Follow Up on HTML5 Video in IE9

發行項
05/03/2010

Our recent post generated many comments and questions. The discussion of intellectual property rights is complex and invites many different points of view. This is a good opportunity to talk through the certainty and uncertainty relative to our goals for IE9 from Microsoft’s point of view.

Developers have consistently conveyed that they want certainty and predictability in the underlying browser platform. We want to deliver a great HTML5 experience in IE9 with great certainty. The goal of certainty informs a lot of choices, such as which of the many standards still under construction we’ll pursue. Browser developers have to make decisions like this all the time.

For many reasons, H.264 video offers a more certain path than other video formats and does so in a way that delivers a great HTML5 experience for developers and end-users. 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. The rights for implementations of the H.264 standard (see this Wikipedia article about the standardization process) are managed by MPEG-LA as part of a program that has been in place for many years. This long-standing licensing program for a codec that is in broad usage today in the industry provides a stable system from which we can support our customers. As experts will note, there is never complete certainty in an area like this one.

Some comments asked for examples to support the statement in the previous post about “The rights to other codecs are often less clear, as has been described in the press.” One comment linked to a Streaming Media article; other examples are easy to find.

Intellectual property is a complex topic. As it’s not an engineering topic and this is an engineering blog, the remarks here are by definition limited. On the topic of whether one person’s codec does or doesn’t use someone else’s intellectual property, the only opinion that ultimately matters is a court’s. Many people seem to assume that availability of source code under an open source license implies that there are no additional costs, or that the code has properly secured necessary intellectual property rights from all rightful owners. Our experience and the experience of others indicate otherwise, and the web standards groups have discussed this issue as well. For other codecs, it’s not clear today how the rights will be determined for commercial scenarios and what the costs will be. By virtue of existing commercial use in a wide variety of products implemented by a large number of companies, H.264 minimizes uncertainty for consumers and developers.

Several comments speculated about Microsoft’s financial interest in the codec. (Microsoft participates in MPEG-LA with many other companies.) Microsoft pays into MPEG-LA about twice as much as it receives back for rights to H.264. Much of what Microsoft pays in royalties is so that people who buy Windows (on a new PC from an OEM or as a packaged product) can just play H.264 video or DVD movies. Microsoft receives back from MPEG-LA less than half the amount for the patent rights that it contributes because there are many other companies that provide the licensed functionality in content and products that sell in high volume. Microsoft pledged its patent rights to this neutral organization in order to make its rights broadly available under clear terms, not because it thought this might be a good revenue stream. We do not foresee this patent pool ever producing a material revenue stream, and revenue plays no part in our decision here.

There were many questions about royalties, and a lot of speculation in the comments about licenses and payments. The majority of H.264 video content on the web today is royalty-free. MPEG has said that individuals can create video files in the H.264 format and distribute them and play them over the internet for non-commercial purposes without further obligation on licensed platforms like Windows. We are aware that this commitment is set to expire in 2016, but fully expect to commit to supporting the extension of this license and associated terms beyond that date. In general, distributing encoders or decoders or offering sophisticated pay-for-video requires a license from MPEG-LA. Third-party applications that simply make calls to the H.264 code in Windows (and which do not incorporate any H.264 code directly) are covered by Microsoft’s license of H.264.

Some comments pointed to language in our Windows EULA that comes directly from MPEG-LA and reinforces many of these terms. As with all licensing programs, there are limitations and issues, which people have pointed out. The functionality we provide is technology we license and we follow the terms of that license.

Several comments asked about Microsoft’s support for plug-ins (like Flash and Silverlight). Of course, IE9 will continue to support Flash and other plug-ins. Developers who want to use the same markup today across different browsers rely on plug-ins. Plug-ins are also important for delivering innovation and functionality ahead of the standards process; mainstream video on the web today works primarily because of plug-ins. We’re committed to plug-in support because developer choice and opportunity in authoring web pages are very important; ISVs on a platform are what make it great. We fully expect to support plug-ins (of all types, including video) along with HTML5. There were also some comments asking about our work with Adobe on Flash and this report offers a recent discussion.

We’ve read some follow up discussion about support for more than the H.264 codec in IE9’s HTML5 video tag. To be clear, users can install other codecs for use in Windows Media Player and Windows Media Center. For web browsers, developers can continue to offer plug-ins (using NPAPI or ActiveX; they are effectively equivalent in this scenario) so that webpages can play video using these codecs on Windows. For example, webpages will still be able to play VC-1 (Microsoft WMV) files in IE9. A key motivator for improving the codec support in Windows 7 was to reduce the need that end-users might have to download additional codecs. The security risks regarding downloadable codecs and associated malware are documented and significant. By building on H.264 for HTML5 video functionality, we provide a higher level of certainty regarding the security of this aspect of browsing and our web platform.

The biggest obstacle to supporting more than H.264 today is the uncertainty. When there’s industry consensus and confidence that the uncertainties are resolved, we’ll be open to considering other codecs. Until then, we’ll continue with our current plans to deliver great HTML5 video in IE9 with certainty for consumers and developers.

Dean Hachamovitch

List of articles referenced AVC/H.264 Licensors Fake codecs that drop widely spread malware Google may face legal challenges if it open-sources VP8 codec H.264 Already Won—Makes Up 66 Percent Of Web Videos H.264 licensing body won't charge royalties for HTML5, other Web streams IE Blog: HTML5 Video JPEG patent case steams forward Mac OS X malware posing as fake video codec discovered Malware Posing As Youtube Codec Media Streaming with Windows 7 Microsoft Security Intelligence Report Volume 8 Microsoft Sued Over JPEG Patent MPEG LA’s AVC License Will Continue Not to Charge Royalties for Internet Video that is Free to End Users NPAPI Open letter to Steve Jobs: Thoughts on Flash

Re: Codecs for <video> and <audio> from Silvia Pfeiffer on 2009-07-29 (public-html@w3.org from July 2009)

