Why IE9 Will Not Support Codecs Other Than H.264 436
jlp2097 writes "There is a new article up on Microsoft's IEBlog explaining why IE9 will support only the H.264 codec: 'First and most important, we think it is the best available video codec today for HTML5 for our customers. Relative to alternatives, H.264 maintains strong hardware support in PCs and mobile devices as well as a breadth of implementation in consumer electronics devices around the world, excellent video quality, scale of existing usage, availability of tools and content authoring systems, and overall industry momentum – each an important factor that contributes to our point of view. H.264 also provides the best certainty and clarity with respect to legal rights from the many companies that have patents in this area.'"
In other words.... (Score:5, Interesting)
Don't be surprised to see a spate of patent attacks on Ogg Theora... which we may or may not fund ourselves.
The patent lawyers succeeded (Score:5, Interesting)
More than anything else, I think the H.264 nonsense demonstrates the lock-down that will mark a new era of the software industry.
So, its for the DRM then... (Score:3, Interesting)
The last phrase quoted is likely the key one - Microsoft is very focused on providing as much DRM as possible, and if this codec has the most potential in that regard from their POV, thats likely why they are supporting it. I am sure the Entertainment industry has been talking to MS about this and urging them to keep pushing on DRM type solutions.
Re:The patent lawyers succeeded (Score:5, Interesting)
I wouldn't worry much. IE is becoming less and less relevant every day. For one it's loosing marketshare on the desktop, but also very importantly is the fact that mobile devices are quickly becoming the preferred medium that people use to interact with the web. I know lots of people who are doing their everyday tasks (check Facebook, email, bank balances, etc) on their phones and are barely touching their computers - if they even bother to have one. Microsoft (and with it, IE) has an absolutely dismal marketshare in that space, and they don't look to be improving.
IMHO, while IE still has a (slipping) majority, if we're talking about something that's going to be used for the next decade, I'd be FAR more concerned with what Safari, Chrome, and to a lesser extent, Firefox, plan to do than IE.
It's already paid for (Score:2, Interesting)
There are several reasons for this decision. H.264 support in Windows is already paid for (if I'm not mistaken $25 million bucks annually) and taking into account the current software patents laws in the US Microsoft doesn't want any more headache facing lawsuits having implemented support for other codecs [read Theora] which patents status isn't entirely clear and there are no powerful organizations which will protect Microsoft if some company [troll] discovers Theora is infringing their patent portfolio.
The last and probably the most important reason is that H.264 is already an unwritten standard on the Internet and this codec has an unparalleled quality and can be used for pretty much any situations (mind that *all* other existing current codecs are inferior).
There's no better choice, unfortunately. (Score:3, Interesting)
MPEG-LA's patent portfolio is sufficiently mighty that a competing video codec would have to be designed from the ground up with the specific design goal of avoiding infringement in order to escape it's shadow. This has not been done with Theora or any other codec that I'm aware of.
Combine this with the fact that MPEG-LA's licensing terms have been sufficiently reasonable that you can get $100-300 gizmos with hardware decoders built in, there's little reason why for anyone to oppose it on practical rather than philosophical grounds.
Re:In other words.... (Score:4, Interesting)
We may find many reasons to "hate microsoft" but I seriously doubt Microsoft will actually assert charges of patent infringement against anyone... ever. Microsoft's involvement in the software patent arms race was quite reluctant and I suspect that is still the case. Microsoft was first bitten by the software patent trend by the people who held the patent on "double-space" back in the day. There were a lot of people who were quite tickled and delighted to see the giant attacked for this. I was among them. I wasn't then able to see down the road to the hell of software patents that we are seeing today. Had we, the IT community at large, sought to limit and even deny software patents from the beginning, we might have less trouble than we have today.
In any case, we might suspect Microsoft of funding attacks against open source technologies, I doubt Microsoft will ever directly assert software patents themselves.
In my mind, in fact, I see Microsoft joining in the fight against software patents. It is as big a pain in their ass as it is for many others... probably bigger because they have a rather big ass.
Who cares? Theora isn't competitive. (Score:4, Interesting)
I just can't get interested in debating this stuff until Google open-sources VP8. Theora is a non-starter. It doesn't perform well and the marketplace already rejected it in enough places (i.e. virtually all portable devices) that it will never be a true competitor.
