- From: Henri Sivonen <hsivonen@iki.fi>
- Date: Sat, 10 Mar 2007 19:30:25 +0200
On Mar 10, 2007, at 15:43, Elliotte Harold wrote: > Alexey Feldgendler wrote: > >> The tutorials will just say "Use <!DOCTYPE html>". > > What are those of us who wish to use XML tools on our documents > supposed to use? <!DOCTYPE html> if you are using XML tools with an HTML5 parser. No doctype if you are using XHTML5 and targeting an XML parser. > We will need a real DTD at some point, to declare the entities if > nothing else. Straight UTF-8 should be used for interchange on the XML side. > <!DOCTYPE html PUBLIC "-//WhatWG//DTD XHTML 5.0//EN" > "/DTD/xhtml5.dtd"> Browsers won't read arbitrary DTDs by system ID. OTOH, if you want to activate the hacks for enabling XHTML 1.0 entities, you need to use an XHTML 1.0 public ID. A new previously unknown public ID won't work. > I know some browser-centric folks here just hate DTDs and schemas > of any kind; but we will need them, even if the browsers don't. For Web content, browsers matter and the reality is that DTDs don't work. > We will create and use them, even if there's no normative DTD in > the spec. The spec doesn't need to say anything normative about private (non- Web) use. If it is between you and your own entity resolver, using a DTD with an identifier of your choice is fine. However, before serving documents on the Web, you'd need to parse and reserialize so that the serialization that is sent to the public network does not have DTD dependencies. -- Henri Sivonen hsivonen at iki.fi http://hsivonen.iki.fi/
Received on Saturday, 10 March 2007 09:30:25 UTC