Comments

Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
To the idiots who think Microsoft has financial gain in h264 support: By implementing h264 in IE9, they will have to pay royalties to MPEGLA (unless IE9 is covered by Win7). Microsoft won't gain money for shipping with h264: they will infact have to pay themselves for every IE9 copy. Ogg would be suitable only until some patent troll surfaces, and sues every major tech company for implementing it. It's the equivalent of shooting yourself in the foot, as far as Microsoft is concerned. Using H264 is a financial loss, but at least a calculated one, while Ogg is like walking on a (patent) minefield. And h264 is way better quality wise as well.
Anonymous
January 01, 2003
@Matt, If you don't know where mpeg4 plays in all this, and especially if you think my link was spam, you shouldn't be commenting in this thread. It makes you look bad and sound like Dean Hachamovitch.
Anonymous
January 01, 2003
It's interesting Microsoft removed my post in response to being accused of posting a spam link yet leaves the false accusation in place.
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
The comment has been removed
Anonymous
January 01, 2003
I am making this comment from Google Chrome. 'nuff said.
Anonymous
January 01, 2003
thank you for this great info, I was looking for, really thanks
Anonymous
January 01, 2003
No one uses IE, only non internet saavy people do, and they are dying out quicker then you think...
Anonymous
May 02, 2010
I hoped that you'll at least support wmv/vc-1 and hope that we'll be able to choose which codecs we'll use in IE9. So far the idea of using H.264 is great - I think that this and other HTML 5 extensions'll change the web as we know it. However, you should not drop plugins yet, as these have some important features, that are not covered in HTML. We hope to see more inventions and samples with the next build of IE9.
Anonymous
May 02, 2010
The least you could do is name the major competing codec and what it's pros and cons are. You just come off as legally afraid to do anything else. Well at least I hope that's it and not that you are supporting this codec to make it hard for new multi-platform independent browsers to start up.
Anonymous
May 02, 2010
This post does not make it clear why VC-1 was not added as well. It is also a standard with industry momentum that has a lot of hardware support and has near identical licensing as h.264. It also has a lot of advantages in using a lot less computing power to decode video essentially making it a greener codec than h.264 and much easier for mobile devices batteries.
Anonymous
May 02, 2010
"At the end of the day, we’re building a browser for the Windows customer. Listening to that customer, in whatever form that takes, is not just important, it defines what we’re here to do." LIARS. WINDOWS CUSTOMERS WANT XP SUPPORT FOR IE9 WHICH YOU ARE NOT DOING. DID YOU NOT READ THOSE HUNDREDS OF COMMENTS ASKING FOR XP SUPPORT WHEN IE9 WAS UNVEILED?
Anonymous
May 02, 2010
The reason is enforced format lock-in, a widely used method with Microsoft products. It has nothing to do with developer convenience, especially for open source based development which this decision will have a direct negative effect, due to the complications related to H.264 licensing. And, considering convenience; it is the opposite of what you would call the situation when your video content is in a different format than the one enforced by the browser. Plugin-based video content is, and this is obvious, not the convenient solution. Except for these reasons, it shows MSIE unwillingness to cooperate with open standards, as H.264 should not be the defacto standard, which it would be, if MSIE exclusively supports only this format. It should only be one of the options.
Anonymous
May 02, 2010
The comment has been removed
Anonymous
May 02, 2010
The comment has been removed
Anonymous
May 02, 2010
Why not support Theora (.ogv)? It is an open-source video format whose codec can be included in a browser or any other programs at no costs. This is clearly stated in its website (http://www.theora.org/benefits/). In my opinion, Microsoft should support a variety of multimedia formats so that developers will not have to spend more time making things work. Should Microsoft limits its support for new web standards, Internet Explorer will lose market share, as we have seen, to other browsers as a consequence.
Anonymous
May 02, 2010
The comment has been removed
Anonymous
May 02, 2010
The comment has been removed
Anonymous
May 02, 2010
Great post, can't wait for IE9!
Anonymous
May 02, 2010
H.264 may be represented by 60% of web video offers, but right now it is only represented by 25% of the market share of web video capable installed browsers. Whereas ogg-theora is represented by 75% of web video capable installed browsers. The two H.264 proponent browsers are Safari and IE9. In order to fill the ogg-theora/H.264 gap, both Safari and IE9 would need to grow by over 25% in browser market share as opposed to IE6/7/8/Firefox. This is unlikely to happen within the next 18 months as Safari doesn't grow that fast, and Microsoft has shown that is incapable of replacing older versions of its browser in a timely manner. Therefore the particular decision of IE9 NOT to support ogg-theora makes life especially much more difficult and uncertain for both consumers and developers. The decision will delay and hinder the arrival of web-video for another 2 to 3 years, because for users and developers there's no single sure way of supporting video in 100% of web video capable browsers regardless of other risks and uncertainties either H.264 or ogg-theora represents and regardless of their technical merits or drawbacks.
Anonymous
May 02, 2010
The real question is: I agree that theora is not clear as of its patents, but if google publishes a new format, with the right licence to use it, would you implement it right away ? I mean, would we have to wait till IE 10. Is your decision set in stone for ie9 or are you allowing it to evolve ? I'd appreciate a clear answer, such as "whatever happens, we will stick to h264 ONLY", or "if a decent format with correct licence is made public, we will consider it". Thanks in advance
Anonymous
May 02, 2010
again, wich audio codecs will ie9 support. video without an audio-stream is a little bit boring. will ie9 support the audio-element?
Anonymous
May 02, 2010
First, thanks for this new post. This explain clearly your choice and your position and answer a lot of questions. I only have the impression that you are taking a shortcut to implement html5 video tag and you do not give any options for the future. Maybe that's not your team goal but the focus should be on fixing the html5 specs to define a clear, royalty-free codec. MPEG LA objectives are commercials = to make as much money as possible : http://www.mpegla.com/main/Pages/PoolBenefits.aspx So implementing this codec only in your next browser and believing that everything will be ok in the future is naive to say the least. I think that your aiming only for the passing grade instead of giving a thoughtful and viable solution for the longterm.
Anonymous
May 02, 2010
The comment has been removed
Anonymous
May 02, 2010
Apple seems to have the same Problems: http://hugoroy.eu/jobs-os.php A browser without codecs is bad, a browser without fast codecs is bad but open source-browsers that uses codecs that are based on illegal copyright violating sources sould be banned. - Well, maybe I just don't like the Mozilla HackFox browser. Even if we have to run wmp, it would be better than no support at all. Just integrate the UAC for that pupose in an easy way, and we're all much happier.
Anonymous
May 02, 2010
Wait, I am a bit confused. We can only run the other codecs when using a plug-in? Why can't the video tag also play any other codec that is installed on the user's computer, and have IE9 specifically make sure that H.264 comes bundled? Security? If that is the case why allow it in WMP/WMC? Don't get me wrong, I think H.264 is a better format than OGG, and I do understand why other browsers support just OGG. I am in the camp of someone that wants it to just work. Right now I will output both formats for my users and would love to be able to cut that down in the future. I just don't understand why the video tag can't be made to work with everything the user has. And can we please get a comment on the audio tag? Will it be supported? What format will that support? If it's something beyond mp3, wave, or ogg I would really like to know as I have code I would need to update.
Anonymous
May 02, 2010
Video aside, why isn't IE9's focus Silverlight? Isn't HTML5 just playing into the hands of Google?
Anonymous
May 02, 2010
Dear Microsoft, I want to hear you mention Ogg Theora. And I want you to talk about On2 VP8 which rumors are Google will unveil in a couple of weeks and free up licence free as HTML5 video codec. You, Microsoft, if you want to sound like you are playing nice to the FOSS crowd, simply need to state that you WOULD be interested in supporting On2 VP8, and that you would even be interested in supporting Google in proving to the industry that that specific open source and free codec would in fact be usable and not be in conflict with the Mpeg LA patents. You need to state that you would find it good for the industry that there were to be an open source and free video codec which also performs well compared to Mpeg LA patented codecs. Especially if adding Ogg Theora 2.0 support in IE9 only requires few megabytes of codec code, then you should state that you will not have a problem supporting that.
Anonymous
May 02, 2010
@tz. Yeah, MS is evil. And the fact that you can post that in a MS blog, no less (but not anywhere in an Apple one, for that matter) is very telling indeed...
Anonymous
May 02, 2010
Current H.264 patent litigation http://bit.ly/9vNGZW
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
I wonder how many of the "MS is evil, Theora is good" comments come from people who actually read what this blog says.
Anonymous
May 03, 2010
Steve Jobs mailed: "A patent pool is being assembled to go after Theora and other "open source" codecs now." http://hugoroy.eu/jobs-os.php
Anonymous
May 03, 2010
Why not support both H.264 and Ogg Vorbis?
Anonymous
May 03, 2010
Rob, your (spam?) link says nothing about h264.
Anonymous
May 03, 2010
This is great news. Very logical roadmap. Thanks.
Anonymous
May 03, 2010
Florian, you've clearly conveyed that you don't know anything about "antitrust and price-fixing laws" so why would anyone believe anything you say? Jadd, please be polite and actually read before commenting. Thanks.
Anonymous
May 03, 2010
"We’re committed to plug-in support because developer choice and opportunity in authoring web pages are very important; ISVs on a platform are what make it great." You guys need to have a talk with the Windows Phone 7 team about developer choice, since there isn't any in WP7. That aside, I like the direction IE9 is going. H.264 is the best choice IMHO if you were to choose only a single codec for HTML5 video. I'm glad to see you are open to other codecs, should other browsers make a different choice. I'm hoping they don't because I think a single codec and a single standard for this will make video support a lot easier for web developers in the long run. Has the IE team communicated with Mozilla/Apple/Opera about using H.264 exclusively? If there's buy-in from the other browser vendors, I think a lot of fears will be eased.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
Ruthsarian, you're plainly confused. MPEG-LA is a patent pool, no "assembly" required.
Anonymous
May 03, 2010
http://lists.xiph.org/pipermail/theora/2010-April/003769.html There you go. Straight from Xiph. The statements from MPEG-LA about all video codecs being patent encumbered are simply not true. It's just a bunch of FUD to keep big corporations (Hello Microsoft) from finding suitable free alternatives and MPEG-LA losing out on another revenue stream. Again. Microsoft, you need to educate yourselves before you make such important changes in what technology you implement. Your rash decisions affect millions of people.
Anonymous
May 03, 2010
ROFL... yeah, xiph is a credible source. Unfortunately, the legal system isn't really based on asking the defendent "Did you break the law?" and acquitting them if they say "no."
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
"This post does not make it clear why VC-1 was not added as well." VC-1, despite its origins, is a codec also governed by the MPEG-LA. Its licensing is pretty much identical to that of H.264, so supporting VC-1 in IE9 doesn't really buy anybody anything. Microsoft also (ironically) pays MPEG-LA for the right to include VC-1 support in Windows. "MY WEBWARE SHOULD WORK THE SAME ACROSS ALL BROWSERS THAT ADHERE TO STANDARDS" Then you should lobby W3C to actually explicitly add Ogg Vorbis/Theora support to HTML5 as it's currently not a part of the standard. Why get mad at Microsoft for not implementing something that's not a part of the standard? Microsoft is 100% following the HTML5 spec. "it shows MSIE unwillingness to cooperate with open standards" Which open standards? Ogg? Theora? These formats and codecs are open SOURCE, but they're not standards. In order for something to be considered a standard it must be ratified by an industry group or standards committee like SMPTE or ISO.
Anonymous
May 03, 2010
As a web developer who watched Real destroy their market lock through greed, it is disappointing to see patent trolls continue to hobble innovation in web technology. Dean, the moment you make the statement that you're an engineer, not a lawyer, your blog post becomes irrelevant. Any argument about the performance or legality of one codec against others must be backed with evidence, and Microsoft's track record on supposed IP violations in Linux makes it virtually impossible to take statements of this ilk as anything other than FUD. If you plan to speak for your corporation, understand that you have debts to pay. FWIW I think HTML5 is for the most part a terrible idea. On the surface it comes across as a convenience for developers, but in practice it cannot be implemented fully by a browser vendor without licensing; I suspect it won't be long before the geolocation feature is hit by a submarine patent. To an end user, this smacks of oligopolistic practices which have no place in a document renderer. You'd be entitled to call this comment paranoia, but remember your employer had to settle out of court with Eolas and will probably have to do the same with i4i.
Anonymous
May 03, 2010
Quote: "For web browsers, developers can continue to offer plug-ins (using NPAPI or ActiveX; they are effectively equivalent in this scenario) so that webpages can play video using these codecs on Windows." What does this mean? Does it mean that someone can make a plugin which allow the video tag to play any kind of Video which the plugin can decode? Or will we again end up with a situation where we have to make special html to explorer 9, to trigger the correct plugin even if the user have it installed?
