'Free' H.264 a Precursor To WebM Patent War?

webmink writes "The MPEG LA seem unwilling to explain why they have extended their 'free' H.264 streaming video policy now. This article unpacks the history of MPEG LA and then suggests the obvious — it's all because of WebM — and the worrying — maybe it's preparing the ground for opening a third front in the patent war against Google."

Comments

Patented Standards

[GuerillaRadio]

It seems an obvious requirement now to me that any 'international standards', as H.264 is described in TFA, should not be written by a consortium that have a collection of patents on the only possible implementation of the standard!

I'm not sure how this would be ensured - maybe the same consortium that pool the defensive patent pool for Linux could start a standards body based around this simple idea.

Re:Patented Standards

[GuerillaRadio]

Unfortunately it's going to be harder for Free software going forward. Try writing an opensource point-of-sale or e-commerce program that can directly process credit cards. You can't without spending around $20,000 for PA-DSS auditing. And I see more of these types of industry barriers to entry popping up.

It won't be harder, it will be impossible - it destroys the mechanism of Free / Open Source software. The way you put it is as if the rise of FOSS is just some kind of unfortunate minority part of the computing world that will be affected, rather than one of the most important, game changing event in the recent history of computing.

Re:Patented Standards

[Anonymous Coward]

ARM and W-CDMA work in similar ways. ARM happens to own the patents and licenses them to whomever for a reasonable fee. W-CDMA works in much the same way as H.264. You have a bunch of companies that decide to share patents into one resource. It makes it easy for other companies to pay 1 fee and then use the technology.

Yes, that's how it works.

And H.264's licensing terms are reasonable

But here I disagree. The difference between H264 and the above is its application space: both ARM and CDMA have a license covering production of equipment, whereas H264 is offered as a user license. The situation would be comparable if an end-user were charged for every communication over W-CDMA, or every programmer was billed for generating ARM machine-code (using a licensed compiler etc).

As it stands, end users are still liable for royalties concerning the codec in addition to the patent fees they may have already paid for by purchasing a digital camera / BR player. That's why the MPEG licensing terms are considered non-free.

Yeah...

[webheaded]

I'm still rooting for Google's format. I don't care about free as in money so much as free as in open source. I don't see how it could possibly be sustainable for every single company that makes a browser from here on out to have to pay a fee to use this codec. If they put H.264 into the HTML 5 spec, that is only going to make it a pain in the ass for browser developers and open source users. It's stupid. This isn't helpful...it's just slight of hand. "Look it's free!" Um no...it actually isn't free at all. I wish people writing all the other articles would acknowledge that a little better. This changes nothing.

Re:It's not "Free" to begin with.

[blackest_k]

The third link has a good explanation of why h264 isn't really free.
First thing to be noted is that any deal they offer today can be withdrawn in five years anyway.
The free bit is only that they will not bill the end user.
the encoding is not free
the streaming is not free
and the decoding is absolutely not free.
The last one means any browser wishing to offer this functionality has to pay for it and unfortunatly it can't pass on the patent protection granted by paying for this so there is no way for firefox to offer this.

So really we should say no to h264 and hope google doesn't get creamed in its patent battles.
     

Re:That makes no sense

[martin-boundary]

In todays heavily legal world, the ONLY kinds of standards you can rely on are ones where members of the standards group hold patents and pledges the protected use of same to people following that standard.

Erm, no. You can't rely on companies and individuals implementing a standard. They'll implement anything they like regardless, and then you'll have interoperability issues, pretty much as we have now anyway.

If you want a standard to be reliably implemented in a wide range of systems, you have to take out the "implementation" part of the equation. That means, distribute a public domain or BSD library which does all the hard work, and let everyone piggyback on that for nothing.

Re:Eminent Domain exists for this

[martin-boundary]

I believe their algorithms should be patentable

Algorithms are just mathematics. I for one believe mathematics should not be patentable, even though it's one of the hardest subjects on earth.

I'm not all that worried

[Sycraft-fu]

Google isn't stupid. They got to investigate the format and patents before they bought On2, and of course after once they owned everything. Also, this is precisely the kind of thing Google would be good at: Looking through large amounts of information and figuring out what is relevant. So My guess is that one of both of the following are true:

1) VP8 (WebM) does not infringe on any MPEG-LA patents, or at least not any real ones. They probably have some overly broad BS ones, but Google probably has examples of prior art. Google did an extensive review and found that there was no infringement, VP8 had been engineered to avoid MPEG-LA patents so that it could be sold without additional license.

2) On2, and therefor now Google, holds patents on critical technologies used in H.264. In the event of any infringement suit, they can pull those out and file countersuit. Having WebM stopped would not be a real big deal to Google. They aren't using it for anything important yet. Having H.264 stopped would be devastating for MPEG-LA. Google could thus force them to license all relevant patents, at not charge, in return for the licenses to the Google patents.

Those are my bets. One or both of those is the case and so Google is confident they can win a game of chicken. This also might explain the move by MPEG-LA to put a permanent licensing moratorium on free H.264 stuff, as well as the fact that there is no suit. They may have looked at things and said "Shit, we can't touch VP8. We could try but we'd almost certainly fail and just wind up with a bunch of legal bills, plug give Google an ironclad thing to point to showing WebM is ok." They may have decided it is better to make H.264 look more attractive and perhaps keep up some nebulous threats to make people think twice about implementing WebM.

Always remember that patent warfare is a dangerous game. The trolls can play it because they don't own anything or make anything losing patents means nothing. In MPEG-LA's case, there could be a lot to lose if things went wrong.

Re:I'm pragmatic

[RAMMS EIN]

``I don't care as much about the codecs being open source or closed source - I'm mainly interested in which format offers the higher quality at the same (or lower) bitrate.''

I, too, care about which format offers the higher quality at the same (or lower) bitrate. But I'm pragmatic enough to not want to have to jump through hoops if I want to install the software on my computer (on whatever OS I happen to be running), fix any bugs that I encounter while using it, or maybe even audit the software. And I don't want to get in trouble with the law over watching or encoding a video.

If MPEG-LA or any other organization wants to offer a video format that I cannot legally use, or for which a codec that allows me to do all the things I want cannot legally be implemented and distributed, that's fine with me. However, I then won't be using the format. Getting slightly better video quality isn't worth the hassle, annoyance, and security risks of typical "we care more about our anti-piracy measures than about our customers" software. If anyone else does want to use such software, fine by me.

The only point where this becomes a problem is when we are talking about standardizing on a format. Standardizing on a format that is not free to implement and use is a very bad idea. Not because it cannot be made to work, but because of pragmatic reasons: there will be barriers, and those barriers will hinder implementation and adoption. All the major players including MPEG-LA recognize this. That's why MPEG-LA is offering this "free for the most common uses" licensing: without that, H.264 would be a huge hassle. Similarly, Flash only became as widespread as it is because the player is freely available for the most popular platforms, TrueType won out over Type1 because of better licensing conditions, and free software is all over the software development world because you can use it without having to jump through hoops.

Ask yourself this: being pragmatic, how much are you willing to pay and how much work are you prepared to do to get to (legally) use H.265, the new and wonderful (and at this point, hypothetic) video format that is even better than H.264? At what point would you, for pragmatic reasons, decide to go with VP12, the (also hypothetic) video format that is almost as good, and free for all to use for any purpose, with free software codecs that you can install with a single click or command?