Streaming video on the Web: a good example of more work to do
Yesterday we announced the HTML5 Recommendation. One of the most significant features of HTML5, and one that has been deployed for some time, is the <video> element, which will make it easier to include video in pages and applications without requiring users to download plug-ins.
There is already strong browser support for video today, but we have more work to do on interoperable support for streaming video. That is why we are working on a number of specifications to support streaming media interoperability, including Media Source Extensions, currently a Candidate Recommendation.
We ran into live stream interop issues as part of planning our W3C20 Webcast today (from 3pm-6pm Pacific Time) and ensuring the widest audience as possible. The deployed solutions we found (and will be using) rely on Flash plugins and other platform-specific approaches such as HTTP Live Streaming (HLS).
Despite that limitation, we are happy to offer the live stream with captions to those who cannot join us in Santa Clara.
Interoperable streaming is just one area where we want to make it easier for developers and users to play video and audio on the Web. We still need Royalty-Free codecs, the ability to play the content on second screens, improved support for accessibility, and more.
I have read the implementation report for HTML 5 in addition to this article.
When HTML 5 first hit the scenes, the idea of native video support was the feature being pushed out aggressively.
All of the sudden, discussions about the problem of codecs appeared and, then, disappeared without any solution.
The implementation report talks about failures with WebIDL and other corner cases (tech jargon without meaning).
Here's my issue with all of this: HTML is not a core language, standard, or even ready for use.
Core standards are as perfect as possible. The laissez-faire attitude for addressing the implementation problems in the implementation report does not support a thing being considered core or standard.
Remove those items not fully implemented or ready and put them in HTML 5.1 or Nightly or whatever, but do not keep them in what is supposed to be core.
HTML 5 is currently Swiss cheese with a lot of holes. "We'll get to them when we get to them" is the attitude over prioritizing these issues.
After 20 years of the W3C and 25 years of the web, I guess I am expecting more from the think tank who bought into HTML 5 in the first place.
Think about it, a core standard is central to everything else built around it. If the core isn't as perfect as possible, then everything built around it will simply amplify those so-called corner cases of minor imperfections. Is this what 20-25 years of collective experience is trying to sell to the public?
Regardless of all the talk, native video was more than 50 percent of the selling point for HTML 5. Without a fully-supported codec, then HTML 5 is simply a lot of fluff with no real substance. I am not sold.
I am so sick of this crap.
XHTML 1.1 was shelved for an inferior HTML 5. RDFa was castrated to RDFa Lite to appease Google and schema.org still decided to push the inferior and less user-friendly Microdata.
ARAI 1.0 was pushed into Recommendation, but causes XHTML 1.1 documents to fail WCAG 2.0 because ARIA will cause those documents to fail validation (Guideline 4).
Additionally, since the only true markup Standard for the web (SGML) was removed from HTML 5, then HTML 5 cannot be a standard. HTML 5 is simply a collection of political compromises and kowtowing.
The fact that the B and I tags were given some bs description and remains in HTML 5 is proof of the aforementioned statement.
In fact there is a lot to do, but considering what is about to happen, I think we're about to get some good result!
For exemple, playing content on a second screen is pretty awesome!
BTW, great article!
The more important thing is a browsers and platforms supporting of these innovations, because, for example iOS devices still doesn't playing flash video and IE is always the only browser who will be "last" for all innovations.