Slashdot Mirror
Nokia Claims Ogg Format is "Proprietary"
a nona maus writes "Several months ago a workgroup of the W3C decided to include Ogg/Theora+Vorbis as the recommended baseline video codec standard for HTML5, against Apple's aggressive protest. Now, Nokia seems to be seeking a reversal of that decision: they have released a position paper calling Ogg 'proprietary' and citing the importance of DRM support. Nokia has historically responded to questions about Ogg on their internet tablets with strange and inconsistent answers, along with hand waving about their legal department. This latest step is enough to really make you wonder what they are really up to."
Fully documentable nothin'! Theora and Vorbis are fully documented. If you can't figure out how to make your own implementation from the docs and/or by studying one of the many existing implementations out there, you need to turn in your geek card and just forget about developing software.
Proprietary would imply that independent implementations cannot be made or cannot be made easily without violating patents or reverse engineering or whatever. Vorbis and Theora are nothing of the sort -- they are fully open and unencumbered.
My blog
There is a plugin you can get for iTunes that lets it support ogg, but last time I tried it there were problems with it (you couldn't stream music to another copy of iTunes for instance because it would stream at the wrong rate and break up every couple of seconds, nor could you stream to an Airport Express).
I read the internet for the articles.
It's funny that you responded to an article about video with a rambling about audio. It's however hilarious that it got modded Insightful.
Nokia and Apple obviously have stakes in determining the codec that people use for video over the web. Apple is pushing H.264, which they point out is a standard, but fail to mention is also proprietary. Nokia mentions this in their position paper, but goes on to recommend H.264 anyway. Ironically, they list their #1 criteria for codec adoption to be "The specifications, and supporting documentation and code (i.e. conformance test suites, example/reference code, ...) are obtainable by everyone, for free or against a reasonable fee (ISO/IEC fees are reasonable in this sense)." You can't get a more reasonable fee than free, which is the case with Ogg. Anyhow, it's clear that Apple wants AppleTV to be a new content-delivery platform. Nokia probably has similar plans.
What I really suspect Nokia is saying in this paper are in criteria #2 and #5: "There is only a manageable risk in implementing the specification. In practice, we prefer specifications that have been developed in a collaborative manner under an IPR policy with disclsore requirements. Examples include specifications developed by the ITU-T, ISO/IEC, or the IETF." and "Compatibility with DRM. We understand that this could be a sore point in W3C, but from our viewpoint, any DRM-incompatible video related mechanism is a non-starter with the content industry (Hollywood). There is in our opinion no need to make DRM support mandatory, though."
Basically, "we think Ogg will get us sued" and "Hollywood won't use Ogg". It's a shame that Stephan Wenger (the author of this paper) has now damaged his own credibility by writing a four-page exercise in being disingenuous.
I'd like to point out that the one really successful proprietary codec, MP3, is a success because of the huge numbers of people who intially implemented the codec without a license and because it didn't support DRM, thus leading to widespread piracy, and establishing the format as the de facto standard for unencumbered audio. I would personally consider the W3C negligent if they did not choose an open (free as in beer and speech) codec.
- No-one knows if Ogg Vorbis or Ogg Theora are encumbered by patents. They were developed to be free of the main known patents, but they could still be encumbered by some submarine patent. If they're accepted as the baseline, Nokia face unknown risk if such a patent emerges after they've deployed the technology in hundreds of millions of phones. With H.261/AAC, the risks are more known because an unknown patent-holder would have sued someone by now.
- There's a lot of content available online (though not directly as part of Web standards). Nokia in concerned that the content producers will will stear clear of Ogg in favour of solutions that support DRM or at least have a known track record. Better the devil you know...
The second concern is probably rubbish, in so far as they are asking for H.264/AAC instead. DRM on these is completely orthogonal to the issue of the codec - you could easily wrap Theora in a DRM wrapper if you wanted (though why you'd want to is beyond me).The first concern though is more interesting. Basically Nokia seems to be saying that they'd rather pay predictable patent licensing fees for H.264/AAC than face unknown risk. That's a business decision, and I don't know of any good argument against it - we really don't know if there are any submarine patents that Theora or Vorbis might infringe on. From what I know about coding, it seems unlikely (especially in the case of Vorbis), but not impossible to me.
Despite this, I think W3C made the right call and should stick to it.
Apart from it not supporting DRM, ogg has only advantages - it's equal or superior to most other codecs (the widely used mp3 and wma are inferior) and it's open-source w/o patents restrictions...
Seriously, does anyone have an explanation for that?