Anonymous
May 03, 2010
EXPLOITATION. IE is and has always been exploitation of the web for none but Microsoft's gain. This is not changing. The more I see things like this in IE, the less I see progress, because with every one step forward, there are two steps back. There's a reason that everybody else (Except Apple) supports Ogg Theora/Vorbis for the video tag. Mozilla supports it because it's patent free. Mozilla can't afford to license H.264. Opera refuses to pay a license for H.264. Google supports it because they "aren't evil". Microsoft doesn't support it and Apple doesn't support it. WHY? Because they are members of the H.264 patent pool. They have royalties to gain. This is completely inexcusable. Without actual standards, the web is fragmented. With only support of H.264, when the HTML5 video tag reaches primetime, IE 9 becomes another IE 6. Microsoft, quit the browser market, or actually work with an existing rendering engine. Obviously there isn't enough competence at Microsoft to realize that an open web is more important than lock-in or profit.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
tim: I think you need to understand what a "corporation" is legally required to do (answer: generate profit for shareholders).
Anonymous
May 03, 2010
The following statement is instructive: ..."H.264 minimizes uncertainty for consumers and developers." IOW: The reason to choose H.264 in IE, exclusive to truly free codecs, is pure and simple FUD. This is why IE is loosing market share hand over fist versus others. MS is not listening to their technical base that wants free and open standards, and we are choosing to not take what is being force-fed to us any more. If you do not support Ogg/Theora -- you loose, because we just say, "Oh, so, you can't play our movie??? -- Hmm, well, here... here's a better, faster, more secure alternative browser for you. It's crunchy, you'll love it!" -- and, they will.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
Microsoft are yet to present any form of web browser to the public which follows any single w3c specification correctly. Achieve that first and then try to support draft specifications. Users are already voting with their feet on IE and it's time Microsoft had a serious re-think about the browser because by the time html 5 is final, IE will be irrelevant as a direct result of not supporting web standards correctly. Now that Microsoft's position is clear and IE's market share is rapidly falling, developers can start to exclude IE from accessing content it is incapable of supporting. Internet explorer needs the web. The web does not need Internet explorer. There is a very deep hatred for IE amongst the professional web development community and developers are now so sick of needing to price support for IE separately, that many developers have started to add features turned off or missing when IE is used to visit the site. Microsoft needs to try to win back the trust of web developers by supporting standards correctly. This announcement goes nowhere towards doing so. Way to go to provide your customers with a slow, irrelevant and featureless web experience. Not a single usable version of Internet explorer released to date and yet still no lessons learned. Every other browser software distributor tries to provide as much choice as they can for their users, while Microsoft is insistent on focusing on what choices it's users and developers should not have. Wake up and start putting your customers first. The web is about standards and choice. Not about what your abysmal web browsing software will not do.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
Why dont you do what micrsosoft allways does best. Give the community an api that we can use to add theora to ie 9 html 5 video tag. Then we could choice of a free webformat or a more closed one. The security issue could always be solved in one way or another ( by digitally signing or only allow codec download from certain vendor site). Best Regards Anders
Anonymous
May 03, 2010
If Microsoft implemented OGG and then got sued would you foot the legal bills? Would you hire the people who got fired or had their careers ended for making that decision. Of course there are no lawsuits against ogg...The guys with deep pockets don't use it. If there are patents that can be used against it then they won't be shown until there is someone with deep pockets shipping ogg. The patent guys don't care about preventing you from watching videos, they care about taking money from Apple. Working for a large technology company (not Microsoft or Apple) I can tell you that many decisions are made about technology on a legal basis for legal certainty. No one with hundreds of millions to lose really wants to see if ogg is unencumbered.
Anonymous
May 03, 2010
Seriously, you’ll point to blog comments made within the same week you made your post to back up your argument about rights issues?
Anonymous
May 03, 2010
You talk about "certainty and predictability", from my point of view as a developer, that would be if each browser uses the same codec. Right now Apple and Microsoft chose H.264, but I'm not sure if Google, Opera and Mozilla will do the same. If they don't, then there is no "certainty and predictability", because then we have to encode in multiple formats to support each browser. I'm ok with H.264, whatever works you know, it's just that I'm disappointed that in of you (all browser-vendors) for not being able to agree on one standard.
Anonymous
May 03, 2010
@Will Why should anybody believe what you say? Now I'll explain to you the term "monopoly". Monopoly is when one company or conglomerate controls one particular market by having their product being the predominant one. Monopolies happen, luckily H.264 is not (yet) a monopoly in Browsers, however it pretty much is the predominant product in the recording/editing/distribution sector. Now I'll explain to you anti trust. This happens when a company that does have a Monopoly (like MPEG-LA already has it on recording/editing/distribution for for video codecs) missuses their strong position to shut out competition of that market. You might ask, well, what's bad about a market without competition, and I leave that to you as an enlightening exercise (hint, you may use wikipedia).
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
@Leo (not Davidson) >Members of the working group really need to stop >making statements and using language that is >clearly chosen in an attempt to influence a draft >specification that is still years from candidacy. That's all a part of being in the working group. You get to influence the specification. The working group is not just some magic impartial council that carves standards into stone tablets without regard for what its members want. Microsoft, along with the other members, is invited to comment on how the standards should be molded.
Anonymous
May 03, 2010
@Ruthsarian: You are out of date with your claim that H.264 fees could start as soon as January 2011. Earlier this year MPEG LA announced that there will be no royalties on internet video that is fre to end-users until Jan 1st 2016, and this post by Microsoft suggests that it may well continue past that date too. For the first time in my life I feel sorry for Microsoft. The voice that seems to be largely missing from these debates is that of all the different companies that provide video services on the web.
Anonymous
May 03, 2010
Serious question: what would Microsoft do if Google decided to encode all YouTube video in a different codec (not H.264)?
Anonymous
May 03, 2010
I'm disappointed that Theora will not be implemented in IE9. This seems like a step backwards. Once we hit year 2014 a lot a bloggers and sites will not be able to finance the large royalty fees set out by the MPEG-LA. The optimal solution would be to support Theora and H264. At least this would allow everyone to use HTML5 video.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
A lot of people are glossing over what might end being more important: The lack of a Canvas tag in MS' IE9 Implementation.
Anonymous
May 03, 2010
Theora is better because it's open. EOD
Anonymous
May 03, 2010
Audio - obviously going to be AAC (AAC and h.264 go together in the .mp4 container) h.264 - clear on legal end (and MS is already paying for it, why not use it?), excellent compression/quality, hardware acceleration, large mobile device support, large support in terms of software editors/encoders/players, large supply in content VC-1 - similar/same as h.264, except not quite as much much mobile device support and very little supply in content ogg - somewhat murky legal end (some person who helped develop could jump in with claims and cause a mess), compression not as good as h.264, little to no hardware acceleration, very little mobile device support, smaller support in software/encoders, very little content uses it This could all change in the FUTURE, but this is how things are NOW
Anonymous
May 03, 2010
@ravewulf: As mentioned previously, IE9's AUDIO element will support MP3 and AAC.
Anonymous
May 03, 2010
@Mike I doubt Google will go with something that doesn't have broad support. They want as many people as possible to be able to use their products, regardless of browser or OS. As W3C adds more codecs in the future, browsers and content will adapt to it. @EricLaw [MSFT] Thanks for clarifying (I hadn't checked, just went with what most use with h.264 video) :)
Anonymous
May 03, 2010
@ravewulf: Regarding theora murky legals, I dont think it even needs to be someone who helped develop it, its about patents that cover any of the concepts used to make the compression & format work. For there to be a totally safe codec from a legal point of view, someone would need to come up with a method of compression that was completely new, not covered by any of the existing patents out there, then they would need to be recognised as the owner of this intellectual property, and then declare that they were giving it to the world for free forever. Either that or wait for all the existing aptents that cover things like H.264 to expire, which will be a very long time away.
Anonymous
May 03, 2010
@ravewulf "ogg - somewhat murky legal end (some person who helped develop could jump in with claims and cause a mess)" It's sad that people have to resort to lies. But shows also fear. The question then is why? Why does this CODEC (ogg theora) that according to them is so bad, has so much legal uncertainty, has so little support causes them so much fear that they have to lie to justify their words and actions?
Anonymous
May 03, 2010
@Jack If you want to believe that legal uncertainty surrounding Theora does not exist then thats your choice, but please dont call people liars on this issue unless you would like to personally cover us all for any legal hassles if it does turn out that any of theoras methods are covered by someone elses patent. It may well be true that there are other reasons why some would be interested in spreading FUD about Theora, but that does not mean that patent woes for Theora are impossible. I have a feeling we will find out before too long.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
@Steve Watkins "If you want to believe that legal uncertainty surrounding Theora does not exist then thats your choice, but please dont call people liars on this issue unless you would like to personally cover us all for any legal hassles if it does turn out that any of theoras methods are covered by someone elses patent." You're seeing things upside down! It's you that have to provide PROOF that Theora infringes. Until then I will do the right thing. Call liars LIARS! If people fear patent suits for using Theora, then they will also fear patents suits for using H264. Both are the same position.
Anonymous
May 03, 2010
@EricLaw [MSFT] Why not support the Vorbis format with <audio> tag? Nobody has been arguing against it, it just seems to get lost in the hubbub about Theora and <video>... And Microsoft ALREADY USES Vorbis in the Xbox consoles, so they obviously feel it a good codec. As for <video>, supposedly, Dirac is going to be VC-2. It's not confirmed, but we'll see. As for H.264, nobody seems to notice that the "free" license for H.264 and AAC only apply if you are making ZERO revenue. This means, no advertisements, no "donations", no monetary "subscriptions", etc. Because of this, it is legally infeasible to use H.264/AAC for most sites aiming to use <video> that don't already pay for the H.264/AAC license. Rather, I'd use either Dirac/FLAC, Dirac/Vorbis, Theora/FLAC, or Theora/Vorbis. Most likely Theora/Vorbis since that has the most support in the HTML 5 browser space, with Chrome, Firefox, and Opera supporting it. I realize that Theora is terrible for HD quality video, which I believe Dirac excels at. So technically both bases should be covered. I'm also aware that Firefox's HTML5 backend does include support code for Dirac, it just is disabled currently. Obviously, Chrome's HTML5 backend supports it, since it uses ffmpeg, which supports a smörgåsbord of codecs. Opera can have Dirac plugged into it since it uses GStreamer for its HTML 5 engine. So, Microsoft, when will you be implementing Ogg support for HTML 5? ;)
Anonymous
May 03, 2010
My vote is for VP8. Understanding that this is not yet "open", if it is opened, it would seem to make a lot of lives easier and just make sense. I wish MS would consider this option but I'm not optimistic.
Anonymous
May 03, 2010
A free and open format (that doesn't require you to pay a license fee for development) would be far better than H.264. Ogg is a very good option that is well supported in many modern operating systems. This is a real chance for MSFT to show they do want to participate, instead of EEE.
Anonymous
May 03, 2010
When Microsoft if making a decision, they should stand on a Microsoft side. From Microsoft point of view, whenever they add a codec into IE, it will aldo be included in Windows. Microsoft must be made sure that they do not need to pay layalty later on for the codec. Otherwise, considering the number of Windows/IE copy on the market, it will be very difficult for Microsoft to manage. This is also the reason that Microsoft did not include Codec for DVD play in Windows XP.
Anonymous
May 03, 2010
JoeJaz, any announcement from Google that they're making VP8 "open" is worthless unless they provide developers who implement it a warranty of indemnification (as Microsoft does for their major products) saying that Google's legal team will defend implementors from lawsuits, and Google's assets will be at risk, not the implementors. It would be a huge win for the internet, but I'm not holding my breath. Instead, expect to see Google make a meaningless announcement that gets lots of press but ultimately means little.
