Ogg codecs dropped from HTML5 [LWN.net]

Ogg codecs dropped from HTML5

Posted Jul 7, 2009 4:59 UTC (Tue) by iabervon (subscriber, #722)
Parent article: Ogg codecs dropped from HTML5

In what way relevant to codecs is the <video> tag expected to be any different from linking to a video file? Chances are that the browser will handle exactly the same codecs with the <video> tag as with URLs visited directly, no matter what the specification said. It's many years too late to be interested in this as a matter of standards. If they wanted to do something useful, they'd provide a way to report codecs in the source URL options in addition to container formats so that browsers can actually tell if they'll be able to play a video without trying it (and maybe finding that it's some obscure codec in an mpeg container, not one of the ones it supports).

If there's anyone who could effectively push open standards in this area, it's the people who do the "Acid" browser tests. If they decided to dock a browser points in Acid4 if it didn't support either Dirac or Theora, or didn't support Vorbis, or didn't support Ogg, then Apple would be pretty likely to come around.

to post comments

Ogg codecs dropped from HTML5

Posted Jul 7, 2009 5:13 UTC (Tue) by gmaxwell (guest, #30048) [Link] (2 responses)

You do realize that the point docking on those tests is directly based on the standards, right?

Ogg codecs dropped from HTML5

Posted Jul 7, 2009 5:58 UTC (Tue) by iabervon (subscriber, #722) [Link]

Theora, Dirac, Ogg, and Vorbis are standards, even if they're not referenced by HTML5. Acid tests have, at times, tested things that were not W3C standards, like CSS 2.1 and CSS 2.0 features that had been dropped from CSS 2.1. They're always tests of those features that the developers think should be available, not of a particular HTML version and its references. HTML4 doesn't specify that ECMAScript/javascipt is supported at all, and the latest revision of HTML4 predates DOM2 and CSS 2.1 entirely. Acid2 tests transparency in PNG, which (like PNG in general) isn't required by W3C standards. So there's no reason Acid4 couldn't simply state that it tests Ogg-related formats as one of the items that they think a browser should support correctly.

Sure, but which standards?

Posted Jul 8, 2009 6:10 UTC (Wed) by khim (subscriber, #9252) [Link]

You do realize that the point docking on those tests is directly based on the standards, right?

Yup - but which ones? ACID3 tests: DOM, ECMAScript, SVG (and SVG fonts), SMIL... These are not HTML5 by any stretch of imagination...

I think Theora and Vorbis fit nicely in the "list of things we need to see in future browsers"...

Ogg codecs dropped from HTML5

Posted Jul 7, 2009 7:20 UTC (Tue) by njs (subscriber, #40338) [Link] (1 responses)

> If there's anyone who could effectively push open standards in this area, it's the people who do the "Acid" browser tests.

Ian Hickson, the HTML5 editor and author of the email you're posting on, is from the (co-)author of Acid2 and Acid3.

I hope Apple comes around, but whatever it is they're trying to achieve by avoiding Theora (legal risk? something strategic?), they may well consider that more important than Acid4 compliance. And if Apple makes some loud and vaguely plausible argument about how Acid4 is so unfair -- they can't possibly comply -- and ignores it, then that runs the risk of making *the Acid tests* irrelevant, not Apple.

HTML5 has a similar problem -- it attempts to make a practically relevant standard (which job the W3C has completely abdicated), but that means that it can't just go around declaring reality to be other than it is. One can see this logic in the email -- it's being dropped from the spec because putting it in the spec wouldn't make a difference either way.

I think Jon's summary is overly pessimistic. This announcement doesn't change anything. The strategy for forcing Apple to bundle Theora was always to create web developer by shipping it in Firefox, and legal confidence by shipping it in Chrome.

Ogg codecs dropped from HTML5

Posted Jul 7, 2009 15:59 UTC (Tue) by iabervon (subscriber, #722) [Link]

It's appropriate for HTML5 to try to be something that people will implement (and, if it's successful, this will be something that the W3C unfortunately never managed). But the Acid tests are intentionally designed to be wishful thinking, rather than reflecting reality. They test a bunch of features chosen to be things that web designers would like to be able to get consistent results out of and currently cannot. And they provide a way for web designers to look at the things that they'd like to use, and compare this against what browsers currently support, so that they can decide how to cope with reality.

The best solution is probably to test, first, that the <video> tag is handled and the browser will, if offered H.263 and Theora, pick one it supports; then to test that, if only offered a single file, it will be able to use it. Of course, there's the question of whether web designers wish everyone could handle Theora or whether web designers wish everyone could handle H.263. I personally wish Theora was what web designers would prefer, but I don't know if that'll be true. I expect the actual answer is that some designers want each.