Ogg isn't a codec. It's a container format. Vorbis is the audio codec in question, and Theora is the video codec in question.
Theora was created using proprietary code and patented techniques developed by On2 and used in their VP3 codec, adapted to fit inside an Ogg container. There are tools to let you convert existing VP3 streams into Ogg streams.
The Xiph.org foundation negotiated free access for all to those patented technology before adapting and adopting it. From the Theora FAQ:
Yes, some portions of the VP3 codec are covered by patents. However, the Xiph.org Foundation has negotiated an irrevocable free license to the VP3 codec for any purpose imaginable on behalf of the public. It is legal to use VP3 in any way you see fit (unless, of course, you're doing something illegal with it in your particular jurisdiction). You are free to download VP3, use it free of charge, implement it in a for-sale product, implement it in a free product, make changes to the source and distribute those changes, or print the source code out and wallpaper your spare room with it.
The paper from Nokia seems to revolve around the fact that it doesn't support DRM from what I can see.
-1 Uncomfortable Truth
Theora video is somewhat based on H.261 and is obsolete in regards with recent developments such as H.264 and VP8 from On2. Can someone knowledgable about Theora make any comment on this assertion?
Monty (the inventor of Vorbis) can comment on it: http://web.mit.edu/xiphmont/Public/theora/demo.html
"Unlike Vorbis and Speex, legitimate best-in-class codecs, Theora's coding quality is obviously poor relative to contemporary competition. This poor performance stems both from implementation and design deficiencies."
I'm sure that I'll be modded down for my following comment, but I post it anyway:
Vorbis is pretty much dead. While its quality is good, Vorbis has quite high performance requirements just for decoding (negligible on current desktop PCs, but not on portables that run on battery). Even Vorbis's developer Xiph.org acknowledged that and instead of trying to "fix" Vorbis, they started development of an entirely new audio codec called Ghost.
While Vorbis and Theora are in no way proprietary, the industry already decided to support MPEG-4. Even Microsoft supports it out of the box on Xbox 360 and Zune. Vorbis was cool when it was released, but it never had a modern video codec as companion.
Well you could always get a mp3 player that rockbox supports and install that. Not only does it support .ogg, but it also supports another feature that is rarely found on mp3 players, true gapless playback.
Mathematics is made of 50 percent formulas, 50 percent proofs, and 50 percent imagination.
Cowon's players support it. They also do FLAC.
Almost the whole range of Samsung has OGG/Vorbis support built-in.
Also, there are a lot of "NoName" asian, or less known brands (most of the time re-packaged asian "nonames") that support Swiss Bull-It is such re-packager, most of their player support OGG/Vorbis out of the box, some other after a firmware upgrade.
I know there are even OGG/Vorbis supporting devices in the "USB stick" form factor (my brother has one).
In fact, appart the few "Big Brands" who usually support only MP3 (because it's such a huge standard that they can't avoid it) and WMA/ATRAC/AAC+DRM or whatever is the proprietary format of their associated shop ; most lesser brands will support OGG because there's no technical limitation preventing it, there's no patent to prevent them, and that enables them to add another bullet point to their list, with very minimal efforts (There's already an open-source integer-math only implementation called Tremor - adding OGG support for a player usually just means recompiling tremor for whatever version of ARM serves as the player's CPU).
Sasmung is more an exception for being both a known brand and providing OGG support.
As a matter of fact, I've always encouraged people to keep a copy of their library in a loss-less format too.
This way, there's no quality loss in case of quality loss, in the event of having to shift formats, or use a newer version of the usual codec with better compression.
Depends on what format the people chosed to save their library into.
I've already had friends with their libraries of WMA changed into coaster because they reinstalled windows, or changed some hardware which triggered windows thinking that it is on a different PC.
On the other hand, all you need to play OGGs is just to choose your player wisely. Either stick only 1 brand (Samsung ), or if you want to go for the cheap, accept having a player with an obscure name that nobody has ever heard about (and which will have changed business before next year)
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
You are correct in saying Ogg is not a codec. But when you compare Theora to VC-1, you must not have been reading the license terms of VC-1 properly. VC-1 is riddled by patents and there are royalites to pay when you use it: http://www.mpegla.com/pid/vc1/ . There is no such thing as royalties to pay for Theora. Also, the only patent on Theora were ones owned by On2 Technologies, who donated their VP3 codec as the *basis* technology for Theora and kindly granted an unrevocable free license regarding those patents: http://www.theora.org/benefits/. As for quality - yes, Theora is a generation behind in compression technology and H.264 is much better quality at lower bitrates. Again - have you read the license conditions? Theora is simply the only open codec standard (as to the definition of Open Standard by Buce Perens: http://perens.com/OpenStandards/Definition.html) with a usable implementation. Mind you, I would watch out for the BBC's Dirac codec http://dirac.sourceforge.net/ which is based on Wavelet technology and is thus opening a whole new space of new video codec developments and improvements - a space H.264 didn't enter. And Dirac is an open standard.