Anonymous
May 03, 2010
Hey Microsoft, are you so proud that you can't even be bothered to mention Vorbis or Theora by name?
Anonymous
May 03, 2010
To the people who are suggesting all the obscure formats, VC-1, Dirac, VP8, Ogg etc etc: H.264 can be decoded in hardware by most portable devices, including the Zune, iPod and many (most) video-capable cellphones. The future is mobile.
Anonymous
May 03, 2010
@Paul Eccles Those formats are not obscure. Ogg Vorbis, for example is used in games quite a lot. VC-1 is used in Blu-ray Discs. (I'll give you Dirac). There is no logical reason to artificially limit IE9. Microsoft may as well go back to IE6.
Anonymous
May 03, 2010
"# re: Follow Up on HTML5 Video in IE9 Monday, May 03, 2010 8:10 AM by Jones111 Steve Jobs mailed: "A patent pool is being assembled to go after Theora and other "open source" codecs now." http://hugoroy.eu/jobs-os.php" Interestingly enough, the patent troll/pool Jobs was talking about has just filed... against Apple! Not Ogg (Which Mozilla, Google and Opera all contain.) H.264 appears to be risky. Looks like no codec is safe.
Anonymous
May 03, 2010
Matt: Microsoft provides indemnification for h.264? Or does MPEG-LA? Neither does? I thought so. They can't indemnify against patents, just like Xiph.org or Google can't: If the patent holder denies a license, that's their right. And with patents there's no need to be "reasonable and non-discrimatory". If in 2 weeks ACME Corp. comes up with a patent that h.264 necessarily infringes on and doesn't want to license it to Microsoft at all, IE9 will ship without h.264. And there's nothing that MPEG-LA or Microsoft could do about it. For real-world examples of this, look at any tech trade fair: Devices are taken from display by customs because of MP3, not for Vorbis. It doesn't matter if those devices and their vendors have proper licensing with Thomson/mp3licensing.com: Sisvel prefers to stay independent with their patent. How is h.264 safer from that than any other codec?
Anonymous
May 03, 2010
@Paul Eccles: "H.264 can be decoded in hardware" Nice try, still wrong: H.264 can be decoded by codec functions in hardware that are driven by firmware. h.264 is too complex to be handled in hardware completely. The accelerated functions will be generic enough to be adaptable to other codecs as well, and those codec functions usually support other codecs besides h.264, such as VC-1, MPEG-4, RealVideo. Support for Theora would be a matter of a firmware update. Also, those h.264 decoders usually only provide baseline support. To be able to decode more advanced h.264 profiles, the devices have to fall back to software-only again, until the firmware is similarily enhanced. When comparing h.264 with Theora for mobile devices with similarily optimized codecs, h.264 will suffer from its higher decoder complexity (yay for committee based designs that needlessly bloat the feature set), so should Theora take off and be supported by the firmware, it will actually be a better choice.
Anonymous
May 03, 2010
What this article says: "We're considering the VIDEO tag to mean ONLY H.264" You might as well come out and implement a <H264> tag yourself. I mean, why listen to other people?
Anonymous
May 03, 2010
What if <IMG> only supported JPEG? Can you see the issues that might arise from that?
Anonymous
May 03, 2010
This was posted earlier in this thread, but it deserves repeating: http://lists.xiph.org/pipermail/theora/2010-April/003769.html Video codecs are NOT any more of a patent minefield than any other area of software development, it is just that codecs produced through industry groups like MPEG-LA are overly encumbered with trivial patents due to the distorted fashion in which they are produced. Do you see Microsoft refusing to implement SVG because of a 'submarine patent risk'? Or Canvas? Or CSS3? No, but implementing these is no less or more risky than implementing a royalty-free video codec such as Ogg Theora. The difference is that Microsoft is invested in the MPEG standards group and while it may not currently profit from the patents it holds for H.264, it still helps to prevent competition from non-members, as they have to pay royalties and cannot cross-licence patents like the members can.
Anonymous
May 03, 2010
I think it's gonna really be great.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
@Bored: You're so boring... :P
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
@PG: "h.264 is too complex to be handled in hardware completely." That's incorrect. Video cards which support DXVA decoding for H.264 (that's most Nvidia and AMD cards) support full GPU decoding of the compressed bitstream. It's called Profile D or VLD Profile, look it up: http://download.microsoft.com/download/5/f/c/5fc4ec5c-bd8c-4624-8034-319c1bab7671/DXVA_H264.pdf The codecs currently supporting DXVA decoding in Windows are MPEG-2, H.264 and VC-1. The support is a combination of decoder hooks, driver APIs and GPU hardware support.
Anonymous
May 03, 2010
Here's a suggestion: Anybody referring to Ogg as a "codec" should henceforth have their comments and opinions entirely disregarded. If you don't know the difference between a file format and a codec, you're probably not qualified to contribute to this discussion so please stop trolling.
Anonymous
May 03, 2010
Okay, let's say Microsoft implements Theora and Vorbis support, then it becomes clear that one of these or both codecs infringe patents in the MPEG-LA patent pools: Would it matter for Microsoft? I don't think so, because MS has these patents already licensed, unlike Mozilla. So the only risk for MS would be if Theora or Vorbis infringe a patent MS has not licensed, wouldn't it?
Anonymous
May 03, 2010
Microsoft is listening. Excellent steps.
Anonymous
May 03, 2010
« When there’s industry consensus and confidence that the uncertainties are resolved, we’ll be open to considering other codecs. » So we have to await the final battle against Theora/Vorbis? Is this what Steve Jobs was talking about in his email?
Anonymous
May 03, 2010
I was pleased to note that this blog entry links to a Wikipedia article. It serves to underscore the relevance and value that Wikipedia has for a great many people. It is, therefore, extremely disheartening to know that Internet Explorer 9 will be incapable of native support for all of Wikipedia's content. Mozilla, Opera Software and Google support both Ogg Theora and Ogg Vorbis in their browser software. Activision Blizzard uses Ogg Vorbis audio in World of Warcraft. Bungie's Halo, published by Microsoft Game Studios, uses Ogg Vorbis audio on Windows and OS X. It is clear that in these cases the risk of using open media formats was found to be acceptable. Internet Explorer 9 has the opportunity to make a significant contribution to the open web through the support of open media formats. I encourage Microsoft to take that opportunity.
Anonymous
May 03, 2010
I am estatic that Microsoft is adopting HTML5 and h.264. This is truly great news for consumers! The move to open standards is much appeciated. My only requests are that you provide a version of ie9 with no activex support compiled in to improve security and that you fully support HTML5's canvas feature.
Anonymous
May 03, 2010
@IEteam, i would like to ask something to you, guys, this has been said lot of times here, but i would like to put again on the table this issue: Please consider the possibility of a radical migration of IE engine. Choose Webkit, choose Gecko , what you like, but please do something to boost standards support in the short term. Google did it ( Chrome Frame ), why don't you? I.e: "Virtually all of Chrome's April expansion came at the expense of Microsoft Corp.'s Internet Explorer, which dropped 0.7 percentage points to finish the month at 59.95%, the first time that IE has fallen under the 60% mark." () Thanks for listening. () http://www.computerworld.com/s/article/9176240/Chrome_again_beats_Firefox_in_browser_gain_race (retrieved 2010/05/03 )
Anonymous
May 03, 2010
A wise decision. This will have the effect of further marginalizing Internet Explorer, and give you guys more of an opportunity to finally lay this beast to rest. I also recommend supporting only AAC audio, and only the WMV container, and possibly finding other ways to shoot yourselves in the foot with views that are polar opposite to Mozilla and just as extremist.
Anonymous
May 03, 2010
ombarg sez "Six out of ten browser users use IE. You're clearly not successful since you've only got 50% more marketshare than all of your competitors combined (including those that run on platforms you don't). So you should adopt the rendering engine used by your weaker competitors." Errr. yeah. That makes sense.
Anonymous
May 03, 2010
"And really, now that we see multiple large companies with experienced legal teams and non-trivial exposure committed to shipping Theora I think we're kidding ourselves when we attempt to analyze this as a legal issue. It's not. It's a business/political decision. The market is now going to battle it out. Enjoy the show." http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2009-May/020015.html
Anonymous
May 03, 2010
Well this makes it easy then: we can just encode all our videos in Ogg Theora, and blame Microsoft (as always) whey they don't work in IE. Nobody would be surprised: IE hasn't supported anything resembling an up-to-date standard in years, so why would anyone expect IE9 to? (Note: I'm not saying Ogg Theora is more modern or up-to-date than H.264, just that not many would be surprised if IE didn't support something most other browsers do.)
Anonymous
May 03, 2010
I don't see how the choice of H.264 would be anticompetitive against Mozilla, as someone claimed. According to the post: "Third-party applications that simply make calls to the H.264 code in Windows (and which do not incorporate any H.264 code directly) are covered by Microsoft’s license of H.264" so there is no reason why Firefox would have to pay to implement H.264 on Windows (which is the only platform where IE competes against it). As for the whole patents/fees issue: discussing what might happen in 5+ years in this industry is bound to be inaccurate and/or irrelevant. For the time being, H.264 seems the most reasonable choice. And if it turned out to be wrong, there is plenty of time to fix it.
Anonymous
May 03, 2010
Mac: You should read the discussion in https://bugzilla.mozilla.org/show_bug.cgi?id=435339
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
I applaud Microsoft for following the ISO standard for consumer audio video. H.264 is the successor to the DVD. When you argue against H.264, that is like arguing against the DVD. Both DVD and H.264 enable consumers to choose any device from any manufacturer and still see video content. It enables publishers to publish one video product and be playable on any device from any manufacturer. This is good for everybody involved. There is no need to support any codec other than H.264 in the video tag. If you're not using ISO standard video, you don't need W3C standard markup. If you are going to be nonstandard, you are choosing to be seen by just a small fraction of the Web. In that case, you might as well use a plug-in.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
@Jordan: Perhaps they are not as short sided as you? There are more codecs than h.264 and Theora. Do they need to list every single codec that exists each post make the whiners happy? What they are are talking about applies not only to Theora but to VP8 (which is even listed in the title of one of their links) and others.
Anonymous
May 03, 2010
I don't see why Microsoft should follow through MPEGLA's lead. Clearly, Microsoft hasn't learnt a lesson after the MP3 licensing ambush.
Anonymous
May 03, 2010
I understand and agree with Microsoft's arguments that H.264 is a great codec and that its legal story is more predictable than that of other codecs. However, I also understand the arguments of others who worry mostly about the fact that they may eventually be required to pay royalties for H.264. How about going one step further, Microsoft? Use your leverage as a leading software company and the fact that you are one of the licensors of MPEG LA to lead an effort that attempts to make MPEG's H.264 baseline profile royalty-free, similar to what JPEG did with its own baseline profile. Notice that this would need to be done only to the baseline profile, and more advanced profiles could continue to be royalty-bearing. This way, we can have the best of both worlds, and other browser makers would have no reasons not to support H.264.
Anonymous
May 03, 2010
Dean, you say "Third-party applications that simply make calls to the H.264 code in Windows (and which do not incorporate any H.264 code directly) are covered by Microsoft’s license of H.264." ... and then "As with all licensing programs, there are limitations and issues, which people have pointed out." This is quite subtle. So all third party apps are "covered" in the sense that Microsoft's license applies to them, but they must still respect the limits of that license, right? In particular, third party apps that wish to use the codec for "commercial use" (whatever that is; a lot of developers will need lawyers, I guess) are actually not "covered" in the normal sense of the word. In light of that, I still think your declaration that "Third-party applications that simply make calls to the H.264 code in Windows ... are covered by Microsoft’s license of H.264" is potentially very misleading. A large number of apps that do just that could find themselves paying royalties or on the receiving end of a lawsuit from the MPEG-LA.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
<blockquote>MPEG-LA could solve the trust issue by legally promising to continue to only charge reasonable prices for reasonable things forever, instead of only for a few years.</blockquote> MPEG-LA does not determine the licenses. That is up to the licensees. And those have already committed to charge reasonable prices as participant in the standardization proces of h.264 or as a participant of the blu ray association. Also the patent pool will get smaller over the next 5 years as some patent overrun the 20 years making the share for other patent holder bigger even without a price rise. So your suggested doom scenario of a steep price rise is not on the cards. The parties involved in the patent pool have every reason to keep prices reasonable.
Anonymous
May 03, 2010
I'm confused about these points: "For web browsers, developers can continue to offer plug-ins (using NPAPI or ActiveX; they are effectively equivalent in this scenario) so that webpages can play video using these codecs on Windows. For example, webpages will still be able to play VC-1 (Microsoft WMV) files in IE9."

