Monday, February 9th, 2009

A couple of weeks in HTML 5

Category: HTML

Mark Pilgrim release a couple of HTML 5 roundup posts in a row. In his first post he talks about:

Big news #1: r2692, a major revamp of the way application caches are defined. Application caches are the heart of the offline web model which can be used to allow script-heavy web applications like Gmail to work even after you disconnect from the internet.

Big news #2: r2684, which redefines the on* attributes in a way that doesn’t suck quite as much. Also, it defines the widely used (but poorly understood) onerror attribute in a way that matches what browsers actually do with it.

Big news #3: r2685 and r2686 defines a whole slew of important events that are fired on the Window object, including onbeforeunload, onerror, and onload. Previously, some of these were defined on the <body> element, which didn’t actually match current browser behavior.

In his latest post he discusses the news this week that is the beginning of the non-normative section on rendering HTML documents. For those of you not up on spec-writing lingo, “non-normative” means “you can ignore this and still claim to be in compliance with the specification.” It’s advice, not commands. On the other hand, it’s generally useful advice, so ignoring it completely is probably not in your best interests.

Posted by Dion Almaer at 4:02 am

3 rating from 17 votes


Comments feed TrackBack URI

Is there a post about how this is better than xhtml 2.0. Not for anything, but I”m sick of browser wars and having to know 50 languages to please every browser. Insights are welcome–I’m a human services researcher who uses networks and not a web developer by trade, so it would be useful.

Comment by symbioticinternetworks — February 9, 2009
Shortly summarized: xhtml 2 is a clean-break redesign that abandons backwards compatibility, html 5 is a clean-up and extension of html 4.

Comment by Joeri — February 9, 2009
HTML5 ships with very nice things inside but his an horrible mess.
Take everything that could have been made wrong on the web one day or another, take a working group, take a few good ideas, mix all of this together and you’ve got the brand new HTML5
On the other hand XHTML2 has a lot of very good features, it’s strict and clear but a pain in the ass to implement and use.
What risk to happen is that we’ll take what’s best in each of theses and make our own tag soup (that’s already what’s happening today).
Href on all elements = brillant, I’ve been dreaming of that for years
Namespaces = so much nicer and future-proof than random/customs html5 attributes (also that’s already in use)
Role attribute = good, especially compared to the predefined HTML5 classnames (yeah mix style and semantics like in the good ol’ time…)
Canvas = do I really need to tell you about that ?

Comment by ywg — February 9, 2009

Leave a comment

You must be logged in to post a comment.