I'm an infovore...
You're missing the point here: Vorbis + Theora is the only major non-patent-encumbered (and therefore legal to use commercially or in free software without paying a bunch of lawyers to figure out what patent fees you owe who) option for video.
MPEG-4 and similar are great for pirates and organizations big enough to have patent lawyers on staff - but standards have to do better than that. Small companies and free software projects have to be able to play too.
-- The act of censorship is always worse than whatever is being censored. Always.
Ogg is like Quicktime or ASF. There's nothing technically stopping anybody from delivering a mp3 inside an Ogg (seriously), Quicktime, or ASF container. Here's proof:
Putting a
Which costs more to distribute, an mp4 stream or an ogg stream?
According to a news release from 2002 which is hosted on the MPEG LA site, the price for mp4 was:
2. In the case of Internet (wired and wireless) or mobile, annual royalties with annual limitations and thresholds will apply: (a) for the manufacture and sale of decoders and/or encoders: US $0.25 per activated decoder and/or encoder subject to an annual cap per legal entity of $1,000,000 for decoders and $1,000,000 for encoders (to be paid by the manufacturer that offers functioning product for sale or distribution, either directly or through a chain of distribution, to the end user), but there is no royalty for the first 50,000 decoders and first 50,000 encoders in a calendar year sold or distributed by a legal entity (applies to no more than one legal entity in an affiliated group); (b) for the use of decoders and encoders to decode or encode MPEG-4 video (to be paid by the party that is the apparent source of such video to the consumer), a licensee may choose to pay US $0.25 per subscriber per year or US $0.000333 per minute of MPEG-4 video used, each subject to an annual cap of $1,000,000 per legal entity, or a $1,000,000 annual paid-up fee (with no royalty reporting obligation), but no royalty is payable on the first 50,000 subscribers during a calendar year (applies to no more than one legal entity in an affiliated group). Subscriber refers to each unique viewer for any part of a year, but where the content provider's remuneration is not directly from subscriptions (e.g., advertiser-supported services), MPEG LA will work directly with Licensees to come up with a consistent method of counting subscribers that works with their business models.
3. In the case of Stored Video (packaged media and video transmitted and stored for viewing for which a transactional fee is paid), the replicator or content provider will pay (a) US $0.01 per 30 minutes or part to a maximum of US $0.04 per movie; (b) US $0.005 per 30 minutes or part thereof to a maximum of US $0.02 per movie where the content of the Stored Video is 5 years or older (after it was copyrighted or subject to be copyrighted), and (c) US $0.002 for a Stored Video of 12 minutes or less.
So, if the current terms even vaguely approach this older release, the difference in price/time sacrifice for the higher file size is more than offset by the pricing. Dollars and cents, free and open makes sense.
Anyone got current/more accurate pricing info?
-1 Uncomfortable Truth
Except for the technical advantages. I can take a PCM file from a ripped cd and encode it with similar settings in iTunes (aac), neroenc (aac), lame, and the vorbis encoder, and everything but lame and vorbis sound like the range is compressed too far, and mp3 sounds poor on some tracks that are difficult to encode, but on the vorbis version those same passages are not as annoying.
I've done this on multiple tracks on multiple machines with good earphones, vorbis is always the least annoying for passages with encoder defects. However i do have an 3gen nano so vorbis isn't a real option, nor is alternative firmware.
I can verify this. However, the iPod causes a loud click when the buffer starts about one second before the current track ends. Rockbox doesn't have this issue. You can also change the buffer length in Rockbox. I think the iPod's is only a half second, to be honest.
Your ad here.
Uhh...Ogg was designed specifically to be open and patent free.
It can never be like GIF.
I dont see what Nokia is talking about however.
W3C is making a tag for html (or similar) and they need a open format which everyone can use.
Why any DRM is required is puzzling because a) everyone has to be able to view it and b) its video over the net.
You probably wouldn't be buying a movie and then streaming it over the net in your browser (or your phone).
Infact Nokia's own selection criteria is contradictory since you can never have a completely open DRM system.
It requires security by obscruity otherwise everyone can bypass it easily.