You say that developers can offer plug-ins using NPAPI, but your hyperlink for NPAPI goes to a Wikipedia article that states that IE no longer supports NPAPI plug-ins!
Does this mean that a plug-in will be required in order to play VC-1? Will the plug-in be provided with IE and installed by default, or will it be a separate download?

Anonymous
May 03, 2010
@Alex: "GPU hardware support." What does this GPU hardware support consist of? Most likely (from looking at implementations of the codec acceleration with GPUs on other platforms) by pushing some shader program into the GPU. So while this program is designed for the special purpose features of a GPU, these features are still relatively generic. Case in point: AMD released a new video driver that provides support for additional h.264 profiles for some of their chips. Were GPU video decoding "entirely in hardware", this wouldn't be possible. So yes, this is software, Theora support can be added (on the OMAP platform they do this already), and Theora on GPU will likely require less resources than h.264 on GPU.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
@hAl, come on, let's not have a stupid argument of semantics. Pretend I said "the members of MPEG-LA" rather than "MPEG-LA" if you must. If you are so certain that all members have all agreed to be reasonable until all their patents expire, where is that in writing, please? If they are all really going to do that and really mean it then it will be no problem for them to make it legally binding, instead of the current vague promises not to charge anyone for certain uses for a few years, after which all bets are off. If it is in writing somewhere then my idea was already done before I suggested it and all is well, but AFAIK there is no such legal promise.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
@Leo Davidson From the MPEG-LA FAQ "Q: What is the Term of the AVC Patent Portfolio License? A: The initial term runs through December 31, 2010. The License is renewable for successive five year periods for the useful life of any Portfolio patent on reasonable terms and conditions. Q: Is there a limitation on the amount that royalty rates may increase at each renewal? A: If royalty rates were to increase, they will not increase by more than 10% at each renewal for specific license grants (except Internet Broadcast AVC Video, which is provided for in Section 3.1.5 of the License)." So yes, patent licensors in the mepg-la patent pool have committed to reasonable terms and conditions and to limited lisense prices on renewals. The scenarios of steep price rises on renewals are just made up stories by open source fans who feel the need to spread FUD on the h.264 licensing. There is no basis to support such stories as that would be against the interests of patent pools which are mainly created to support broad support for a technology with one stop licensing.
Anonymous
May 03, 2010
To Leo Davidson: Really? Is the Windows codec system so bad? But it's Microsofts own system, so it should be easy to run the decoder in its own process. Then you can simply restart it if it crashes. (Extra bonus if you resume the video where it crashed). But I must admit I have never seen a windows user complain about codecs crashing. But even if it is that bad, that's not an argument because anyone can still make a 3 party codec which decode video in IE9 they just can't use the video tag to do it. So we once again have to make special html to IE9 -( What will most likely happend is that someone make a plugin, which use the windows codecs to decode video, and then we end up with this situation anyway.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 03, 2010
But the problem with IE9's codec plugin support, is that it require non standard ie9 only html to use. And why no vorbis audio support for <audio> Do you fear patents for that too(And if you do, remember to inform Microsofts game section, which does use vorbis).
Anonymous
May 03, 2010
@hAl, Thank you for the reply. If that is accurate & binding, and there are no hidden caveats, then it seems reasonable and was (more or less) what I was suggesting. @Martin, It's not really the Windows codec system that's at fault. The problem is that people writing codecs/demuxers tend to do a poor job of checking for unexpected inputs and produce code that crashes (or has security flaws or whatever) when faced with them. You can move all your video rendering out-of-process and sandbox it for a bit more security (assuming the codecs are well-written enough to run in the sandbox; and even then sandboxes usually only protect from writes, not reads (data theft)). Even then, though, you'd still be asking your users to put up with unpredictable video performance and stability. If videos randomly crash or do not work in a web browser then it's better than taking out the entire web browser -- for sure -- but it's not exactly a great user experience. We don't really need a hundred different video codecs anyway. A handful of good codes with known-good quality implementations is much better than a free-for-all. I mean, there are a million and one different image formats but the web gets by pretty well with just PNG, JPEG, GIF and SVG. I'm not saying there should only be one codec (though it'd be nice if everyone could agree on one which they all support) but there's no need for ten, either.
Anonymous
May 03, 2010
@hAl, Another thought: Given that the rate for non-commercial use is currently 0, according to that FAQ it can never be non-zero. :)
Anonymous
May 03, 2010
[quote]It's not really the Windows codec system that's at fault. The problem is that people writing codecs/demuxers tend to do a poor job of checking for unexpected inputs and produce code that crashes (or has security flaws or whatever) when faced with them.[/quote] What? If you do the decoding in its own process, that process should not have any read/write access to any other user data. And remember the alternative to the possibility of bad codecs are a blank screen, because the browser can't decode the video at all. I know what i prefer. I assume that Microsoft have written their own H.264 windows codec, so we know that just using the codec system will at least work for MPG4. So as i see it, the alternatives for non H.264 are between no video, and possible bad video. But I really don't think that there are that many problems with the existing windows codecs. And I agree we don't need support for any insane format(Flv anyone?) but the problem with the current solution is that it's H.264 only and there is no way for third parties to fix that. Using windows codecs would atleast allow users the possibility to view the videos at wikipedia.org and other places which use other formats.
Anonymous
May 03, 2010
"(except Internet Broadcast AVC Video, which is provided for in Section 3.1.5 of the License)."). This is the problematic section. So i think they plan to ask money for all H.264 streams 1 jan 2016.
Anonymous
May 03, 2010
The comment has been removed
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
@martin No that is not a problematic section because the licensing fee for Internet Broadcast AVC Video is zero till 2016 and after that will be maximum the fee of free TV broadcasting which is currently 10,000 dollars per broadcaster(site). I guess people spreadig the FUD forgot to mention that little detail... From the MPEG-LA license summery I have (which is not updated to 2016 yet)

