Recently there was a symposium at Seneca College, Canada called the Free Software and Open Source Symposium (FSOSS). The name tells you what the speakers were supposed to address.
Unfortunately, and contrary to the name of the event, software freedom discussion is apparently not welcome and the recordings are needlessly encoded such that many free software users can’t play them.
See the updates and comments at the end of this post for some news on the struck portions of the post.
The recordings of the talks came online the other day and so far this much is clear: The videos are distributed exclusively in a patent-encumbered format you cannot play with free software in the US. It makes sense that a conference focused on free software and open source issues would distribute media in a way one could play with free software. The videos are encoded in DiVX format. DiVX and OpenDiVX are non-free because they build on patent-encumbered codecs:
DivX uses lossy MPEG-4 Part 2 compression(and so does the so-called “open source” OpenDiVX) according to Wikipedia as of the time/date stamp on this post.
Users in countries whose governments grant patents on algorithms used in computer software (also known as “software patents”) can’t play DiVX or MP3 files without breaking patent law or getting a patent license. Patent licenses for such things are notoriously incompatible with free software. The only semi-saving grace here is that the videos are licensed under a Creative Commons license that allows one to convert these videos to a free format. The future doesn’t bode well for people who care about freedom, as far as these videos are concerned. According to David Humphrey, they’ll only choose non-free formats for copies of the recordings to be distributed later:
Over the next little bit we’ll be adding WM, QT, Real, and MP3.. Humphrey has turned off comments for his blog entry. Apparently there are no plans to distribute Ogg Vorbis+Theora versions of these recordings. For those of you who aren’t so technical, that list of formats includes nothing which can be played on a free software system on which nothing but free software is installed. One might have to violate patent law (depending on the patent regime the user is under) or install proprietary software to watch any of the videos the symposium plans to distribute.
But why didn’t the distributors of the video choose to distribute free codec versions of these videos in the first place? Why should freedom take a back seat to sidling up to patent-encumbered and proprietary formats? I think it’s because freedom wasn’t in the mindset of the people who put the videos together. Like I’ve shown you with so many other recordings aimed at a free software and open source audience, software freedom just doesn’t matter.
A friend of mine showed me the keynote speech from Nat Friedman (co-founder of Ximian which is now owned by Novell).
- Friedman gives no credit to GNU at all, for him “Linux” is an operating system and (even though I’m sure he knows better) he doesn’t say anything to correct the misperception that Linux is an operating system (consider these points before you post with your counterpoint to saying GNU by name). If all you have is the Linux kernel, you won’t be able to do the things Friedman brags about as useful on a “Desktop Linux” system. One can see why Friedman gives no credit to GNU—it might raise questions of freedom, and thus place software freedom within the allowable limits of discussion.
- Friedman started interestingly enough by pointing out that a number of people in the audience had MacOS X machines and that MacOS X has proprietary software in it (starting at 1m30s into the recording—audio or video, wide angle video). Thus, MacOS X, as a whole, is not free software; it’s a mix of some free software and some proprietary software. But his own system (the machine he used in his talk) has the same flaw. In fact much of the software he demonstrated exhibited the same characteristics. The pitch he made was aimed at convincing you that his GNU/Linux system is convenient and modern. Yet he bragged about a deal with Real Networks (49m) that lets him play patent-encumbered and proprietary audio files “out of the box” (50m10s). Around 49 minutes into the recording, Friedman mentions that other GNU/Linux systems only play free formats (like Ogg Vorbis and Theora , ironically the very formats this recording wasn’t in) but he didn’t mention why that is so (again, a discussion that might bring up freedom). Perhaps he was also using a proprietary video driver to run his video card to do graphical effects (he mentioned how easy it was to install the non-free ATI driver software (57m40s), not framed in terms of losing your software freedom, of course). It’s hard to take someone’s message seriously when they’re willing to abandon the restrictions they brought up earlier out of inconvenience.
The FSOSS reminds me of the difference between the two movements imposed by those who either don’t like software freedom or think technical matters matter more than freedom. I’m also reminded of how so many open source advocates basically don’t demonstrate that they care about software freedom; when a proprietary way to do something comes along they’ll encourage others to use it and they’ll use it themselves. And in so doing argue against even their own relatively unimportant line about why anyone should favor “open source” at all. That movement stresses developing software faster, with fewer bugs, and for less money (an argument aimed at managers, not users). These points are nowhere near as important as building community and defending the freedoms which make those points possible—the freedoms to run, inspect, share, and modify software. Yet the use and display of proprietary software argues against open source points entirely. Talking about treating your neighbors well by giving them the permission to run, inspect, share, and modify is immaterial and inconvenient to many open source advocates. Hence, no need to mention GNU, no need to show off what a free software system can do, no need to distribute archived recordings in formats a free software user can play.
I don’t think breaking patent law is unethical like encouraging others to install and run proprietary software is, and it’s clear that the practical limitations aren’t the same. It is unethical to give others no way to avoid breaking patent law. So if you’re distributing media files, distribute copies of them in a format free software users can play right along side the proprietary and/or patent-encumbered copies. If you distribute exclusively one kind of file, pick a free codec. After all, everyone can play the free media, not everyone can play the non-free media. We don’t need MP3 and the various non-free video codecs anymore. We can do everything FSOSS is doing here with free codecs. The FISL demonstrates this every year (and I’ve mirrored a number of their talks). When you’re speaking to a free software crowd, you have an obligation to make sure your intended audience can play what you’re distributing and you ought to be sensitive to the reasons why things are the way they are with media codecs today.
Update (2006-11-03): Boing Boing points to the files for the conference without noticing the disparity between the format the recordings are distributed in and the subject matter at hand.
Update (2006-11-06): The FSOSS recordings referred to above are now available in formats free software users can play! And they’re licensed to share too! Clearly, with this change, some of the points in this post are no longer valid. Those points have been struck. My thanks go to the FSOSS for reconsidering this decision and adding free format versions of the talks.
All videos have been converted to Theora and the audio to Vorbis. These are being mirrored as we speak and will be available shortly.
Check this page for updates.
The video and audio is now available in Ogg Theora and Ogg Vorbis formats. More formats are on the way to accomodate a wide audience.