“Google confirmed today that it plans to buy digital video compression outfit On2 Technologies in an all-stock agreement, valued at $106.5m. Mountain View said On2 shareholders will be handed 60 cents worth of Google class A common shares for each outstanding share they hold in the company. Clifton Park, New Jersey-based On2 employs around 60 staff. It started life as the Duck Corporation in 1992. On2’s VP6 codec is licensed by Adobe, for its Flash Player and Flash Lite 3 for mobile phones. Other major customers include AOL, Freescale, Nokia and Sun Microsystems.”
does that mean Google is going to support Theora development by removing some of the limiting Patents? and put its wight to support it as the main codec for HTML5.
What patents are holding Theora back? Wasn’t theora, or rather VP3 from which it is based, completely released and all patents upon it disclaimed entirely by On2?
you are right but all of the development took after opening it is still being held back by avoiding all patented codec compression logarithms.
compression logartihm (base 10):
100 -> 2
1000 -> 3
….
Interesting compression technique..
Theora already has full rights to the VP3 codec patents, garnted to Theora by On2. Theora is based on VP3.
I would say that this step means that Google wants to open VP8 or maybe VP7 as well so that Mozilla will be able to include it in Firefox, and Webkit will also be a ble to include it (and hence also Chrome).
This way, sights like YouTube will be able to host better quality video, using HTML5, at lower bandwidth requirements and yet not have to be beholden to other parties IP, and not have to pay a solitary red cent of royalty payment to either MPEGLA or Adobe.
That is my take on it.
Edited 2009-08-06 03:13 UTC
Youtube’s video library is encoded in H264 and On2’s VP6 video codec. It’s in google’s interest to have this codec in all browsers – whether they have Flash plugin installed or not. So it is quite certain that they will make VP6 an open standard and try to make it the HTML5 video codec.
All the reasons why the HTML5 working group failed to choose a codec will be moot.
Mozilla and Opera dislike H264 because it is not an open standard. Google doesn’t like Theora because it’s bitrate is not on par with H264. Apple dislikes Theora because the iPhone’s hardware cannot efficiently decode it. Now everyone should be happy.
Edited 2009-08-05 16:55 UTC
“it is quite certain that they will make VP6 an open standard and try to make it the HTML5 video codec.”
I wonder if Google is really THAT keen on getting rid of flash. I mean… they are potentially gonna pay $106.5m for this. This strikes me as quite a lot of money for just that. Even for Google!
Another possibility might be that they are going to build their own kind of player, spread it free of charge and reap all the benefits of having a widely used player. For example to spread their adds and to gather viewer data.
Edited 2009-08-05 17:40 UTC
I would say that is a paltry sum of money to get rid of the Flash virus. I think if that was all it took, many companies/organizations would poney up the cash. I’d chip in!
Simple, Flash support is shit on Linux, and they now have Android and soon Chrome OS coming. They can make special plugins for their devices for Youtube videos, but their devices will look like crap if Flash is hobbled everywhere else. The solution: make Flash a non-issue.
Not sure about this. It would have been far cheaper to strike a deal with Adobe. Linux support is flaky because they don’t make much money from it. Give them a few millions and I’m quite sure that it would become a first-class citizen in no time.
Remember that Flash won’t go away as soon as the codec issue with HTML5 is resolved. It might even take years. And Flash isn’t only about videos!
I think it is more likely that Google wants On2’s VP8 to become the agreed codec for the web. They would want Mozilla and Webkit browsers to be able to include it natively, under HTML5, in a similar fashion to the way that Theora is already embedded.
That way, Google gets to deliver better quality videos at lower bandwidth using HTML5, pay no royalties, and is no longer beholden to either Adobe or MPEGLA.
Given that YouTube hosts many thousands of videos, and given that YouTube daily serves many thousands of videos uniquely to many thousands of customers, and given also that the licensing language of MPEGLA reads like it will start to charge a royalty per transmission (i.e as if it were television) for h264, Google could be looking at saving perhaps $100 million per month by using their own codecs rather than h264.
Pretty decent rate of return.
The only thing is, Google would need acceptance of their codec (which looks like it will become VP8) by all browsers for this to work.
To do that, Google would have to release the codec under open source terms.
Edited 2009-08-06 03:30 UTC
There is an enterprise cap of 4.25M $/year for 2008-2009 and 5.0M $/year for 2010.
Right now, fees for free television broadcasts aren’t bad: a maximum of 10000$/year for reaching a market of 1M+ households or a one-time fee of 2500$ per encoder. I’m not sure that that services like YouTube would qualify as “broadcasts”, though. There’s no word on free title-by-title transmissions.
Those fees are not affordable by many websites, but Google could easily afford the maximal fee.
That said, buying On2 for freeing up their technologies might still make sense. It would be a fantastic publicity stunt. They could promote a complete video solution for corporate customers. As I said before, I don’t think Google is here for charity!
Well, H.264 is an open standard… The fuss comes from its patent-encumbered status.
If Google opens the VP6 standard and grants an irrevocable free license to its patents, then it might be a great solution for the HTML5 codec issue.
That said, I don’t think Google will buy a company for 100M $ just for charity towards open source. They must have plans in mind, even if it’s only brand name reinforcement. We’ll see in the following weeks.
Edit: who the hell created a filter to convert M $ (with no spaces) to Microsoft?
Edited 2009-08-05 18:35 UTC
Microsoft
M $
Microsoft
Microshaft
Testing
[i]Edited 2009-08-05 22:06 UTC
Strange definition of open.
More typically, “open” would be held to mean “unencumbered” (i.e patent free, or at least royalty free for anyone to implement), rather than simply meaning “published” or “not a secret”.
Edited 2009-08-06 03:21 UTC
Not really. Some consider patents fees as acceptable for ‘open’, some don’t. It’s ‘open’ for most standard organisations like IUT, IETF, even ISO.
No matter what, it’s quite different from a proprietary standard!
As far as the important body of people are concerned, i.e. that body with the most votes in the matter if all interested parties are consulted, RAND terms are not open.
Making something actually open (and most decidedly not RAND) reduces costs. Spending a bit of money to reduce costs long term is a perfectly reasonable approach to running a business.
http://www.theregister.co.uk/2009/08/06/google_vp6_open_source/
A number of people have concluded that this is what Google are probably up to.
Makes sense.
By this definition, there isn’t many open standards beyond the software world…
As for making sense, I don’t know. Are they making any money from YouTube/Google Video? About 99% of their revenue comes from advertising and there’s only an handful of adverts in YouTube/GV.
There are at least two ways to improve one’s margins … charge more money for your products, or reduce your costs in production of your product.
Giving a rights for implementing a codec to open source is therefore not an act of “charity” … it is actually just improving your margins via reducing production costs as opposed to improving your margins via raising your prices.
YouTube does not use VP6 at all. It was using H.263 and now it’s using H.264. The “YouTube uses VP6” is just an old myth that some people keep repeating over and over again.
This is interesting news for at least four reasons:
– Internet: it could mean a major boost for the HTML 5 video codecs
– Internet: changes in the things underlying YouTube
– Android: use of video embedding, sharing etc.*
– Competition: though still far ahead on Silverlight, a challenge for both parties is always productive
(* – A few years back, Google bought YouTube. This was criticized by many as “a risky step” since YouTube might be “just another fad”. This turned out not to be true, though I do believe Google will be thinking of a new way of sharing videos.)