"For (b) (2) where remuneration is from other sources, in the case of free television (television broadcasting which is sent by an over-the-air, satellite and/or cable Transmission, and which is not paid for by an End User), the licensee (broadcaster which is identified as providing free television AVC video) may pay (beginning January 1, 2006) according to one of two royalty options: (i) a one-time payment of $2,500 per AVC transmission encoder (applies to each AVC encoder which is used by or on behalf of a Licensee in transmitting AVC video to the End User) or (ii) annual fee per Broadcast Market12 starting at $2,500 per calendar year per Broadcast Markets of at least 100,000 but no more than 499,999 television households, $5,000 per calendar year per Broadcast Market which includes at least 500,000 but no more than 999,999 television households, and $10,000 per calendar year per Broadcast Market which includes 1,000,000 or more television households.13 In the case of Internet broadcast (AVC video that is delivered via the Worldwide Internet to an end user for which the End User does not pay remuneration for the right to receive or view, i.e., neither title-by-title nor subscription), there will be no royalty during the first term of the License (ending December 31, 2010), and after the first term the royalty shall be no more than the economic equivalent of royalties payable during the same time for free television."

Anonymous
May 04, 2010
In addition to the above license text video broadcast of less that 12 minuten are licensed for free anyways. So as I can see it a internet video broadcaster that plays h.264 video longer than 12 minuten and has a marketshare of more than 1 million people (I asume that would be unique visitors to the video) would pay

nothing untill 2016 and then a maximum of 10,000 dollars. A site with video less than 12 minuten would pay:
nothing A site with videos that have less than 100,000 visitors would pay:
nothing.

