I was a fan of XHTML for a while, largely because it made it easier to use XML tools to verify my documents. But that became a non-issue as HTML parsing tools (specifically what I could use in Ruby tools while developing) got better.
I wasn't serving my pages as application/xml, and on consideration of what Ian Hickson wrote (http://hixie.ch/advocacy/xhtml) I went back to using HTML 4.
I'm puzzled by suggestions to use HTML 5. it seems a peculiar reversal from the "design with Web 'standards'" mantra. It's not a W3C recommendation, it's still a draft.
If XHTML is not working out as planned, why not use HTML 4.0.1, a current W3C rec that works?
Using HTML 5 now would only mean using wellformed documents and dropping "deprecated" attributes and elements from your HTML document (e.g. not using font, cellspacing, etc attributes). You don't have to use the new elements introduced (since IE pretends such elements don't exist).
"You can already feature detect and use audio tags where available etc."
OK, but isn't this the grounds for people griping about various flavors of IE, that you have to have behavior checks for X number of browsers because they do not follow a W3C rec?
I think you'll find that the "X number of browsers" you need to check for and use fallbacks from HTML5 are "various flavours of IE". So please, let the griping continue.
This is because Microsoft has, for a long time, dragged its feet regarding the web because the web is a threat to its core business. In fact two main strands of HTML5 development are reverse engineering IE peculiarities and making them a standard so that others can interoperate and investigating ways to fix up IE with javascript shims so that it can support HTML5 without any help from Microsoft.
This is not the "right way" to do things but when the monopoly browser (and OS) producer varies between apathy and antagonism towards the web that's how progress gets made.
I was a fan of XHTML for a while, largely because it made it easier to use XML tools to verify my documents. But that became a non-issue as HTML parsing tools (specifically what I could use in Ruby tools while developing) got better.
I wasn't serving my pages as application/xml, and on consideration of what Ian Hickson wrote (http://hixie.ch/advocacy/xhtml) I went back to using HTML 4.
I'm puzzled by suggestions to use HTML 5. it seems a peculiar reversal from the "design with Web 'standards'" mantra. It's not a W3C recommendation, it's still a draft.
If XHTML is not working out as planned, why not use HTML 4.0.1, a current W3C rec that works?