- From: David Woolley <david@djwhome.demon.co.uk>
- Date: Wed, 16 Apr 2003 22:13:17 +0100 (BST)
- To: w3c-wai-ig@w3.org
> and 508 - W3C does not limit itself to "Dominant" or any other kind of > browser.... Lynx isn't an XHTML browser either, and serving XHTML properly breaks the page for most users++. The basic argument is that you cannot serve XHTML 1.0 properly to IE, which means that you largely lose its advantage as you will get away with serving not well-formed XHTML, therefore making it essentially the same as HTML 4.01. The particular concern expressed is that a large amount of not well-formed XHTML will get written, resulting in problems when XHTML browsers become common (possibly putting commercial pressure on the developers to drop the rejection of not-well formed xml+html documents). Most people using XHTML these days are doing it for fashion reasons, rather than because it gives them any advantage. Incidentally, XHTML 1.0 is not compatible with HTML. The compatibility rules rely on the browser not processing HTML strictly according to the definition. / at the end of a tag is not ignorable in a strict implementation of HTML; it can't be ignored through error recovery as it has a specific meaning that is defined - it's not necessarily a syntax error. There is no strict backwards compatibility, even though there is a de facto one for many browsers. Basically, if you serve XHTML any time in the next 10 or so years, you will be breaking the "any browser" principle, and encouraging the propagation of bad HTML habits to XHTML. ++ IE either doesn't understand application/xml+html at all or gives the unknown XML document format treatment. The version of Lynx I have would probably give a download or cancel prompt. Served with a text/html type, IE treats it as HTML, including all the standard error recovery, which results in the typical person authoring for IE being likely to introduce numerous undetected structural errors.
Received on Thursday, 17 April 2003 01:48:41 UTC