orderedlist.com starting to publish articles on how to use structural tags and the new doctype in existing browsers and following (and following up on) “html5” and “html 5” on Twitter almost becoming a full-time job are indicators HTML5 is getting more stable. The issues graph suggests the same.
As you may recall, the plan of the editor is to get to Last Call by the end of this year. As a result, Ian has been focusing on addressing the remaining feedback and splitting out parts of the HTML5 draft. Not alone actually, Adam (security fame, Google Chrome) is helping with splitting out content sniffing and Dan (former HTML WG co-chair) intends to work on splitting out the URL section.
All in all, that removes content sniffing rules, URL handling, and Web Sockets from the HTML5 draft. (For what it’s worth, XMLHttpRequest was at some point part of HTML5 as well.) This is probably a good thing as they are quite orthogonal from HTML5 itself, e.g. the XMLHttpRequest specification currently references HTML5 URL handling as they are in fact the same. Ian made a list of other sections that could be taken out, though it is unclear at this point how much of that will happen.
I’m curious to see what will happen when some of these drafts hit the IETF (e.g. URL handling and content sniffing) as I assume they are quite controversial. Running code is there, but rough consensus remains to be seen.
I think that all this cutting is welcome by most people; myself included. Even if HTML5 might be delayed because of the shrinkage, it will be a better specification in the end.
The interesting thing now that Ian has fulfilled his part of the deal is to see how many of the proposed parts will be picked up by dedicated editors. I'd love to pick up one or more had my time permitted it. Highly unfortunate, it does not. :-(