit
external alternative to things that are included as content. If people have
used Xlink, then testing is particularly important for links which are
embedded content rather than replacing the page.

Testing both the DTD/Schema and some instance documents is helpful, since in
DTDs it is not clear that an attribute value is a URI (as I understand it),
but if that is typical of instance documents then it can be assumed. This
test seems likely to give false results complaining about elements that are
no problem, but fewer when testing can be based only on included content by
inspecting the xlink. It  also seems likely to miss content that is included
directly - for example I wrote a paper using a DTD that required me to
directly include TIFF or EPS image format data, so there was no URI. I am not
sure how to test for that, although some could be done by looking for
well-known strings such as identifiers for image types.

A second test would check for the containment model where elements that
included media could appear - if they have to be inside another element which
can have variable content then that could be used as the method (this is what
SMIL does).

In some cases passing checkpoint 1.3 would suggest that the author has met
these checkpoints. There isn't a formal proof, but since it could be a valid
technique for meeting these checkpoints it is probably a positive indication.

I think these checkpoints, and therefore these proposed techniques are
closely related to checkpoint 2.1

chaals


-- 
Charles McCathieNevile    http://www.w3.org/People/Charles  phone: +61 409 134 136
SWAD-E http://www.w3.org/2001/sw/Europe ---------------- WAI http://www.w3.org/WAI
 21 Mitchell street, FOOTSCRAY Vic 3011, Australia       fax(fr) +33 4 92 38 78 22
 W3C, 2004 Route des Lucioles, 06902 Sophia Antipolis Cedex, France
Received on Friday, 16 August 2002 20:42:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 13:15:37 GMT