Slashdot Mirror

← Back to Stories (view on slashdot.org)

Ogg Vorbis Update: Thomson Trouble

Posted by jamie on from the do-I-put-this-under-patents-or-music dept.

3.1415926535 writes: "In this article on C|Net, Thompson Multimedia's vice president of new business Henri Linde openly threatens the Vorbis project. The quote is, 'We doubt very much that they are not using Fraunhofer and Thomson intellectual property. We think it is likely they are infringing.'" Considering Ogg Vorbis is GPL, you'd think they'd already know.

6 of 315 comments (clear)

  1. It's not very likely at all. by gschmidt · · Score: 5

    > It's actually quite likely that Vorbis infringes on several of Fraunhoffer/Thompson's patents

    Actually, no. Monty, Xiphophorus, and iCast did their homework, and also hired some clueful technology IP lawyers to look things over, some of whom I've met -- these guys are sharp, and they grok psychoacoustics.

    Have you read the actual patents? If you work out exactly what is being claimed, FhG doesn't actually own the farm, as I've heard it told.

    Yeah, I was skeptical too, but I've been convinced.

    Note, also, this: FhG has never actually claimed any infringement by Vorbis. FhG's lawyers could certainly read the source if they wanted to know.

  2. Re:Sheesh by ChadN · · Score: 5

    Or maybe they have a patent on "A method of compressing audio to preserve quality but reduce space usage.

    Well according to this patent, obtained by Thompson for his "invention", that may be exactly what they claim. The patent would seem to cover any audio compression method that converts from time domain to frequency domain, does quantization, then entropy coding.

    The other Fraunhofer patent is at least a bit more focused, and specifies a breakdown into frequency groups, followed by quantization, then compression. The Ogg Vorbis scheme avoids the first stage of prefiltering into smaller frequency bands, and does the transform in one feel swoop. This requires more work for the transform, but arguably gives better results.

    In short, the first patent I mentioned seems difficult to defend against, unless it can be shown to be overly broad or invalid. The second is exactly what Ogg Vorbis was avoiding.

    --
    "It's overkill, of course. But you can never have too much overkill." - Anonymous Slashdot Coward
  3. Info about the Patents by RachaelAnne · · Score: 5

    From this url given elsewhere by another poster, I looked up all the patents that Thomson Multimedia and Fraunhofer have in the US (apparently some weren't approved in US but in other countries). With all the hub-ub about overbroad/silly patents I thought I could go read some in more detail. The list of patten numbers is:

    • 5,742,735
    • 5,455,833
    • 5,579,430
    • 5,559,834
    • 5,703,999
    • 5,706,309
    • 5,736,943
    • 5,701,346
    • 4,942,607
    • 5,214,742
    • 5,227,990
    • 5,384,811
    • 5,321,729
    • 4,821,260
    (You can look up any of them at the patent office. Just enter all the numbers in the search field separated by spaces.)

    Some interesting things I noted:

    1. I know something about math (and signal processing) and I couldn't decipher what was meant by a lot of them. In other words, how can the patent office people decide if an application is a valid patent if the application is incoherent? I know that specialists are supposed to look at applications for patents in their field but they might not be up to speed about everything in their field. Yeah, they might reject it for that reason but aren't they just as likely to approve it, thinking "maybe I'm just stupid/ignorant about this sub-field."
    2. They seem to be fairly specific (although specificity doesn't necessarily mean something is patentable).
    3. These "ominous words" were found at the end of patent #5,579,430, titled "Digital encoding process":
      • Although the invention has been described and illustrated in detail, it is to be clearly understood that the same is by way of illustration and example, and is not to be taken by way of limitation. The spirit and scope of the present invention are to be limited only by the terms of the appended claims.
      Since most of the patents I found did not specify that the encoded signal had to be audio, this seems like they have a patent on any use of whatever their algorithms are trying to do (which I found not very clear...) In other words, it is almost like somebody patented a specific hash table function (I'm sure someone has) and then patented it specifically for application X, but didn't rule out the possibility of "owning" it in any application that uses hash functions.

    4. All of the above must be taken with at grain of salt because the legal-ese in the patents (especially the beginnings where the claims are listed) is very weird and I had trouble deciphering what kind of math they were getting at. Not to mention one could spend days if not weeks reading them all and all supplemental material. Overall it looks like Ogg would have to include some very specific algorithms to be infringing (unless just the fact that the patents claim to patent one method of doing a certain type/part of encoding signals is enough to claim infringement--i.e. one form of encoding algorithm counts as owning them all...but that doesn't seem very reasonable.)

      Rachael

    --
    "Go Forth Ye Lemmings and Propagate"
  4. The horse's mouth: Time to calm down folks by xiphmont · · Score: 5

    Linde said what he said only because it was an opportunity for some free FUD. Between the lines, it says, "we're worried/scared, they're on our radar, and we need to make some noise. Words are cheaper than lawyers."

    Big surprise.

    If Slashdotters didn't expect that already, well, shame on you. Sudden worried speculation about Fraunhofer's and Thomson's 'newly ominous tone' is just the snowball they'd like to start (while pressing full-steam ahead with the new webcast and download licensing). I'd be annoyed if they managed to start it with a single public sentence (we've known they didn't like us for quite a while in private). Let's not be a herd of sheep being maneuvered into the chute.

    Thomson and FhG both have a reputation of a loud bark, but tend to pursue relatively little litigation in practice and they'll have to work hard to find basis against Xiphophorus. When we did our patent review, we focused on the FhG/Thomson MPEG patents. Our counsel advises us we don't infringe, what we knew already.

    In other words, nothing's changed from yesterday except that Linde has decided to bluff before the call.

    Monty

  5. Eric Scheirer is on our side by xiphmont · · Score: 5

    Don't attack Eric for telling what he sees to be the truth. That's his job.

    Behind the scenes, he's a friend of the Ogg project and has been for some time. He's doing his job by calling it how he sees it, and we don't ask him to spin the facts toward our favor. He also doesn't have control over which quote a reporter will choose.

    Monty

  6. We should catch this earlier next time by stevens · · Score: 5

    These delayed-action patent issues are becoming predictable. The community ought to keep its collective eyes out for this in the future. While not exactly the same, the similarities are striking:

    1. gif/unisys/lzw Unisys waits until the file format is literally covering the world, and then threatens lawsuits.
    2. Rambus We've been using SDRAM for how long, and they're just waking up to the idea that they have patents on it?
    3. CDDB Don't worry, it's "free." Right. Sure.
    4. Now, Fraunhofer ...

    We should all be experienced enough with this phenomenon to see it coming a mile away. From this perspective, things like Windows Media are not competitors to mp3, they are just different complainants in the patent lawsuits.

    I strenuously suggest people use png, .ogg, and anything other technology that isn't trying to strangle open standards.

    The Internet wouldn't have existed if they played by these rules at the beginning.

    Steve