Update: John Sullivan of the Free Software Foundation responds in the Talkback section. In addition, the FSF has revised its FAQ page. See the update at the end of this post for details.
In the great human tradition of rooting for the underdog, it is tempting to want to stand up for the tiny nonprofit against the giant corporations. After all, big corporations lie and spread fear, uncertainty, and doubt (FUD). The little guys just want to spread the truth. Right?
Not always. While researching the recent battle over H.264 as the default HTML5 media format, I’ve discovered that open-source advocates are just as capable of spreading FUD as those mega-corporations. And the single worst offender is the Free Software Foundation. The FSF has started a campaign called PlayOgg, whose goal is to encourage the use of open-source Ogg media formats. On the PlayOgg FAQ page, I found one whopper after another, starting with this one:
Unlike MP3, Ogg Vorbis is not restricted by patents. Microsoft had to pay $1.5 billion after being sued for using MP3 without a license. With Ogg Vorbis, they would have been safe!
That is an outright lie. Microsoft did not have to pay a penny to anyone as the result of a lawsuit on the MP3 format.
The facts? A jury in February 2007 found for Alcatel-Lucent and against Microsoft in a patent case that lasted seven years and involved Microsoft's use of the MP3 format in Windows Media Player. Six months later, United States Senior District Court Judge Rudi M. Brewster dismissed that verdict, which called for Microsoft to pay $1.52 billion in damages to Alcatel-Lucent. The judge ruled that “the jury’s verdict was against the clear weight of the evidence”—a phrase which is repeated six times in the published decision to grant Microsoft’s motion to overturn the verdict (PDF). Judge Brewster also found that Microsoft had properly licensed the MP3 codecs in 1997 from Fraunhofer Inc., which was a co-owner of the patents.
Microsoft began using the MP3 format in Windows Media Player in 1998. Alcatel-Lucent filed its lawsuit in 2000. The FSF says “they would have been safe” if they had chosen Ogg Vorbis. That overlooks the inconvenient fact that the first stable version of the Ogg Vorbis reference software (version 1.0) was not released until July 2002. It’s hard to imagine how Microsoft could have chosen the “safe” open-source option when it didn’t exist yet.
Later in the same document, the FSF flings out another big lie:
RealPlayer, Windows Media Player, iTunes, and other popular formats require people to use non-free software: controlled by companies, not by the users. The companies that control the software design it to restrict the users and spy on them. [emphasis added]
That statement is FUD in its purest form. It is also technically absurd and factually dead wrong. The three programs listed are not formats, they are players, which handle a variety of media formats. I’ll ignore RealPlayer and resist the urge to bash iTunes and instead concentrate on Windows and Windows Media Player. Here are the facts:
- Anyone can write a media player for Windows and can build in support for whatever media formats they want. No one is “required” to use Windows Media Player—exactly the opposite. In addition, the Default Programs menu allows you to replace Windows Media Player with your preferred media player. One alternative is VLC, which I have praised before (and which has a link to the PlayOgg campaign on its homepage). In an e-mail to me, one of the core developers of VLC specifically praised Microsoft last year for its assistance, noting that “Microsoft … funded our Windows 7 compatibility program participation.” VLC, by the way, plays back files encoded in Microsoft’s Windows Media formats and in Apple’s AAC formats. [Clarification: As several readers have pointed out, AAC formats are not owned by Apple but are part of the MPEG-4 standard. They are most widely associated with Apple because they are the default format for music sold in the iTunes Store.]
- If you choose to use Windows Media Player, you can play audio and video files saved in any Ogg format (Ogg Vorbis, Ogg Speex, Ogg Theora, and Ogg FLAC) by installing the required DirectShow-compatible filters for 32-bit and 64-bit Windows versions. According to the download page at Xiph.org, which maintains the Ogg formats, the most recent release (February 22, 2010) adds “support for Windows 7 [and] HTML5 <video> tehnical [sic] preview for Internet Explorer.” I downloaded and installed the filter pack on a 32-bit Windows 7 machine and was immediately able to play back a sample Ogg Theora file in Windows Media Player.
- You don’t need to use any media player to play back streaming files in the various Ogg formats. If you have the Java virtual machine installed on your Windows PC, it will download the Cortado applet automatically and play back Ogg-formatted files in Internet Explorer 8. When you install VLC, it adds an ActiveX plugin that will play back Ogg Theora files in an Internet Explorer window. You can install Firefox, which has had Ogg Theora support built in since version 3.5. Or use Google Chrome, which also supports the Ogg Theora format natively. See for yourself using this test file at MediaWiki. In short, you’re not restricted in any way by Windows.
And then there’s this:
Best of all, [Ogg Vorbis] is designed to be completely free of patents and does not require the use of proprietary software.
That statement is almost charming in its naïveté. Except I don't believe that the author of that sentence was naive but instead is doing his or her best to mislead. As I’ve already shown, the default media file formats included with Windows don’t require proprietary software. But being “completely free of patents” isn’t a design decision, it’s a business decision. The group that maintains the Ogg formats have made a business decision to not apply for patents on any innovations they come up with for their software. Good for them—and I mean that sincerely. But making that business decision doesn’t exempt the developers of the Ogg formats from having to abide by existing software patents that are owned or controlled by other organizations, and it certainly does not give them a free pass from patent lawsuits.
So is Ogg Theora "safe"? A reasonable observer might conclude exactly the opposite, if they read this statement from Larry Horn, CEO of MPEG LA, the consortium that controls the AVC/H.264 standard. In a March 2010 interview with StreamingMedia.com, Horn said:
The license offered by MPEG LA charges a royalty for decoders wherever or however deployed. In addition, no one in the market should be under the misimpression that other codecs such as Theora are patent-free. Virtually all codecs are based on patented technology, and many of the essential patents may be the same as those that are essential to AVC/H.264. Therefore, users should be aware that a license and payment of applicable royalties is likely required to use these technologies developed by others, too. [emphasis added]
The real irony is that the FSF used previous patent litigation (the Alcatel-Lucent vs Microsoft suit) as a key argument that its own formats and codecs are “safe.” But that’s another lie: the Ogg formats and codecs simply haven’t been tested yet. As a nonprofit foundation, Xiph.org is a pretty unattractive target for a lawsuit. But if a big company—an Apple or a Microsoft or a Google—decides to build native support for that “safe” codec into OS X or Windows or Chrome OS, the game changes radically. All of a sudden those patent holders have someone to sue along with Xiph.org—someone with very deep pockets and billions of dollars of cash in the bank. And if they run into a jury like the one that tried to ding Microsoft for $1.5 billion, guess who’s left holding the bag?
From everything I can tell, the developers at Xiph.org who are working on the Ogg family of codecs (including Theora) are hard-working people who deserve a chance to tell their side of the story. But they should reject the FUD and the lies from the FSF.
* Update: The PlayOgg FAQ page has now been revised to correct some of the inaccuracies noted here. Except for a change in the revision date, there is no indication to the reader that the page has been changed. The accusations of "spying" by RealPlayer, iTunes, and Windows Media Player remain unsupported by any evidence.
Click the thumbnail below to see a cached copy of the page as it existed at the time I wrote this post. According to the note at the top of the page, it had previously been updated on April 7, 2010.