Anonymous
May 04, 2010
And I am inclined to expect that MPEG-LA leaves the fee at nothing even after 2016 anyways as it hardly worth it to collect those fees from the few sites that regularly have long video's that get more than a million viewers. What is at least evident that there is a lot of FUD being spread about the h.264 licensing fees that could have been avoided with some simple surfing tto the MPEG-LA website and doing some minor fact checking.
Anonymous
May 04, 2010
Some links for the readers AVC/h.264 licensing FAQ: http://www.mpegla.com/main/programs/AVC/Pages/FAQ.aspx AVC/h.264 license summary: http://www.mpegla.com/main/programs/AVC/Documents/AVC_TermsSummary.pdf
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
To all those saying Microsoft is ignoring standards by choosing not to support Theora and Vorbis:

HTML5 specification doesn't actually require either of those codecs. To be fair, it doesn't require H.264/AAC support either.
HTML5 specification is still in draft stages which means no browser's HTML5 implementation is "more" complete than others because the specification isn't done yet.
Neither Theora nor Vorbis are standards because they haven't been ratified by any industry or standards organization (such as SMTPE or ISO). Open source != standard So to suggest that Microsoft is somehow ignoring web standards for video is completely ridiculous. How can they ignore something that doesn't even exist?!? One could just as easily make the argument then that Firefox and Chrome are ignoring standards too!

