- From: Chris Lilley <chris@w3.org>
- Date: Mon, 7 Jul 2008 11:40:37 +0200
- To: Cameron McCormack <cam@mcc.id.au>
- Cc: public-svg-wg@w3.org
On Monday, July 7, 2008, 3:33:29 AM, Cameron wrote: CM> When running that script, it generates three fewer files than currently CM> exist in spec/mobile/1.2/1.2NG/master/rng: CM> a CM> langspace CM> switch CM> Both ‘a’ and ‘switch’ aren’t included in any of the spec chapters with CM> an <edit:schema/> element, so they can probably be deleted from CM> spec/mobile/1.2/1.2NG/master/rng. In both cases, I suspect this is because the RNG is a bit more complex - the content model of these elements is the content model of their parent. CM> (Interesting note: the references appendix is generated in two versions, CM> XHTML 1.0 transitional and XHTML 1.1, the latter using ruby annotations CM> of some authors’ names in the bibliography.) Yes. Its not currently allowed to serve xhtml 1.1 as text/html, and its not currently allowed to put ruby in xhtml 1.0 despite the fact that the language elements are carefully chosen to have a reasonable fallback behaviour if unrecognized. Sigh. If you are looking at streamlining/refactoring, it may also be worth looking at what 1.2 modules do. Currently these all have a copy of a set of scripts, which means bugs have to be fixed in multiple places (but limits the impact of local customisations that particular modules might need). A configurable central script that is called from a per-module stub which specifies the customisations would be a better architecture. I'm not sure if these module scripts are derived from the one for the 1.2T spec, but I think they are IIRC. -- Chris Lilley mailto:chris@w3.org Interaction Domain Leader W3C Graphics Activity Lead Co-Chair, W3C Hypertext CG
Received on Monday, 7 July 2008 09:41:42 UTC