HTML5, H.264 and Free Software: it's the wrong game | Hacker News Reader

HTML5, H.264 and Free Software: it's the wrong game

jan.rychter.com

17 pointsjwr7 comments

The blog claims that you can't develop a video codec without tripping over patents but then links, approvingly, to an article written by the lawyer who okayed using Theora in Mozilla, talking about why he thinks there isn't any patent problem with Theora.

(At least the original author points out this patent situation is a terrible situation, I've seen this same point made repeatedly with responses that range from fatalistic acceptance through to a bizarre sense of triumph over freetards.)

If you read that article carefully ("What’s the Problem with Theora?"), you'll notice that while significant research has been made, no one can actually be sure there are no submarine patents. The author doesn't say he "think there isn't any patent problem with Theora", he very carefully says that he hasn't found any.

And that is precisely my point — since you can never be entirely sure, using Theora (or VP6, or any other video codec for that matter) won't solve much in the long term.

BTW, as a related note it should be mentioned that at least Theora developers chose the right license for this kind of code — BSD. This avoids the huge landmine that x264 has set up for itself: the anti-patent clause in the LGPL/GPL (section 7 of the GPL).

> And that is precisely my point — since you can never be entirely sure, using Theora (or VP6, or any other video codec for that matter) won't solve much in the long term.

That's just silly.

It's easier to defend against a submarine patent of which you knew nothing about, than to defend against MPEG-LA in case you just breached their license. Not to mention that nothing guarantees that you won't get sued with a submarine patent even when you have a legally licensed H.264

Plus ... we simply don't have too many patent threats that go to trial. The best case scenario would actually be for Theora to become a widely popular standard, and then for Mozilla to get sued for it ... the public needs such trials, and the supreme court of justice also needs such trials ... how else would could they realize that software patents are so dumb? The reality is that such threats are usually settled out of court.

Not to mention that when a company has the resources and the trial goes public gathering lots of supporters, the patent troll tends to get screwed (see SCO vs Novell).

As a general reply, since you three all used the phrase submarine patent, I'm fairly certain that term gets thrown around in a misleading manner.

A submarine patent isn't just someone coming along and saying I have a patent on what you're doing so stop it and/or give me money. The technical term for that is just "patents", that's what they're for.

A submarine patent is a particular type of patent hi-jinx which is no longer possible in the US (though some may still be waiting to surface, from the time it was possible to create them):

http://en.wikipedia.org/wiki/Submarine_patent

So in the US all software development should stop because you might run into submarine patents. It seems such a sad state of affairs: a true deadlock.

There were some psychological tests that showed that under surveillance people behave differently, they self-censor, they act unnatural. I wonder how this applies psychologically to software development in the US. When everything you could write might infringe upon some patent, how much do you end up self-censoring ?

How much progress is lost because people are afraid ?

You can never be sure about that for anything. That is the sad truth about the patent system.

Eben Moglen [1] predicts major patent reform is coming. I thought this video a great perspective on historical patent law in the US and the absurdity of software patents. It's lengthy but well worth the viewing.

[1] http://www.isoc-ny.org/?p=1009

Reply on news.ycombinator.com