Comments on: Is HTML on the web a special case?

By: zcorpan

zcorpan — Sat, 03 Feb 2007 00:40:53 +0000

Test cases often have “completely bizarro markup that no author or tool I can imagine would ever produce”. The purpose of test cases is to test things, not to reflect what authors or tools produce. I didn’t choose which browsers were to be tested; I asked on a forum if people could test their mobiles, so the list of browsers should reflect what people actually use. The “confusion about what browser is really in use” was due to some testers not reporting the UA strings along with their results.

If my research didn’t convince you then I encourage you to do your own research. Henri Sivonen has a more complete set of tests at http://hsivonen.iki.fi/test/mobile/toc.xhtml

Cheers,

By: Anne van Kesteren

Anne van Kesteren — Sat, 03 Feb 2007 00:08:51 +0000

Micah, ah, that’s certainly true for HTML parsers. If there’s ever an XML 2.0 though with graceful error handling as proposed it will most certainly be able to handle namespaces.

This can’t really be merged with HTML parsing though. Although I suppose parts could be shared.

Regarding mobiles and XHTML. What are you basing your statements on? Mobiles can just as well handle HTML. That they perhaps only support a subset of the elements is a different issue and has not much to do with the syntax you express these features in.

By: mdubinko

mdubinko — Fri, 02 Feb 2007 17:23:56 +0000

Hi blog.dlade.net,

Sorry if I wasn’t clear. I wasn’t aiming for a tone of ’surprise’. :) My goal is to get lots of folks to look at the issues here, rather than digging in behind one side or the other.

I agree with your premise that “For a mobile developer, there is a clear reward in going with XHTML. You are much less likely to have your page break on a random phone if you stick to XHTML MP than you are if you go with HTML.”

That’s why I disagreed with the methodology of that little mobile experiment I linked to–it’s true that many browsers fail to implement XHTML flawlessly, but despite this, XHTML is still farther ahead in mobile. In that respect, XHTML is living up to it’s goal of enabling smaller, simpler devices. -m

By: mdubinko

mdubinko — Fri, 02 Feb 2007 17:07:47 +0000

Hey Anne,

You’re in a better position to answer that than I am. :-) But all the HTML parsing rules I’ve seen have avoided namespaces and URI baggage. At best, they allow meaningless xmlns attributes. (which is a good thing in my book)

This is *probably* a rich enough topic to merit a future article…

Thanks! -m

By: blog.dlade.net » XHTML adoption in the mobile world

blog.dlade.net » XHTML adoption in the mobile world — Fri, 02 Feb 2007 13:35:16 +0000

[...] Micah Dubinko mentions with a tone of surprise, both in his article Is XML 2.0 Under Development and in a subsequent blog post, that despite the poor browsers, XHTML adoption is still farther ahead on the mobile web then the desktop web. [...]

By: Anne van Kesteren

Anne van Kesteren — Fri, 02 Feb 2007 11:50:19 +0000

Why wouldn’t it have namespaces? I think it would support the full “infoset” in so far I understand the infoset.