- From: Shadi Abou-Zahra <shadi@w3.org>
- Date: Mon, 20 Apr 2009 19:59:56 +0200
- To: Christophe Strobbe <christophe.strobbe@esat.kuleuven.be>
- CC: public-wai-ert-tsdtf@w3.org
Hi, Christophe Strobbe wrote: > Dear TSD TF participants, > > One of the issues that were discovered during the implemenation of the > new test samples database is that it would be desriable to have a > "handle" for technical specs referenced by the 'technicalSpec' element > <http://bentoweb.org/refs/TCDL2.0.html#edef-technicalspec>. > > The reasons are > (1) that not every spec has an official URL (although it would be > possible to use e.g. Wikipedia URLs if there is no URL from a standards > body or a vendor for a spec) and > (2) that the 'specName' element is an element type with mixed content > that is prone to accidental changes (insertion of line breaks during > XSLT processing or "indent XML" actions, conversion of entity references > to actual Unicode characters) that make string comparison in the > database less reliable than expected. > > We could address this by adding e.g. an attribute called 'handle' to the > 'technicalSpec' element. We should then also agree on the values to be > used in this new construct (consistency can be enforced by means of > Schematron). For example, we could use values such as the following: > * w3c_xhtml10 (XHTML 1.0), > * w3c_css20 (CSS 2.0), > * w3c_css21 (CSS 2.0), > * ecma_262 (JavaScript), > * ms_wmv (Microsoft Windows Media Video), > * ms_asf (Microsoft ASF), > * ... > > Of course we could also choose more "human friendly" values. Any thoughts? I suggest W3C shortnames or MIME-like names where applicable. Examples: - XHTML10 - CSS20 - CSS21 - ECMA262 - WMV - ASF I think we should maintain casing but not depend on it. For instance, "Java" rather than "java" or "JAVA" etc. Acronyms can be in caps. This will help when presenting these handles to DB users. Best, Shadi -- Shadi Abou-Zahra - http://www.w3.org/People/shadi/ | WAI International Program Office Activity Lead | W3C Evaluation & Repair Tools Working Group Chair |
Received on Monday, 20 April 2009 18:35:58 UTC