The Looming Video Codec Fight 235
itwbennett writes "With both Apple and Microsoft promoting HTML5 standards, you'd think that there would be joy in software freedom land. But instead there's another fight brewing. 'While it is true that HTML5 video is a step in the right direction, we also have to take into consideration the underlying codecs used to deliver the video content,' says blogger Brian Proffitt. The problem, says Proffitt, is that Microsoft and Apple's browsers will be supporting only the proprietary H.264 video codec by default. But Google supports only the WebM (VP8) and Ogg Theora codecs. 'So, basically, if Ogg Theora content starts making a dent in Apple and Microsoft's bottom line, or that of the MPEG LA's, then expect to see a lawsuit or two headed Google's way after 2015,' concludes Proffitt."
It's a very real problem (Score:4, Interesting)
The $25-PC Raspberry Pi project is struggling about which codecs to include right. Even at a few bucks or less a codec, that's a huge cost when you try to hit $25 for a functional PC.
Re:No more than 100,000 people (Score:2, Interesting)
As a software developer, I'm generally happy when my 'problem' is that I have to start paying royalties because my software has hit 100k users.
I don't know about what sort of ultra-awesome-kickass programmer you are, but by the time most of us start selling 100k units/year we've got enough accountants involved that dealing with MPEG-LA licensing is a drop in the bucket.
Re:Bring it on (Score:5, Interesting)
The issue with GIF wasn't even Compuserve, but rather some patent trolls at Unisys who discovered that they held a patent on the LZW compression algorithm. Unisys went to Compuserve with a legal threat, but also a licensing scheme where they and their customers could be absolved of past and future infringement if they would submit to the license.... and they capitulated.
In fairness to Compuserve, the developers who created the GIF standard intended it to be patent and royalty free. Their mistake was picking up an ACM Journal and discovering the LZW algorithm when they were trying to establish the format, thus adopting that algorithm into the standard. Typically at the time, algorithms published in such a manner were considered "in the public domain"... or at least assumed so if you were formally publishing how they were working in an internationally recognized journal of that nature. Certainly in the article where this algorithm appeared there was no mention of intellectual property other than a copyright on the article itself. Software patents were presumed to be something you would do to something that you also protected via trade secrets and deliberately tried to keep from your competitors except for a begrudging filing with the USPTO that was usually obfuscated enough that you couldn't really figure the algorithm out from the patent application anyway.
It should be noted that IBM also seemed to have a patent on LZW algorithm, so the claim by Unisys could have been challenged from multiple fronts. The problem was that nobody wanted to take them on so it was either ignored or people capitulated to Unisys.
The largest problem with the GIF format was the issue that web browsers had adopted the GIF standard as a de facto image standard and the only universal image format across multiple operating systems and browsers that displayed images. JPEG images were starting to come into use, but there are limitations on JPEG images and it wasn't nearly so universal. Furthermore, most image editing software at the time Unisys started to demand royalty payments supported the GIF standard.... in part because it was thought to be patent and royalty free.
This stands out because it was one of the first software patents that really impacted a broad swath of software developers and pretty much hit nearly every internet user at about the same time. Few companies are willing to let such "submarine patents" languish without enforcement any more because too much money can be made through enforcement before the concept becomes essentially an international standard. Unisys got lucky after a fashion too, as they certainly didn't contribute to the development of the GIF standard.
In terms of "official standard", I guess that is in the eye of the beholder. I do believe that the W3C did archive the GIF standard and provided links to Compuseve's documents on the format as a recommended image format for web browsers before the patent issue hit the fan. It certainly was hard to avoid the use of GIFs even if you tried, and was a pretty universal image format even before web browsers were created.
Re:Or we could just fix patents and be done with i (Score:2, Interesting)
Normally this situation is pretty cut and dry. Unfortunately, the problem here is that H.264 is substantially better technology than WebM [multimedia.cx], it's not clear that WebM is not trampling on MPEG-LA patents, Google is not indemnifying WebM users against it, and Ogg Theora is noticeably worse than WebM.
I am obviously in favor of eliminating software patents, and thereby eliminating the problem of patent trolls like MPEG-LA. I am also obviously in favor of open standards. But this is not like most software where we can just round up some of the gang and knock out a new better thing in a few weeks. There are not a lot of people who can design a modern, performant, high-quality video codec, open source or no (notably, both Google and Xiph have managed to do a shitty job). To do so within our legal constraints really might be asking too much.
Since it looks like we don't get to live in the world we want to live in, we have to ask ourselves which of the worlds we can have we would like. I don't relish the idea of having inferior technology foisted on me any more than the idea of closed standards. But it's looking like we're going to have to pick one, and I don't think when it comes down to it I'd rather have the inferior technology.