YouTube Embraces HTML 5 Video, Experimentally youtube html5 YouTube recently stated that they want to help push HTML 5 to the forefront and have begun a more robust experiment to use it on their site. You have to opt-in to beta program in order to see the HTML 5 pages and there are some serious limitations to what it can do right now.

In a recent blog post, YouTube has stated that the number one request from user (hard to believe) was that YouTube do more with HTML 5. Why users would even want them to is beyond me since only three browsers currently support their version of it (with MP4 as the file container) - Safari, Chrome and ChromeFrame for Internet Explorer. Of course, firefox does support HTML5 with theora but the YouTube version does not.

YouTube Embraces HTML 5 Video, Experimentally browser market The install base of those three combined is probably in the sub-20% of all browsers. In fact, according to Market Share by Net Applications, Chrome just recently surpassed Safari with a whopping 4.63% install base (see image). It shows that Chrome and Safari combined don't even have 10% of the market at present so you'll forgive me my healthy skepticism I'm sure. I have no data on the ChromeFrame plug-in install base, but even 10% would be astounding.

Perhaps YouTube users are those 10% of the world that use Safari and Chrome? Anyway, so they've started this new test using it as the player. When you get into the test it will begin to utilize the HTML 5 player for videos. However, it won't work for (here are the serious limitations) videos with "ads, captions, or annotations." Now if you're a Partner on YouTube do you really want anyone looking at your Ad-supported videos with HTML 5? I certainly wouldn't if I were. It defeats the whole purpose of the ads. And really, what YouTube is saying is "we can only get this to work for the most basic of videos with no fringe benefits or fancy stuff."

ALSO ►  List of Video Search Engines and Video Search Web Sites (Updated)

So what's their game? Perhaps they don't want to continually rely on Adobe Flash anymore? Perhaps they really do want to, as the blog post said "be part of moving HTML5 forward on the web." But isn't it still years from acceptance, implementation and widespread use?

Of course you'll also need the proper codecs for video and audio which means an h.264 codec and more as well as one of those browsers.

How to get involved?

To try it out, go to the HTML5 page via TestTube or visit this page and join the experiment. This will enable HTML5 video for your browser, provided that it's one of the browsers mentioned above and fits in with the parameters we already referenced. (If you've opted in to other experiments, you may not get the HTML5 player.) You can also enable Feather watch along with HTML5 video for an even simpler, faster YouTube experience.

Feather is another of their projects to reduce the viewing pages to their bare essentials which means a lot less features for the users but faster loading etc.

I'm rather perplexed by their decision to start pushing for HTML 5 adoption so early. I hope this attracts their attention so they an tell us their logic behind it aside from their mention that users were requesting it (not that this is not important). I'm very much interested to hear about it.  I'm not ANTI-HTML 5, but as I said in my previous article, HTML 5 Won't Kill Flash Video Player (at least, not anytime soon).

  • Christophor Rick

    I agree, there's plenty of potential but shoving isn't going to make it work any faster or better. The W3C is notoriously slow. HTML 4 was ratified in what? 1997 or something? What we really need is a completely open source solution. I think I talked about that in a post some time ago but can't remember which. Aside from performance issues and general wonkiness from time to time I see nothing wrong with Flash video players. Everyone has it installed practically so the widespread adoption is going to be key. When Firefox and IE have fully non-beta or plugin-based solutions for HTML 5 then the tag will really be useful.But, it could also be neutered during the ratification process and become just another semi-useless tag at which point we're still left with Flash...

    Yes, Flash could certainly use a lot of work in performance and features. But I don't know that HTML 5 will really push Adobe to do it.

    • Alex

      HTML 5 is motivating Adobe to faster work on better Flash. Right now beta of new Flash Player 10.1 has GPU support for video encoding (h264) - performance is much better as actual version of Flash Player. I tested it with HTML 5 on Chrome and desktop player Kmplayer - both has worse performance than new Flash Player 10.1 (h.264 encoded video).
      Problem with Flash - lot of developers build banners/applications not optimized (with lot of bugs in code) so then performance going down. Lot of Flash developers are not programmers but comes from designers and that's the problem.
      Some solution: ad companies should test banners on performance before acceptation - then developers will be pushed to build optimized code. I'm frustrated when I see banner without complicated animations but with high level of processor consumption...

  • Mark Robertson

    Im actually a bit more optimistic as to the whole HTML5 Video stuff, but we certainly do have a long way to go. What was meant to be an attempt to make web video easier has become somewhat of a complicated issue with regard to the codecs/formats. Vimeo announced HTML5 support as well yesterday. They too are only supporting h.264. It will be interesting to watch this space as it develops.

  • Ken Price

    Nice post Chris. I'm perplexed by it as well. Granted, I've got some research to do on the merits of HTML 5, but it would seem that as 5 gets adopted Adobe will be making improvements to flash at a must faster rate. So, as far as video goes, most will want the advantages a new/improved flash player/container can provide.