Once Google open-sources VP8 and makes it free (gratis and libre) then we'll have a real horse race. I'd love to see VP8 hardware support fast-tracked for all devices (mobile and otherwise) so we can have a competitive free solution for video.
Re:Someone explain this to me. (Score:3, Interesting)
Because Microsoft made a deal with MPEG-LA, that's why. MPEG-LA makes money off patent licensing.
Re:The patent lawyers succeeded (Score:4, Interesting)
And it's not just video codecs. Want to create an application that stores, processes, or transmits credit card data? You had better have about $50k in cash ready ready to pay for PA-DSS or PCI-DSS Level I certification. And that is the starting point. The documentation process pretty much means that Opensource, by its very nature, will never be PA-DSS certified. We're in the process of taking an opensource project we forked and getting it PA-DSS certified. Small development team of 4 people and it is a nightmare. While we ship the source code on every install CD the development process itself is pretty much restricted to a BSD-like invite only approach.
90% of PA-DSS is documentation. A lot of that documentation revolves around your development process including interviews with the developers to make sure that things like code reviews are indeed implemented and that requires at least 2 developers since the person who writes the code can't review the code, technical support cases are documented, if any cardholder data is used for troubleshooting, it is properly and securely deleted, etc..
And I see more of this coming down the line in the name of "data security". While it won't kill Opensource, it is going to make it pretty damn hard for a weekend hacker to create something.
Now don't get me wrong, after a year of dealing with PCI-DSS and six months of PA-DSS, I fully understand why their standards are the way they are and for the most part it's mostly a good thing. However the fact that it takes at least $10k (and as much as hundreds of thousands of dollars for PA-DSS) pretty much means that the project has to be maintained by a company that is making money off the product in some way.
BTW, the only Opensource project I know of that is PA-DSS certified is Magento. And ONLY the Enterprise edition ($8995). The Community Edition is NOT.
Re:HTML5 will be a screw job. (Score:4, Interesting)
Alright, answer me two questions : HTML5 is really the flash killer, yes? Isn't an open replacement for Flash an improvement over flash? I'd assume that HTML5's openness will help avoid Flash's spammyness, right? In particular, all the pop-up ads that circumvent the "Block Pop-Ups" button are using Flash now, so they'll all go away right?
I'm not sure that HTML5 will beat the Flash plus FlashBlocker combo, but that's not realistic for most users, and variations on NoScript could accomplish the same ends.
Re:It's already paid for (Score:3, Interesting)
Microsoft doesn't want any more headache facing lawsuits having implemented support for other codecs [read Theora] which patents status isn't entirely clear
Theora is patented by On2, a Google company. These patents are licensed permissively to the public.
and there are no powerful organizations which will protect Microsoft if some company [troll] discovers Theora is infringing their patent portfolio.
What organization will protect Microsoft, Apple, and other MPEG-LA members if some NPE not in MPEG-LA discovers that H.264 infringes?
Define "better". (Score:0, Interesting)
h264 is heavily patent encumbered and *anyone* that uses it is subject to its licensing terms (which for consumers, includes NO commercial use, if you take a movie of aliens coming to earth with your h264 camera and it makes a billion dollars on youtube, they will come after you and take it all).
Theora is significantly better in that there are no patents that apply to it. Companies like Apple and Microsoft like to spread FUD about their being some theoretical patents that apply, but not a single one of them has EVER come up with even a single example, because they know those patents would be quickly dissected and invalidated if they did. Opposition to Theora is firmly entrenched in the realm of fear uncertainty and doubt.
Re:H.264 (Score:2, Interesting)
Microsoft has an opportunity here to support a codec that free browsers-- such as Firefox-- may not be able to support, given the codec's licensing restriction.
If YouTube never works with free browsers, the proprietary browser makers all get a major boost.
Army of MPEG-LA bots? (Score:1, Interesting)
Funny. Every time this debate comes up, I see this huge stream of either "but H.264 is oh-so-much-better than Theora" (which doesn't matter: HTML5 standard dictating Theora as baseline wouldn't force anyone in using it!) or "but MPEG-LA has patented everything-and-your-dog", which is most probably FUD.
