- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Thu, 12 Feb 2009 11:28:43 +0100
- To: Robin Berjon <robin@berjon.com>
- CC: Robert J Burns <rob@robburns.com>, Jonas Sicking <jonas@sicking.cc>, Lachlan Hunt <lachlan.hunt@lachy.id.au>, "L. David Baron" <dbaron@dbaron.org>, HTML WG <public-html@w3.org>, plh@w3.org
Robin Berjon wrote: > > On Feb 12, 2009, at 10:30 , Julian Reschke wrote: >> The problem is that you do not always have context. >> >> For instance, XHTML elements can appear in many other XML documents, >> re-using the document markup semantics. In this case, you frequently >> have a single element, and no context at all. > > That's a problem of the host language. Some elements need context and > that's the end of their story. I don't think that <li> is broken because > you don't know what type of list item it is without its container, or Where's the problem here? If some vocabulary uses <xhtml:li>, that means "this is a list item". The container format may define its own list, for instance. > that svg:rect is broken because while it retains its semantics it cannot > be rendered without a container to define its viewport. > > If a host language reuses from XHTML, then it's up to it to provide the > necessary context for interpretation. But that's a new situation. Until today, to find out what xhtml:foo means, you checked the HTML 4.01 or XHTML 1.whatever spec. Once we allow two different parties to continue development of the language, there's no single answer anymore. So this breaks something that worked just fine before. BR, Julian
Received on Thursday, 12 February 2009 10:29:31 UTC