Some of the titles of articles about WebM have been hilarious. CNNTech has an article titled ‘Patent cloud looms over Google Web video plan‘ by Stephen Shankland. While the title is wildly inaccurate, the article is actually well balanced. But it does miss a couple of points. First, WebM is at present a software solution, and in over 90% of the world you cannot patent software. The only place the article might be topical is the United States – note that I said might. The Supreme Court of the United States is due to hand down a ruling in ‘Bilski” which could possibly wipe out software patents in the United States as well. If so, the so-called patent cloud becomes clean air. Of course the group, MPEG.LA, which is claiming that there are patent issues with WebM (previously known as VP8), has acted as a patent troll over H.264, which is why that codec hasn’t been more widely adopted. Like Microsoft and The SCO Group, they are prone to extravagant claims, and a total lack of proof. Also the company which originally developed VP8, the basis of WebM, has never been sued for patent infringement, so it appears likely that MPEG.LA is out of luck.
One of the real issues with WebM is a lack of hardware acceleration, which H.264 has on some chip sets. Intel has stated that it’s considering implementing hardware acceleration for WebM, which means that everyone else will follow. This will help in adoption of WebM, and it is possible that we’ll see manufacturers dumping H.264 hardware acceleration as no longer required.
And there is rising support for it. Mozilla wants WebM included in HTML5 and it is likely that Opera will back this. Microsoft probably won’t like it, and it’s certain that Apple won’t, Apple is a huge backer of H.264, and Apple President Steve Jobs appears to be unimpressed with WebM. The problem for both Microsoft and Apple is that both of their web browsers allow for plug ins, like Adobe Flash, so a WebM plugin for both browsers is quite possible. So even if Apple and Microsoft decide not to support WebM, someone else will using their browsers. This is really a no-win situation for both. Microsoft for years has dreamed of being the gatekeeper for video and audio on the web, collecting a small fee for each video or song played. Apple to a certain extent has been a gate keeper with the ITunes application, making billions of sales. A change in video codec could hurt both companies.
Videolan plans to support WebM with it’s 1.1 release, which is currently in beta. Viewcast is adding support to it’s Niagara video streaming box. Support from others is expected shortly. Apple and Microsoft will be late in announcing support, if they decide to do so at all.
The upper part of this article was written in June. Due to various things, I’ve let it sit, and in many ways that was a good thing, because a lot has happened, and some of it is really important.
First, there is a report that the WebM codec ‘outperforms’ H.264. The report, and the response to it misses an important point. A new technology doesn’t have to be ‘BETTER’ than the old technology, it just has to appear to offer some advantages. Consider the switch from CP/M to Microsoft DOS that happened during the early 1980s. DOS did not offer any technical advantages of CP/M. In fact it wasn’t as good. But it had IBM backing it, and like many other people back then I though that IBM had to know what it was doing. It turned out we were wrong, but with IBM’s support, DOS ended up dominating the market.
In this case rather than IBM, it’s Google who is the big backer. It has been reported that YouTube (which is owned by Google) will be using the WebM Codec. Since YouTube is the most popular video site, that puts a lot of momentum behind WebM.
If WebM is even 90% as efficient as H.264, it will likely do well.
Second, H.264 is a proprietary standard. While proprietary standards have a long history, a proprietary standard which is in conflict with an open standard will loose unless it offers significantly more value. As mentioned above, if WebM is close to H.264 in performance, WebM will win.
Third, H.264 costs. Oh, the end user doesn’t see a separate charge on their invoice, but the manufacturer does. And when you are producing a high volume product, this can make an immense difference to your profitability. This won’t matter to Apple, who is one of the companies behind H.264, but it will matter to other manufacturers. Google has just given them a way to cut their costs significantly, and they’d be fools not to use it. After all, if you are manufacturing a competitor to the IPod, do you really want to be lining Apple’s pockets? Or Microsoft’s for that matter?
Fourth, while MSNBC (partially owned by Microsoft) thinks that a truce has been declared because MPEG.LA now says that use of the H.264 codec by websites that provide free video streaming will be free in perpetuity. This does nothing for the hardware manufacturers, who still have to pay, and have the most to gain from switching. While Microsoft and Apple won’t admit it, if the camera/camcorder manufacturers and the video device manufacturers switch, they won’t have much choice.
Why is this important? Simple. If you buy a camcorder, like my son’s Sony, and actually read the documentation, you’ll find that if you want to use the device to make money, you have to buy a separate license from MPEG.LA. Yes, I’m serious. If you attended an event, and recorded something newsworthy, legally you CANNOT SELL YOUR FOOTAGE unless you pay MPEG.LA. Admittedly the odds of them noticing, and coming after you for payment are low, but why would you want to take the chance, when a free option exists? For that matter, you paid for the camcorder, why should someone else have a right to limit what you can do with it?
The vast majority of consumers don’t know about this issue. When they are told, the typical reaction is shock.
In effect the MPEG.LA consortium could attempt to control video culture. I’m not saying that they want to, or plan to. But they could, and that would have a negative impact. Culture is supposed to belong to all of us, not to a for-profit consortium.
As a result WebM has my vote for ‘Disruptive Technology of the Year.’
Thursday September 2, 2010
Note 1: A lot of articles mention the patent issue. Patents are mostly an issue in the United States. Patent litigation is relatively rare in other countries. The United States Patent System is undergoing a variety of changes driven by recent court cases, including the infamous Bilski. It is quite possible that the MPEG.LA patents may be end up being limited, if say for example software patents are blocked. Also while MPEG.LA claims that their patents also cover WebM, they have at the present provided no proof that this is so. Since Video Codecs are essentially mathematical algorithms it is quite possible that any attempt to enforce the patents could lead to their invalidation.
Note 2: A lot of people have pointed out that there are no hardware accelerators available for the WebM Codec at present. While this is true, Intel, NVidia, AMD, and other manufacturers have stated that they will support hardware acceleration of WebM. It is not known how long it will take for new hardware which supports WebM to make it onto the market.
Note 3: The Bilski ruling has been issued, and while it did not directly address software patents, some parts of the ruling can be interpreted as not allowing them, and other parts appear to allow them. There is a lot of confusion over how Bilski will play out in the court system, because it was so vague. Patent lawyer Robert Plotkin wrote that Bilski will still allow software to be patented. It appears that this issue will end up before the courts again.