I can't get rid of the impression that MPEG-LA (or some of its members) have hired a spin-clinic.
Re:HTML5 will be a screw job. (Score:4, Interesting)
Professional web dev here. I first heard about HTML 5 a year or two ago, in the context of their adding a bunch of new elements (<nav>, <header>, <sidebar> and so forth) and removing all the presentation markup.
Overall, HTML 5 is great. There are a few things from XHTML 1.1 which aren't going to be present which would be nice, but I can't name them offhand. The <video> tag was, to me, just a nice convenience. The war that's erupted over this is, IMO, kind of ridiculous; everyone should obviously support both if they can and Theora if they can't, unless legal issues materialize. And I think that's 100% FUD; the Xiph guys are meticulous about legality since it is basically the reason they exist. If anyone litigates Xiph, Xiph will win.
More than that, the <canvas> tag is a big deal. I hope all of CSS 3 gets implemented too. Things are looking pretty good overall. I think this video hysteria will probably blow over, and Theora will be widely available, if not installed by default, available as a plugin.
Re:HTML5 will be a screw job. (Score:3, Interesting)
The BBC were doing just fine, even with iPlayer - I was using XBMC to watch beautiful HD content until they switched on swf verification on their streams. If they disabled this, or just offered up h.264 without the flash wrapper I would be happy again.
Re:HTML5 will be a screw job. (Score:3, Interesting)
You are right, but at the same time you are completely ignoring the elephant in the room. Microsoft is putting HTML5 and *only* h.264 into IE9. This means that as HTML5 gets rolled out, it *will*have*patent*problems* for anyone who wants to do 'Free' video and doesn't want to convince their users to download a different browser.
Meta-rants aside, do you see the problems coming down the road? This is the topic of the article, after all.
Re:HTML5 will be a screw job. (Score:3, Interesting)
This is a very Apple-esque move on Microsoft's part.
Sure, it makes sense for them to favor h264 over anything else. There is really no good reason for them to pretend that other formats do not exist.
H*LL there could be legacy video files that people don't want to transcode. This isn't just about open systems zealots. Forcing one codec can be a nuissance in a number of ways.
Re:HTML5 will be a screw job. (Score:3, Interesting)
Your external video player can scale videos anyway you want, or not at all. I'd also suggest that a one click "open with" dialog is a nicer interface than a web browser that starts a video as soon as you hit the page. Why is launching another app such a big deal?
An external video is still problematic, unless we can agree on a universal format that every external external player on every platform can handle.
There is one thing I'll say. It would be nice if there were a one click way to pass an URL to an external video player for streaming, instead of "open with" which downloads the file first and passes that file to the video player. This could easily be done with no changes to the protocol.
Yep, agree on this point.
Software should do one thing, and do it well.
It depends on how you defined "one thing". Wouldn't you define "displaying web-based content" as "one thing"? Shouldn't then, by your definition, everything be handled by the browswer? Why not launch images in a separate viewer? It completely depends on whether you interpret video as part of the web content, or something external and separate. Why the arbitrary distinction between still and moving images?
Re:HTML5 will be a screw job. (Score:3, Interesting)
See Chris Dibona's explanation [whatwg.org].
Re:HTML5 will be a screw job. (Score:3, Interesting)
The open source browsers will either have to get on board with h.264 or risk losing the majority of their users, and we'll see a return to the bad old days of IE dominating the web.
I think you missed my point. While I agree with your premise above I disagree with your conclusion. Yes the other browsers, including the open ones like Firefox and Chrome, will have to support h.264. But they'll also support the other free alternatives. New startups tend to operate on a shoe string budget until they get popular enough for funding. So they'll tend towards using free codecs. If one of these start to get popular even more people will be switching off IE. It's not like the bad old days where MS had the power to attempt to fragment the web into MS only or everyone else. MS doesn't control h.264 and can't keep it secret and change it at a whim so no one else can support it. That's what they attempted originally with IE. In this case any browser is free to implement the only standard MS is going to support but they can support other unencumbered standards that are much cheaper to use. It will just take one web site that uses ogg to get popular and IE's numbers start dropping faster. Once IE's numbers drop low enough Google can switch youtube to ogg and MS will have to adapt or die. Either way I only see this as a plus in that IE's numbers will drop.