Ogg codecs dropped from HTML5

Posted Jul 9, 2009 16:13 UTC (Thu) by lambda (subscriber, #40735)
In reply to: Ogg codecs dropped from HTML5 by dwmw2
Parent article: Ogg codecs dropped from HTML5

The HTML 'standards' have always been mostly theoretical, and it's naïve to think that this is going to change with HTML5.

Actually, part of the point of the HTML5 standards process was to move away from this, and actually try and write standards that would be implemented. The W3C tried to move HTML to XHTML (which in practice doesn't achieve much more than pages completely failing to render if you have one syntax error), and then write backwards- incompatible standards (XHTML 2) that did very little that anyone actually wanted while making a few architecture astronauts happy.

The WHATWG groups was started by Mozilla, Opera, and Apple to try to actually improve HTML in ways that would benefit the web and be implemented by browsers. This involved "paving the cowpaths," or writing specs for all of the things that had become de- facto standards like XMLHttpRequest or contenteditable (which helps make the job of new implementations easier, as you don't have to reverse engineer what the other browsers do), as well as specifying new features that authors actually want and vendors are willing to implement.

Now, Microsoft has not been very involved in the HTML5 process. They have actually implemented some new HTML5 features in IE 8, but there are a lot of issues (like the <video> issue) that they haven't made any statement one way or the other on. Apple has been highly involved, has helped invent features like <canvas> that have become very popular. Apple saying that they cannot implement something carries some real weight with the WHATWG, and Ian Hickson really does not want to go down the path of previous HTML versions in which features are added to the spec just because they make the standards committee feel good while having no chance in hell of actually being implemented by several major browser vendors.

If Microsoft committed to Theora, I'm sure that you could turn the heat up massively on Apple, and probably get them to implement it. However, given that Microsoft already licenses H.264, and hasn't said anything, it's more likely that they fall on the other side of the debate. With two major implementations refusing to support Theora, it means that writing it into the spec is just wishful thinking, and wishful thinking is one of the things that HTML5 is explicitly trying to avoid.