Hacker News new | past | comments | ask | show | jobs | submit login
Did Google Just Kill Ogg Theora? (newteevee.com)
37 points by jasonlbaptiste on April 15, 2010 | hide | past | favorite | 19 comments



The goal of the Xiph.org Foundation was never make Vorbis or Theora take over the world. The goal was to provide royalty-free multimedia codecs. If VP8 is open sourced and they make the required patent grants, the world gets a royalty-free video codec. We win.

We worked on Vorbis and Theora because no one else was going to do it. Here we are a decade later on the cusp of much of our dream finally coming true.

disclaimer: I co-founded, managed, and still sit on the Board of Directors of the Xiph.org Foundation.


I think it's fair to say your efforts paved the way to a world where a major corporation could actually consider this. You may not have struck the final blow for video codec freedom, but the victory is very much yours as well.

A sincere bravo to you and the Xiph.org Foundation.


Xiph's Vorbis is the only audio codec that's suitable for use alongside a hypothetical Google royalty-free video codec.

Well it's not the only one. Xiph's CELT and Xiph's FLAC are probably interesting too for certain use cases, and maybe the upcoming IETF Wideband codec that Xiph are helping standardize. And they appear to be working on their next-generation audio codec, Ghost, again too, so that's one for the future.

So yes, well done Xiph.


On a related note, somebody once explained to me why India develops satellites, missiles, rockets, fighter planes, nuclear reactors etc which are sometimes half-assed at great expense over many decades when they can probably buy the same things ("comparative advantage" and all that jazz).

The reason is to force sellers to sell their highest-end goods, or to negotiate on price. Because India has it own reactor tech, it has a stronger hand when negotiating with GE/France/Russia to buy reactors. For instance, they can ask for technology transfer and license to manufacture too instead of just a reactor.

So Ogg Theora wins even if it dies, because it kept the Free video codec idea alive until a big backer such as Google emerged for the idea.

Thanks for all the work, Ogg!


India is also making their own nuclear reactors because they have a lot more thorium than uranium, and in order to use it, they need reactors which support thorium fuel. </nitpick>


I wouldn't say so much "making their own" as they are cloning & tweaking the 2 CANDUs we sold 'em a few decades ago.

And yes, CANDU reactors can operate on thorium fuel cycles (lucky for them).


And they need to keep reactors that are not under IAEA oversight in order to build more nuclear weapons.


Thanks for that interesting perspective. I guess it succeeded because western world effectively stopped development of space programs and nuclear technology. Similar strategy for semiconductor (http://en.wikipedia.org/wiki/Bharat_Electronics_Limited) has not worked out very well.


"...indisputably superior to H.264"

Is that really the case? Indisputably? I'd like to learn more about that.

Also, even if VP8 takes over Ogg Theora's position as a competitor to H.264 as a web standard, it will still face some problems.

One is that in moving to HTML5, many videos being served through flash are already encoded as H.264 and could be served through HTML5 without re-encoding. VP8 would still require conversion for a full transition away from flash to be made.

Another problem is IE support for VP8... hopefully they'll be willing to support it in a timely manner.


Why do people keep stating it like an eternal truth that Flash can't support VP8. They've added video codecs at least three times in the past.

All those codecs have been bought in so it's not NIH syndrome. One of them was VP6! And if you think they're allergic to open source codecs then note they just adopted Xiph's Speex as a better alternative to NellyMoser for voice a year or so ago. I'm sure they appreciate not having to pay licence fees as much as the next giant corporation, and understand that customers appreciate that too.

(edit: just realised the post I'm replying to is actually saying something different i.e. that existing Flash video is often H.264 and so can be moved to HTML5 without touching the video file itself. I think the point is that Google doesn't want to move from Flash to HTML5, they want to move from royalty-bearing codecs to open source and open standard friendly codecs, so yes replacing those H.264 files is required regardless. Flash will become optional as a side-effect.)


you fail to realize that windows has codec management built into the system.

a browser should never concern itself with codecs. that's what system APIs are for.

you can teach IE6 about VP8 by just installing a codec into the system. if that's not done by the IT dept. or your neighbor's kid, you either don't have the need to play back VP8 or whoever gives you VP8 content also tells you how to install a codec.


My point with IE is that a lot of content providers don't want to use something that everyone doesn't have access to out of the box. Yes, people can install the codec, but for websites to start using it in a widespread manner it would help for all the major browsers to package it.


Didn't Google just commit to funding Ogg Theora on ARM devices, not two days ago?

http://www.linux-magazine.com/Online/News/Google-funds-OGG-T...

Not sure how that would qualify as 'killing' it..


I've always felt that a factor holding back more widespread adoption of Ogg Theora (or Vorbis for that matter) were the product names. I'm very glad that someone out there was providing a royalty-free codec, but it's frustrating to see them get held back by a decision that made their battle harder than it needed to be.

To some extent - especially in the last decade - a codec is/was a consumer-facing technology, and I feel short, simple names like "FLAC", "MPEG", "VP8", etc make it easier to brand things than a long foreign names that leave most users puzzled.


1. This is nothing but a rumour, this instance seemingly made up out of thin air by that website.

2. There is no evidence whatsoever that VP8 even exists, let alone performs anywhere close to H.264.

3. Even if it does exist, and even if it is opened, there is no reason to believe the usual suspects do not hold patents covering at least parts of it.


I'll still view it as a rumor until Google confirms, personally.


Google started the rumor, in their PR when they bought On2 last year:

“Today video is an essential part of the web experience, and we believe high-quality video compression technology should be a part of the web platform,” said Sundar Pichai, vice president, Product Management, Google. “We are committed to innovation in video quality on the web, and we believe that On2’s team and technology will help us further that goal.”


"indisputably superior to H.264"

Whoo, whoo there, Ben Moskowitz, jump off your stampeding horse. Stop counting beans and pennies and go get some "hands-on" with these things before you talk.


Let's hope.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: