Weekly Edition
Return to the Front page
| |||||||||||||
GStreamer's MP3 for Linux
The MP3 audio format is a pain. It is patent-encumbered, making it hard for
Linux distributors to package (or Linux users to use) legally in various
parts of the world. It doesn't even sound all that good, compared to some
of the alternatives. Yet MP3 is hard to avoid; digital audio players often
prefer it, and much of the interesting audio content to be found on the net
is encoded as MP3. So Linux users who do any amount of audio listening
with their systems generally end up with MP3 software on their systems even
if their distributor refuses to include it.
The hassles of tracking down unofficial repositories, configuring a system to use those repositories, and installing MP3-capable software are something that many Linux users take in stride. Using Linux has often required some of that kind of work, after all; you newcomers should just be happy that you don't have to come up with your own XFree86 modelines anymore. But the lack of native MP3 capability is an impediment for potential users who want things like audio to simply work without a bunch of fiddling around. Such people tend to be uninterested in discussions of the evil of software patents and the superiority of Vorbis audio. None of that helps them listen to their favorite Norwegian reggae Internet radio station. The folks at Fluendo - the main force behind GStreamer - have made an attempt to improve this situation. Fluendo has bought a patent license for the MP3 technology, and has used it to make a couple of different items available:
There are other freely-licensed MP3 decoders available, but the Fluendo release is still worthy of note due to its use of the BSD license. Most MP3 codecs are licensed under the GNU GPL, which includes this language:
If you cannot distribute so as to satisfy simultaneously your
obligations under this License and any other pertinent obligations,
then as a consequence you may not distribute the Program at
all. For example, if a patent license would not permit royalty-free
redistribution of the Program by all those who receive copies
directly or indirectly through you, then the only way you could
satisfy both it and this License would be to refrain entirely from
distribution of the Program.
This language makes it hard to distribute patent-encumbered, GPL-licensed code, so the redistribution of any application containing a GPL-licensed MP3 codec is legally dubious. The use of the BSD license is an attempt to avoid this particular clause of the GPL. One should not forget, however, that the result of linking BSD- and GPL-licensed code is normally a derivative work which must retain the GPL license. So it is not clear that any GPL-licensed GStreamer application can be shipped with the new decoder. The binary-only plugin has a different intent. This plugin is fully licensed, so any Linux user (on a supported version of a supported distribution on a supported architecture) should be able to install it and play MP3 files without patent worries. Distributors can also sign a contract [PDF] with Fluendo which allows the binary plugin to be included with a distribution. There are plenty of restrictions in the contract, including an end-user license which prohibits further redistribution. So, while a distributor can gain the right to legally distribute the binary-only MP3 decoder, any derivative distributions must sign their own agreement with Fluendo to obtain the same right. The contract also prohibits "embedded" use, so projects like iPodLinux seem unlikely to be able to ship this plugin. So Linux distributors can now ship MP3-capable distributions, as long as they don't mind the little fact that any such distribution is not 100% free. Which distributors will accept this deal remains to be seen; in a few cases, some guesses can be made based on the discussion (or lack thereof) on the relevant mailing lists:
Most other major distributions do not currently appear to have a public discussion going. In an ideal world, our systems would include free codecs for all of the widely-used audio and video formats. The world we actually live in, unfortunately, requires that we set our expectations a little lower. While many of us can do nicely with formats like Ogg much of the time, the simple fact is that missing MP3 support makes Linux less useful for many people. And this is not a problem that can be solved by coding. The contributions from Fluendo do not qualify as a solution, but they could well help make Linux work for people who were not able to do what they wanted previously. That's a step in the right direction, even if it is not ideal. (Log in to post comments)
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 21:49 UTC (Mon) by spot (subscriber, #15640) [Link] I doubt that Fedora will include the binary only component because it would be a major break from existing standards.
Fedora Core and Fedora Extras currently only include free, and unencumbered open source packages. The binary only mp3 component would not meet this criteria.
We do mind this "little fact", as it is the underlying foundation for open source development. If Linus had been tolerant of using the Minix kernel without source, would we be where we are today? If Richard had decided that he didn't need to share the source code for his utilities?
What if Red Hat took this same stance towards RHEL, and didn't release any of the source code packages? Would it still be a "little fact"?
It's not that hard to add mp3 support to Fedora post-install, the internet is covered in easy to find documentation and prepackaged rpms. But for Fedora to endorse a closed source world? No thanks. It just puts us farther away from where we want to be, and tells software patent holders (and large hardware companies) that binary only components are good enough. This is Arjan Van de Ven's slippery slope: http://lwn.net/Articles/162686/
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 22:09 UTC (Mon) by jdub (subscriber, #27) [Link] Note that there's no "binary only component", just a patent-licensed binary and the MIT licensed source to go with it. You can use the patent-licensed binary from Fluendo, or if your distributor signs a distribution agreement with Fluendo, a patent-licensed binary built by them. In all cases, the code is 100% Open Source.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 22:16 UTC (Mon) by piman (subscriber, #8957) [Link] The point is that, since only the binary is patent-licensed, it does no good for free software. I can't link it to GPLd software. And if I choose to "risk" violating the patent, I can just use MAD, which is a better decoder than this.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 22:21 UTC (Mon) by spot (subscriber, #15640) [Link] The limits on redistribution would tend not to make it open source compatible. Breaks clause #1 of the OSI definition:
"1. Free Redistribution
The license shall not restrict any party from selling or giving away the software as a component of an aggregate software distribution containing programs from several different sources. The license shall not require a royalty or other fee for such sale."
In addition, Fluendo doesn't have permission to grant patent-licensing to anyone besides themselves:
"Fluendo has got the relevant licenses to cover its own
So, you can look at the source code, but not legally use patent infringing code without a license from the patent-holder (Thomson). So even the "patent-licensed" binary isn't really kosher in the US (other parts of the world vary).
Believe me, I'd love to have legal, uninfringing, open source mp3 playback, but this isn't it.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 22:42 UTC (Mon) by jdub (subscriber, #27) [Link] Well, firstly - this is the patent-license for the binary, not the source. The code is MIT licensed, and thus, 100% Open Source. Sure, Fluendo can't grant permission to distribute binaries based on random use of the MIT code - you must enter into a contract with Fluendo to receive a patent license.
MP3 is hard. Patent licenses are hard. Here, at least, is a legal option for those who care about solving the problem legally, and with fairly minimal intrusion for users and distributors... *if* reality means more to you than some of the more abstract notions of software freedom (ie. if you were being serious about software freedom, you wouldn't be using MP3 in the first place).
MAD has a serious problem: Look at its license (GPL), then read section 7 of it. Bad news for everyone.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 22:47 UTC (Mon) by piman (subscriber, #8957) [Link] > MAD has a serious problem: Look at its license (GPL), then read section 7 of it. Bad news for everyone.
Of course, most MP3-playing software is also under the GPL. Most useful software on GNU/Linux is under the GPL, period. So regardless of this plugin's license, if you distribute this plugin with software the end result is probably going to be under the GPL.
Jon mentioned this in the article -- Rhythmbox has no license exception yet, and I'm fairly sure Muine doesn't either. I'm less sure about amaroK, but at least its debian/copyright file doesn't have an exception.
So you can't actually ship the plugin with any of the major GStreamer-based music players. It solves nothing.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 23:07 UTC (Mon) by jdub (subscriber, #27) [Link] Sure, no argument: You can't distribute the plugin with a GPL user of it. That puts the responsibility on users - and they can do whatever they wish, because they're not distributing it. So even with GPL software on top, the plugin has a good use case, which it solves legally and well.
MAD being licensed under the GPL is another story: It is incompatible with itself and not distributable at all. Read section 7.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 3, 2006 1:19 UTC (Tue) by JoeBuck (subscriber, #2330) [Link] Section 7 of the GPL means that people in countries where the MP3 patents apply to software (for example, the USA) cannot distribute GPL MP3 code. However, it does not constrain those in countries where software patents are not yet valid. This means that, for now, sites like livna do not violate the GPL, even if they allow Americans to access them, and Americans downloading MP3 support from sites like livna aren't violating the GPL, though they may be infringing patents by using the code.In practice, though, only the copyright holder can enforce a GPL violation, and they won't in this case, so the only danger is from the patent infringement. And that applies to MP3 code regardless of license.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 11, 2006 20:00 UTC (Wed) by pimlott (guest, #1535) [Link] Section 7 of the GPL means that people in countries where the MP3 patents apply to software (for example, the USA) cannot distribute GPL MP3 code.Interpretations of Section 7 have gotten mad! This statement is as misleading as Section 7 of the GPL means that you cannot distribute the code you stole by hacking into Microsoft under the GPL.Neither is legal, but not because of Section 7. Indeed, the GPL itself considers Section 7 a no-op: This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License.A summary of Section 7 would be: If you cannot distribute a body of code under the GPL, you cannot distribute the code under the GPL.A shorter summary would be: True.I'm being extreme--probably Section 7 has some legal significance. However, I don't believe that it forbids distribution in any cases where it would otherwise be allowed. I know that Eben Moglen has said or suggested otherwise, but I haven't heard a plausible explanation. And to be clear, I consider the "for example" in Section 7 a complete non-sequitur.
reality & freedom Posted Jan 3, 2006 1:39 UTC (Tue) by pjm (subscriber, #2080) [Link] The previous poster suggests that the availability of this binary is clearly a win in real terms, but this is far from clear in the longer term.
It would clearly be contradictory to think that software freedoms have no connection with reality: each of the freedoms in question is in the physical world, reality.
Inability to share this binary among others is one real-world result. Other real-world harms come from perpetuating the idea that it is wrong to help one's neighbour by sharing one's software.
Inability to modify the source and use the resulting different binary has yet other harms: for example it is harder to take advantage of bug fixes & feature improvements of others.
Additional convenience in using the mp3 format reduces the rate of transition to better formats. Thus, Fluendo might instead consider distributing a command-line utility to convert mp3 files to better formats (allowing separate, unencumbered, gui interfaces to that tool). This would have more chance of helping the transition to other formats; whereas a gstreamer plugin will clearly reduce this transition.
Paying the patent license fee and slowing down the transition to better formats each increase monetary payments to the patent holder, and increase the attractiveness of patenting things (to the detriment of everyone else), increasing the likelihood of future patent harm.
Of course this binary does provide real-world benefit as well: one immediate benefit of easing the playing of some music files, and future benefits through increasing the use of other free software for people who would use mp3 software no matter what.
reality & freedom Posted Jan 3, 2006 15:30 UTC (Tue) by Felix.Braun (subscriber, #3032) [Link] Can Fluendo also write a program that lets me play vorbis files on my portable media player? Didn't think so.
Of course, people who care about software freedom could vote with their euros and buy media players that do play vorbis. But try telling that to the novice computer users that some linux distributions are trying to attract. To these people software freedom looks like an inhibition to use gadgets that run quite well under competing setups.
So, by lowering the inhibitions to such users, Fluendo is arguably helping Linux adoption by a wider public. Which might even help convince hardware vendors to support alternative audio formats in their firmware.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 3, 2006 13:06 UTC (Tue) by rickmoen (subscriber, #6943) [Link] jdub wrote:The code is MIT licensed, and thus, 100% Open Source. Sorry, Jeff; that doesn't follow. Being under an OSI-approved licence is necessary but not sufficient to make the codebase open source. Additionally, source code must be actually available (which is true in this case, but needn't be the case for somethine merely declared to be "MIT licensed"), and there must not be external legal obstacles preventing exercise of the rights in that OSI-approved licence. Such as, notably, patents. Thus, in parts of the world where such a codebase is patent-encumbered (rights not exercisable on account of patent-based restrictions), it's proprietary by definition, as long as those conditions persist. That's not to say I might not, in some cases, ignore a patent I felt to be either meritless or basically harmless. But it's simply wrong to say that the licence terms are all that matter. No, it's those terms and the ability to exercise them that matter. Rick Moen
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 3, 2006 16:48 UTC (Tue) by ranger (guest, #6415) [Link] Such as, notably, patents.Thus, in parts of the world where such a codebase is patent-encumbered (rights not exercisable on account of patent-based restrictions), it's proprietary by definition, as long as those conditions persist. By your definition, the Linux kernel is proprietary, since it includes patent-encumbered code (such as RCU). But, the Linux kernel is not proprietary, and considered Open Source (and Free Software etc etc). Why? Because of the license IBM has granted for the use of the RCU code, which most people agree has acceptable conditions. So, the real issue is whether the Linux distribution in question is agreeable with any license terms on an MP3 decoding license. For example, Mandriva has negotiated with Thompson that it (and others using any Mandriva-based distribution for a non-embedded device) may provide royalty-free MP3-decoding, so MP3 playback works out-the-box. While it would be nice to have a policy on this in each Linux distribution, people are still arguing the wrong points in this thread. If you don't want any patent-encumbered code, better not run a 2.6 kernel.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 3, 2006 17:07 UTC (Tue) by rickmoen (subscriber, #6943) [Link] "ranger" wrote:But, the Linux kernel is not proprietary, and considered Open Source (and Free Software etc etc). Why? Because of the license IBM has granted for the use of the RCU code, which most people agree has acceptable conditions. You seem unaware that you are actually in strenous agreement with me -- presumably because you didn't read what I wrote attentively. Allow me to reiterate: "patent-encumbered (rights not exercisable on account of patent-based restrictions)" and "external legal obstacles preventing exercise of the rights". I'm extremely well aware that by no means does the mere existence of an applicable patent (or other legal encumbrance) does not, in itself, preclude exercise of open-source licensing rights. My comments concerned particular patent encumbrances (and other legal obstances) that do happen to be imposed in a fashion that blocks exercise of those rights. You, unfortunately, didn't notice that I'd already carefully observed that (elementary) distinction in my post, which is why I'm now obliged to waste time repeating myself. Oh well. Rick Moen
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 22:17 UTC (Mon) by elanthis (guest, #6227) [Link] Arjan's "slippery slope" has little to do with this plugin. The binary nature is not meant to hide anything from the user, stop them from recompiling the plugin, stop them from redistributing the plugin's source, or anything like that. The patent holders require that binary MP3 decoders be licensed. This plugin makes it possible to legally (in many countries) ship MP3 support.
Also keep in mind that every single one of those "easy to install" Fedora MP3 add ons are (1) in no way easy to install unless you're a Linux dork with nothing interesting to do with your life besides sit in front of a computer and learn how things like yum and rpm work, and (b) are all illegal in many countries, including the the USA.
The only "easy to install" that actually exists is "already installed and configured." If the distribution doesn't come with MP3 support enabled in its default media player(s), then there many/most users are not ever going to get MP3 support, and are likely to just blame Linux since "it sucks and can't even play MP3s" and will use a [non-free] OS that actually "works."
This plugin from Fluendo can also be a vehicle to achieve Freedom. Assuming you want to do things legally (which many large companies sure as heck do, not to mention some overly-honest/anti-anarchist home users), this plugin from Fluendo is the only way to digitally convert an existing set of MP3s into unencumbered formats such as Vorbis.
Finally, this plugin is also a decoder *only*. Fluendo is not releasing an MP3 encoder under this terms, as they do not want users to either intentionally or inadvertantly create new content in an encumbered format.
The plugin exists to assist in migration to a Free platform and to unencumbered formats. Even if some do consider the plugin a "slippery slope," it's the *only* [legal] slope available to many people.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 2, 2006 22:29 UTC (Mon) by spot (subscriber, #15640) [Link] There are distribution restrictions, and Fedora would have to ship a binary without matching source code. This is ignoring the legal issues raised in the above comment. Of course, IANAL, so the Fedora Foundation legal team might think its all fine and dandy (but I seriously doubt it).
I'm not convinced that the Fluendo plugin is legal in a Fedora distribution without Fedora (re: Red Hat) paying for the patent license from Thomson, since Fluendo doesn't have the ability to "resell" the patent license.
This is also a slippery slope in that it tells Thomson that their practice of collecting royalties on software patents is ok. It opens the door to all sorts of binary components going into the distribution, because they "work"... and then we're no better than Windows.
I know that this sucks for end-users, but we can either sweep this under the rug, or stand for open source principles.
GStreamer's MP3 for Linux: Thanks, but no thanks. Posted Jan 3, 2006 15:23 UTC (Tue) by Uraeus (subscriber, #33755) [Link] To clarify something in regards to our abilility to 'resell' our patent license. We have the ability to let our business partners redistribute our plugin with their products. So anyone who signs a contract with us can freely redistribute our plugin with their own products. The 'reselling' part you are mentioning is that we of course can't give everyone the same right we have, meaning that people could sign a contract with us, then have other people sign contracts with them and so on. Everyone have to either sign a contract with us to distribute our plugin, or get a contract with Thomson/Fruenhoffer to distribute their own plugin (which they could build from the source we provide).
Christian Schaller
GStreamer's MP3 for Linux Posted Jan 2, 2006 22:14 UTC (Mon) by piman (subscriber, #8957) [Link] This plugin offers no real benefit for free software that the existing MAD plugin didn't. More likely, it's slower and lower-quality than MAD. And gstreamer-mad is already in most distributions' "external" repositories.
GStreamer's MP3 for Linux Posted Jan 3, 2006 1:25 UTC (Tue) by JoeBuck (subscriber, #2330) [Link] What evidence do you have that Fluendo's plugin is slower or of lower quality? After all, they are a company that specializes in streaming audio and video. In the absence of benchmarking, we don't know the answer, but even if there is a quality problem, there is no legal barrier in Fluendo's way if they want to improve it.
GStreamer's MP3 for Linux Posted Jan 3, 2006 1:53 UTC (Tue) by piman (subscriber, #8957) [Link] Fluendo says: "The other drawback is that our code is probably not yet of the same end quality as madÂ’s." -- http://thomas.apestaart.org/log/index.php?p=333
Is that good enough for you?
GStreamer's MP3 for Linux Posted Jan 3, 2006 20:04 UTC (Tue) by JoeBuck (subscriber, #2330) [Link] Note the word "yet".
GStreamer's MP3 for Linux Posted Jan 3, 2006 12:25 UTC (Tue) by job (guest, #670) [Link] Exactly. I don't get it. If I'm and end user, I still have to install extra software to play my MP3s so the situation hasn't improved one bit. If I'm a Linux distributor I wouldn't be interested in anything that restricts redistribution. So no gain.
I may be talking about things I don't understand now, but since it is apparently legal to discuss and distribute source code because sites such as sourceforge can host decoders, can't a Linux distribution just fetch and compile an mp3 decoder as part of the automated post-install?
GStreamer's MP3 for Linux Posted Jan 3, 2006 12:41 UTC (Tue) by louie (subscriber, #3285) [Link] "If I'm a Linux distributor I wouldn't be interested in anything that restricts redistribution. So no gain." Wha...? That's a fairly optimistic, naive view of Linux distributions. Every for-profit distro (that is, most of them) have been dying to have a well-supported, legal mp3 solution for their enterprise desktop products for years. Assuming the quality is good and the pricing is right, this will likely be in every enterprise desktop distro within a year. Tangentially, the 'this is still a problem for derived works' argument is accurate, but gstreamer have written some licensing language which authors of gstreamer-derived players can use to supplement the GPL and indicate that they are OK with the use of binary-only/non-GPL compat gstreamer modules. I believe most gstreamer-based players use this exception language, so the problem is covered. I do wish LWN had actually asked the gstreamer guys about this before posting this article- sloppy research on their part.
GStreamer's MP3 for Linux Posted Jan 3, 2006 18:40 UTC (Tue) by piman (subscriber, #8957) [Link] > I do wish LWN had actually asked the gstreamer guys about this before posting this article- sloppy research on their part.
Or, sloppy reading on yours; from the article: "There is some discussion of whether it would still be legal to include rhythmbox, which is GPL-licensed with no plugin exception, in such a work."
GStreamer's MP3 for Linux Posted Jan 4, 2006 21:56 UTC (Wed) by spot (subscriber, #15640) [Link] RHEL already "includes" RealPlayer, which has licensed mp3 support. The reason there are quotes around the "includes" is because it is provided to customers as an unsupported addon, since Red Hat doesn't have RealPlayer source.
My favorite radio station... Posted Jan 2, 2006 22:16 UTC (Mon) by rknop (guest, #66) [Link] ...online is WCPE. Classical music, public radio. *And* they've got an Ogg Vorbis stream! What more could you ask for?!?!?!?
Well, to be able to get it in my car, that's the main thing I'd like.
-Rob
My favorite radio station... Posted Jan 3, 2006 9:07 UTC (Tue) by nurhussein (guest, #16226) [Link] Can I have the linky? :)
My favorite radio station... Posted Jan 3, 2006 14:40 UTC (Tue) by rknop (guest, #66) [Link] Website: WCPE.org
Page of links to streams: http://theclassicalstation.org/internet.shtml
(The Ogg stream is the second one they list; they also have an MP3 stream, but I ignore that, of course :) )
-Rob
My favorite radio station... Posted Jan 13, 2006 1:34 UTC (Fri) by andrel (subscriber, #5166) [Link] The list of Vorbis streams on the Xiph wiki lists a number of stations streaming in Ogg Vorbis.
MP3 patent expiration? Posted Jan 4, 2006 11:32 UTC (Wed) by Duncan (guest, #6647) [Link] Unlike copywrong, patents aren't yet virtually eternal. AFAIK, it's17-20 years here in the US. (I believe now 20, extended three as a tradeoff for something else, don't know the details, a few years ago. However, I believe we're still in the period during which some 17 year patents haven't yet expired.) It would have been nice if the article had included the expiration date of the MP3 patents, so we'd know how much longer we had to worry about it. Obviously, the format has been around for several years already, so it's quite possible we're already half-way there. This is of practical interest once one compares it to the GIF patents which recently expired. Along that line, Gentoo has recently deprecated libungif, and all new ebuilds depend on libgif, tho it's quite likely folks' stable installations will have libungif dependencies for a few more months, anyway. So... you see, it /does/ happen that patents expire within a software's lifetime. =8^) If they'd just make the lifetime of software patents something more realistic in terms of average software life, say five years or so, I have a feeling there'd be far less pressure to do away with them entirely. Maybe that'll be what will actually happen, eventually, as the forces for serious patent reform continue to mount. Of course, doing away with software patents altogether would be better, but five years would certainly be better than twenty -- probably /enough/ better to take the pressure off of killing them entirely. So when /do/ the MP3 patents expire? Anybody know? Duncan
MP3 patent expiration? Posted Jan 4, 2006 14:00 UTC (Wed) by frazier (guest, #3060) [Link] I was curious so I did some hunting and found this page at wikipedia:http://en.wikipedia.org/wiki/Talk:MP3
Look under "patent".
Here's a patent portfolio from Thompson:
...and some more talk on the subject...:
Seems like a good topic for the EFF to have a write-up on, and maybe thay have one, but I wasn't able to find it.
-Brock
MP3 patent expiration? Posted Jan 4, 2006 21:08 UTC (Wed) by Duncan (guest, #6647) [Link] Thanks!It appears from your links that there are a collection of patents claimed to apply, mid eighties thru mid nineties. The MPEG standard completely defined decoding in 1991, so no decoding patents can apply beyond 20 years after that (US, I understand others expire faster), so 2011, five years off. That's not too long, and some may have already expired or will be expiring between now and then. Certain parts of the encoding process may be covered thru 2015-ish. Apparently, however, LAME was one of the earliest decent quality encoders, in good part responsible for the popularity of MP3, and Thompson appears to have learned something from the GIF/PING thing, and hasn't gone after LAME, tho they probably could. They probably realize that were they to do so, they'd only push competing formats such as ogg vorbis, just as GIF enforcement pushed PING. Early on, that would have discouraged adoption and popularization; now, with the patents on the way to expiry and other, better quality codecs available, they probably just don't want to rock the boat. Rather, they have pushed the commercial license side, and seem to have been fairly successful in doing so. Thus, while in the gray area, FLOSS MP3 implementations appear to be relatively safe, as long as they don't push the commercial side. In any case, 2011 isn't /that/ far away, and we can all rest easier when it arrives. After that, 2015 won't be far away. So.. a few more years of gray area, but it won't be /that/ long, from a user's perspective, anyway. Of course, being a commercial distributor puts things in a far different light, but the clock is still counting down, either way. Duncan
MP3 patent expiration? Posted Jan 5, 2006 10:17 UTC (Thu) by nix (subscriber, #2304) [Link] PING? You mean PNG, I think.
(PING, of course, is a duck.)
MP3 patent expiration? Posted Jan 6, 2006 11:29 UTC (Fri) by Duncan (guest, #6647) [Link] Well, pronounced "ping", but you are correct, it's officially PNG. (Youprompted me to look it up, so now I know. Thanks!) Duncan
|
|||||||||||||