Anonymous
May 04, 2010
@Charlie: Semantics? I talk about the difference between necessarily having to buy a new device for any other codec support, or installing a driver update should that come out. That is, the difference between a possible update and an impossible one. The other aspect is that with the GPGPU APIs it should be feasible for third parties to provide GPU-accelerated video decoding. Probably not as efficient as the vendors (that can natively access device functions), but still an improvement. As written, that happens on OMAP. Indeed the question is, if it will be done. Simple answer: It will be supported when there's a business case. Should Theora become common on the Web, it will be supported. Microsoft (and others) seems to be very interested in making sure that this won't ever happen. For "Vorbis and Theora aren't standards", a standard is a standard once it's set in stone by some governing body that declares it stable. For Vorbis and Theora, this standard body is Xiph.org - simple as that. Following your argument that only SMTPE and ISO are supposed to declare standards, HTML should be universally ignored (That's W3C, not ISO),SMTP should be ignored (that's IETF, not ISO). ISO gave us (in cooperation with ITU-T) OSI networking. We're on IP now, surely something must have gone wrong? I think I'll stick to standards that are actually out there (Theora for nearly 6 years by now)
Anonymous
May 04, 2010
@hAl The 12 minute limit you mention only applies to video services where the user pays for individual videos, so ad-supported services such as YouTube would not benefit from this. Your figures around the royalties for internet broadcasters do not correspond to the language in the summary. You talk about 'unique viewers', but the summary makes no mention of actual viewers, only to the size of the 'Broadcast Markets' in which people COULD receive the content. This map: http://www.dishuser.org/TVMarkets/ should give you some idea of the number of 'Broadcast Markets' in the USA alone. Now, imagine you post a video on the internet, people COULD watch it in every one of those areas (and the rest of the world!), and theoretically the 'economic equivalent' royalties promised by the summary could be millions upon millions of dollars. Interestingly, the licence summary has a footnote that indicates the royalty equivalence between internet broadcasters and free television is expanded on in section 3.1.5 of the full AVC Patent Portfolio License, but for some reason they don't actually make the full license available online: http://www.mpegla.com/main/programs/AVC/Pages/Agreement.aspx Quite apart from broadcaster royalties, there is also the issues of royalties for distributing decoders. Browser vendors such as Mozilla and Opera would have to pay several million dollars per year to distribute versions of their browsers that included a h.264 decoder, or drop support for HTML5 video. This would be a tremendous strain on their finances, particularly for a non-profit foundation like Mozilla. It would effectively mean only rich companies could make web browsers that could browse the entire web and show HTML5 video.
Anonymous
May 04, 2010
@PG: You're interpreting my example of ISO and SMPTE literally. I listed them as examples of industry and standards organization, but the list certainly isn't limited to them. 3GPP, IETF, ITU, DECE... These are all valid examples. But Xiph.org isn't, I'm sorry to say. Xiph is a non-profit corporation, not a governing body or an organization of industry professionals. If you follow that logic, than RealVideo is also a standard. Let's repeat that again for everyone to hear: Vorbis and Theora are not standards.
Anonymous
May 04, 2010
<<<Browser vendors such as Mozilla and Opera would have to pay several million dollars per year to distribute versions of their browsers that included a h.264 decoder>>> And where exactly did you pull this per-year figure from? (FWIW, Mozilla likely spends more per year on Firefox than Microsoft and Apple spend on IE and Safari-- they can afford to, because Google plays them almost $100M/year).
Anonymous
May 04, 2010
@Amtiskaw: "Browser vendors such as Mozilla and Opera would have to pay several million dollars per year to distribute versions of their browsers that included a h.264 decoder, or drop support for HTML5 video." To be fair, this is not an issue on Windows 7 and Vista because access to the native H.264 decoder is open to any DirectShow or Media Foundation based application. It's exactly how players such as PowerDVD work, so there's no reason why Firefox or Chrome couldn't take advantage of the same decoder.
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
@FactChecker I 'pulled it' it from the blog of Mike Shaver, chief evangelist for the Mozilla Corporation: of http://shaver.off.net/diary/2010/01/23/html5-video-and-codecs/ To quote: "Mozilla has decided differently, in part because there is no apparent means for us to license H.264 under terms that would cover other users of our technology, such as Linux distributors, or people in affiliated projects like Wikimedia or the Participatory Culture Foundation. Even if we were to pay the $5,000,000 annual licensing cost for H.264, and we were to not care about the spectre of license fees for internet distribution of encoded content, or about content and tool creators, downstream projects would be no better off."
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
@Charlie Sure, and I believe they considered doing just that, but decided against it because such a strategy would undermine the cross-platform nature of the Firefox browser, and the web itself. As FactChecker pointed out, Mozilla certainly have the cash to license H.264, and probably any other codec they like, but they're making a stand based on the principles open web. The open web is about more than just following standards, it's also about open participation. The great strength of the web has always been that anyone can make a website or develop a browser. Adopting H.264 for web video undermines that principle, so Mozilla have taken a stand against it. Not because it's easier, or because it's cheaper, but because it's the right thing to do.
Anonymous
May 04, 2010
This test in 2010 show that to achieve similar quality a Theora file would have to be twice the size of a h.264 file. http://keyj.s2000.ws/?p=356
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
@Charlie: Did you consider PNG a standard at the time it was implemented by IE 4.0 and 32-bit Netscape 4.04? That was when it was merely described in a RFC, years before being standardised by the ISO. @Amtiskaw: The end of 2012 is also when all patents relevant to MP3 decoding should be expired, then Firefox might implement native MP3 support for the audio tag so we have at least some sort of common denominator for native implementations.
Anonymous
May 04, 2010
The comment has been removed
Anonymous
May 04, 2010
@Amtiskaw In a broadcast the video is send to all people in an area. On a site a html video is only send on a pageview to one person. So looking at unique visitors to a webpage containing a html video is a good measure for the number of people a video was broadcasted to.
Anonymous
May 04, 2010
@hAl That's certainly a reasonable view, but the point is it's YOUR opinion, not what's actually written in the licensing summary. The summary itself is ambiguous. If they've nothing to hide, then MPEG-LA should publish exactly how internet broadcasting royalties would be calculated, so we could have a properly informed discussion.
Anonymous
May 04, 2010
When a company starts choosing to avoid risks and sits on its own success, well I think it's a sign of decadence. I believe a company like Microsoft should look a little bit further and pursue not only its own interests (money) but also oversee -whenever possible- the interests of the entire community. Imho copyright as it's today is only about a way to make money regardles it is right or not. Please rethink your decision and add native support for the Ogg Theora technology in IE9. Kind regards, druss
Anonymous
May 05, 2010
"When a company starts choosing to avoid risks and sits on its own success, well I think it's a sign of decadence." They're paying for it as IE's market share continues it's inevitable downward slide. I believe we've passed MS' tipping point.
Anonymous
May 05, 2010
The comment has been removed
Anonymous
May 05, 2010
Dan, it's both foolish and impolite to ask questions when their answers are contained in the original posting.
Anonymous
May 05, 2010
@Charlie: The difference between RealVideo and Theora is that there's a specification (versioned, stable for 6 years, open to implement for third parties) for Theora. There isn't such a thing for RealVideo. Were Real to release a specification with such characteristics, RealVideo would be standardized, yes. The only question for standards is if they're properly managed: That is, clear specifications, stable release engineering, open access for third parties. Xiph.org has shown to be able to provide that. If someone wants to provide their own video standard, they're free to do so. And prove over time that they're doing a good job managing a standard. But that's probably a matter of belief, similar to the difference between "big government" (ISO and their ilk) and "individual reponsibility" (lots of standards by whoever cares enough to manage them). Given the track record of ISO, I'll stick to the Xiph.org model.
Anonymous
May 05, 2010
Here's all anyone needs to know about MPEG-LA: they write one thing and say another.
Anonymous
May 05, 2010
@PG Theora is not a standard an it takes nearly twice the bitrate to deliver the same quality video as the h.264 which is an actual standard. Worthless as an option for internet video.
Anonymous
May 05, 2010
I think one big thing that people would like to hear is using plugins, can you extend what the video tag supports? The post says you can use plugins to use codecs, but nothing mentions extending what the video tag supports. So could there be an outright yes or no answer for this? (This should help keep the theora people quiet knowing that IE isn't limited to just h264 if you can extend the video tag.)
Anonymous
May 05, 2010
@Noname Microsoft themself will not make such a plugin but anybody else could.
Anonymous
May 05, 2010
I know that, but the whole point of my post was to have at least a semi-official yes or no on the video tag can be extended to support additional video types. With who writes the plugin, I don't care, in the end I know that support in a lot of cases means "if you can do it, good for you, but don't come crying to us if it breaks." All I care about is whether you can do it.
Anonymous
May 05, 2010
The comment has been removed
Anonymous
May 05, 2010
FYI, here's an excellent overview of H.264 licensing: http://www.engadget.com/2010/05/04/know-your-rights-h-264-patent-licensing-and-you/
Anonymous
May 05, 2010
The comment has been removed
Anonymous
May 05, 2010
The comment has been removed
Anonymous
May 05, 2010
Every so called advantage of H.264 mentioned would be equally true of a true open and royalty free industry standard. Clinging to these proprietary formats harms the public. They are covered by patents of dubious quality forcing people to purchase licenses directly or indirectly as a form of tax on everything. All evidence that these proprietary formats are a greater benefit than a harm to the public is purely subjective. People believe or disbelieve these things as they suit their agendas. As an engineer, I do not believe that any patent on a file format is legitimate. Encoding video into a file is not patentable in and of itself since the need for it makes the innovation obvious. And just what is innovative by a compression technique? These have been known for decades. Nothing new there either.
Anonymous
May 05, 2010
<blockquote>Every so called advantage of H.264 mentioned would be equally true of a true open and royalty free industry standard</blockquote> Then name us which true open and royalty free standard:
Is an ISO/IEC standard
Combines the same quality/compression ratio
Is a current Blu ray standard
Is a codec used for digital TV broadcasting
Has hardware support in hundreds of millions of devices
Is widely supported in video en entertainment software Possibly if a codec provides all of the above advantage and also is open and royalty free there could be reason for complaints about the descision to use h.264 as the codec for the video tag in IE9.
Anonymous
May 05, 2010
@Matt: "Dan, it's both foolish and impolite to ask questions when their answers are contained in the original posting." It doesn't say why Microsoft have suddenly decided to worry about codec patents in IE when every other program of theirs uses DirectShow and works with whatever codec the user wishes to install. It doesn't say why Microsoft have decided that using user-installed codecs to access video on the Internet can be dangerous in IE when Media Player does exactly that already. It doesn't say why Microsoft have decided to block everything that isn't H.264 on quality grounds when they're happy to have Media Center play anything. The answers are not contained in the original psot.
Anonymous
May 05, 2010
@Charlie, Some people are angry with the W3C for not mandating a codec. However, if you read up on the issue you'll find that the W3C wanted to specify a codec in the standard but the participants could not agree on one. Apple refused to support Theora. Mozilla refused to support H.264. Google were happy to support either or both. Microsoft, meanwhile (AFAIK anyway), had not weighed in on the issue until now. So it's not that (different groups of) people have not been angry with the other players; it's just that the anger with Microsoft is fresher. The W3C decided that there was no point specifying things in a standard which they knew implementers would actively ignore. I'm not sure that was the right thing to do but I can see their reasoning. A big goal of HTML5 seems to be to create a spec which really is (eventually) implemented in its entirety, unlike HTML4. For better or worse, standards don't actually mean much when they conflict with business (etc.) interests and the W3C has no power to dictate anything to anyone. Leaving the codec out of the spec. punted the responsibility for choosing a common codec over to the browser makers. Those browser makers have failed to agree on a common codec. Every player (except maybe Google) has attracted some anger from some people over the matter.
Anonymous
May 05, 2010
nobodyinparticular sez <<<Encoding video into a file is not patentable in and of itself since the need for it makes the innovation obvious>>> Your remark makes your complete lack of understanding of patent law obvious. If you're going to play fake-lawyer, you need to do at least a little bit of research.
Anonymous
May 06, 2010
@hAl: Whether Theora is a standard depends on the definition of a standard. It definitely fits mine (stable, properly specified, accessible to third parties). For quality, the specification doesn't say anything about quality. See the improvements between the libtheora 1.0 and 1.1 implementations, and those that happened after 1.1. It's catching up, and it definitely has a need to do, but it has more room for improvement than h.264. All those improvements happen within the standard. Your list of devices that use h.264 (Bluray, digital broadcasting and all those devices) is actually an example of the issues h.264 faces: Thanks to the complexity of "design by committee", where everyone tried to get their own pet feature in, to improve leverage in the licensing business, h.264 is segmented into various profiles. So you have a valid h.264 data stream on Bluray but can't play that stream on the iPhone (and most other portable devices) as only the baseline profile is supported on the latter. digital broadcasting uses another set of h.264 features, as does the encoding in studio workflows. For the web, this means that only h.264 baseline will be universally used (to support the growing segment of mobile internet). So any quality comparison between codecs in the context of web-video must be with h.264 baseline to be meaningful. When you get your hands on a h.264 stream you'll have to look first if the used features are all supported by your codec (ever had those spurious defects in mpeg4 video? Those might have come from different support for specified features). Theora will just work.
Anonymous
May 06, 2010
The comment has been removed
Anonymous
May 06, 2010
@Dan: Your original question of "Or would it be something to do with Microsoft using IE as a tool to help cash in in the future on the H.264 patents that they own?" WAS in fact addressed in the original post. It's been pointed out that Microsoft pays more to MPEG-LA to license H.264 than it actually gets back from it in royalties. BTW, anybody who thinks that codec royalties amount to some significant amount of cash - you're mistaken. We're talking about pennies per unit. For a company the size of Microsoft the amount of money it gets from codec royalties is a drop in the bucket - certainly not enough to be worth changing entire IE strategies for. @PG: "Theora will just work." Really? A mobile phone will decode 1080p Theora video without a hitch? I sincerely doubt that. Your point about H.264 profiles and their fragmented support is valid, but you're missing the real point: not all playback devices have the same processing power so profiles were defined in order to create conformance guidelines. In other words, profiles are not what make H.264 support fragmented. Even if you took the profiles out of the spec, you'd still need some way to say "this device will play H.264 video but only up to 720x576 25 fps progressive with 4:2:0 subsampling because it doesn't have the CPU power to handle anything more". If all decoders could handle 1080p 60 fps 4:4:4 video, there'd be no need for profiles.
Anonymous
May 06, 2010
@Alex: This is true, until the licence is changed in 2015 and Microsoft can decide how that licence is changed as they form a part of MPEG-LA. I'm sure they're willing to invest in corning the market by making something free for the consumer before reaping the rewards later on, it's not as if it's the first time it's happened. The decisions taken now seem to point to that end, if it were otherwise they would use DirectShow like any other piece of Windows software to allow the user the freedom to view whatever website they want.
Anonymous
May 06, 2010
What is the top speed at which browsers can run SunSpider? I guess the answer is going to be ZERO SECONDS. I just ran this test on my 2 yr laptop on Chrome 5.0.396 dev channel build and 10/26 of the tests are actually running in 0 seconds (See the Chrome reults here at http://tinyurl.com/25eeswh) Come-on IE, catch up. It should not be hard to do all the work in no time ... !!!
Anonymous
May 09, 2010
The comment has been removed
Anonymous
May 09, 2010
@Ryan Smyth: « but that does not mean that they have been legally vetted against patent infringement » Theora is basically just an improved version of a former proprietary codecs older than 10 years. The original was very likely carefully avoiding any patents and so did the contributers to the open source project. The chances for patent infringement are low, the chances for infringement of patents Microsoft has not already licensed are even lower. Really, how can anybody believe that patents are the issue here. Google had no problem including Theora/Vorbis support, why does Microsoft.
Anonymous
May 09, 2010
<<<The original was very likely carefully avoiding any patents and so did the contributers to the open source project.>>> Your overwhelming naivete would be quaint were it not so dangerous. Please leave the lawyering to those who have a legal education.
Anonymous
May 12, 2010
The comment has been removed

共用方式為

Follow Up on HTML5 Video in IE9

Comments

其他資源