- From: Nick Kew <nick@webthing.com>
- Date: Wed, 07 Sep 2005 13:44:03 +0100
- To: www-validator@w3.org
Christoph Schneegans wrote: >>While some of them may be fixable with the current parser, the plan >>for addressing these shortcomings long term is to make use of a >>specialized XML processor. This long term is a frustratingly long time coming. > A specialized XML parser is the only reasonable choice for XML > documents. I think there's absolutely no sense in trying to make > OpenSP an XML parser. Indeed. It was IMO a valid stopgap choice for a time. But how long does that extend? > >>This requires some fairly big changes in the code - which is one >>reason why it's taking so long - and is not without its own issues >>(determining when to use the XML processor and when to use the SGML >>parser, for one). > > When in doubt, let the user decide. <http://valet.webthing.com/page/> > offers two parsers. Exactly my view (surprise:-). But Page Valet also has some pretty good heuristics, which apply when the user leaves it set to auto. There's no reason W3C shouldn't adopt the same. Basically you select on content-type, and just explicitly sniff for the XHTML1.0/AppendixC doctypes as a special case in text/html. But we do that already! Schneegans is offering a different service to any of the DTD-based services. As with the Relaxed service, we should welcome it as a useful tool that _complements_ DTD validation. But for HTML and XHTML we still have a bottom line that they are expressed by DTD, so this must remain the bedrock of the validator, notwithstanding its limitations. -- Nick Kew
Received on Wednesday, 7 September 2005 12:44:21 UTC