Page 4 of 5 FirstFirst ... 2345 LastLast
Results 31 to 40 of 49

Thread: Samsung Accidentally Leaked The exFAT Linux Driver

  1. #31
    Join Date
    Feb 2013
    Posts
    441

    Default

    Quote Originally Posted by tga.d View Post
    A couple points in response to quite a few comments:
    Just because code violates the GPL doesn't mean it gets released under the GPL. I haven't seen anyone make this mistake explicitly, but I feel as though it has been implied to some degree. So for example, if Samsung did violate the GPL with this code, and it was supposed to be licensed under the GPL, there are multiple legal resolutions, of which licensing the code under the GPL is only one. More likely, they would pay a fine and rewrite the code or something similar. Because of that, even if the code that rxrz released supposedly under the GPL was supposed to be under the GPL originally, doesn't mean it actually is for certain legally licensed under the GPL now. In fact, since it violates Samsung's license, it's not. A violation of a license that violates a license isn't necessarily legal.
    And secondly, it's entirely possible Samsung cannot license the code under the GPL because it contains code or patents that aren't theirs to license (namely, Microsoft's), which were licensed to Samsung to use and maybe even distribute in binary form, but not relicense.
    You are right that the Samsung code hadn't explicitly been released under the GPL, however the GPL contains an implicit acceptance clause:

    5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.
    Since Samsung has modified and distributed the derivative work (in binary form), according to the GPL they have already indicated their acceptance of the GPL. If that is not the case, they need to explicitly say that they haven't accepted the GPL, but then they are liable to large fines and several years jail for criminal copyright infringement.
    Last edited by chrisb; 07-24-2013 at 04:27 AM.

  2. #32
    Join Date
    Feb 2013
    Posts
    441

    Default

    I wonder if anyone has contacted Samsung over this? It would probably be a better idea than guessing as to their intentions. Not releasing the code may have just been an oversight.

  3. #33
    Join Date
    Nov 2008
    Posts
    25

    Default How to workaround

    Someone will read the code, write description of how the driver looks like, and smoebody's going to write a clean-room driver. Patent research would also help, if it's possible to workaround them.

  4. #34
    Join Date
    Dec 2008
    Posts
    152

    Default

    Quote Originally Posted by chrisb View Post
    I wonder if anyone has contacted Samsung over this? It would probably be a better idea than guessing as to their intentions. Not releasing the code may have just been an oversight.
    I contacted Samsung's Open Source Release Center.

    We're pretty much guaranteed a response now.

  5. #35
    Join Date
    Jan 2011
    Posts
    193

    Default

    Quote Originally Posted by tga.d View Post
    A couple points in response to quite a few comments:
    Just because code violates the GPL doesn't mean it gets released under the GPL. I haven't seen anyone make this mistake explicitly, but I feel as though it has been implied to some degree. So for example, if Samsung did violate the GPL with this code, and it was supposed to be licensed under the GPL, there are multiple legal resolutions, of which licensing the code under the GPL is only one. More likely, they would pay a fine and rewrite the code or something similar. Because of that, even if the code that rxrz released supposedly under the GPL was supposed to be under the GPL originally, doesn't mean it actually is for certain legally licensed under the GPL now. In fact, since it violates Samsung's license, it's not. A violation of a license that violates a license isn't necessarily legal.
    And secondly, it's entirely possible Samsung cannot license the code under the GPL because it contains code or patents that aren't theirs to license (namely, Microsoft's), which were licensed to Samsung to use and maybe even distribute in binary form, but not relicense.
    but is rxrz has copyrights in the kernel, then he does have a pretty good counterclaim, that may just end up resulting in a bilateral injunction against distribution, however Samsung has a lot more to lose as the result of such an injunction.

  6. #36
    Join Date
    Jan 2011
    Posts
    193

    Default

    Quote Originally Posted by chrisb View Post
    You are right that the Samsung code hadn't explicitly been released under the GPL, however the GPL contains an implicit acceptance clause:



    Since Samsung has modified and distributed the derivative work (in binary form), according to the GPL they have already indicated their acceptance of the GPL. If that is not the case, they need to explicitly say that they haven't accepted the GPL, but then they are liable to large fines and several years jail for criminal copyright infringement.
    Don't know if it will fly in court, but worth a shot, however there are several licenses that satisfy the requirements for linking to GPL 2.0. hower the GPL 2.0 is the most restrictive license that satisfies the requirements of the gpl 2.0.

  7. #37
    Join Date
    May 2013
    Posts
    556

    Default "Bilateral injunction" indeed

    Quote Originally Posted by WorBlux View Post
    but is rxrz has copyrights in the kernel, then he does have a pretty good counterclaim, that may just end up resulting in a bilateral injunction against distribution, however Samsung has a lot more to lose as the result of such an injunction.
    Hmmm-Samsung has IP from MS mixed with GPL kernel code, thus two incompatable licenses. Samsung cannot distribute the module under any license but the GPL, but MS probably did not intend their code to be distributed under any license but a restrictive one, so they can't distribute it under the GPL either. Thus they can't distribute it at all. I've seen the evidence concerning Linux kernel code, somehow I doubt that Samsung clean-roomed the MS portion of the code. The MS patent is of no effect in most countries and can be ignored for global purposes.

    Unless either the FSF or MS is willing to turn a blind eye (or in the case of MS, license their code under the GPL), Samsung cannot distribute this module. Since it also contains some Samsung original code, nobody else can distribute it either except as Warez. MS or Samsung would breach the GPL, any GPL release would breach any MS or Samsung copyrights. The MS patent is of no effect in most countries and can be ignored for global purposes.

    In other words, we've got a copyright thicket on our hands. MS might want to remember that hurting Samsung benefits Crapple, their most dangerous adversary (not just competitor) in the mobile space. If MS chooses to permit use of this code under the GPL, they keep Crapple from sticking it to everyone else. Probably they aren't that smart. If they do, Samsung can then GPL the driver or withdraw it, if they withdraw it others will fork it.

    With luck, it will get reverse-engineered, and people will use the resulting "OpenExFAT" kernel module exactly the same way they use ffmpeg/avconv, and for exactly the same purpose: being able to use their cameras without buying Windoze or Crapple. One lawsuit against an individual over this, millions of people stop buying AVCHD cameras, maybe all digital cameras, maybe even new PC's. It would be the GIF lawsuits on steroids, could sink MS if people become afraid to own digital electronics at all.

    As for Samsung, they are now in a position to face copyright suits possibly from MS and the FSF at the same time, which would truly be a first. One possible resolution: Samsung must buy a transferrable patent license from MS, and buy transferrable rights to any MS code as well. Samsung must then use those rights to release the code under the GPL.

  8. #38
    Join Date
    Mar 2011
    Posts
    378

    Default

    Not sure if I understand everyting correctly:
    1) Samsung writes a driver that violates the GPL.
    2) They accidentially put it on GitHub.
    3) Somebody took the code and re-licensed it under the GPL.

    So basically samsung wrote a exFAT driver for us? Cause the GPL clearly sais the code has to be GPLed, too. If they sue Linux for using the code under the terms of the GPL Linux can simply say: "Wrong! You used GPL code! When this code breaks some agreements you have with MS it's your problem, go fix that."

    If all that is true could somebody please A) report to the FSF that the first report (against the GPLed driver) was invalid and B) report the original driver (from samsung) to the FSF as a GPL violation?

  9. #39
    Join Date
    Mar 2013
    Posts
    49

    Default

    Quote Originally Posted by TAXI View Post
    Not sure if I understand everyting correctly:
    1) Samsung writes a driver that violates the GPL.
    2) They accidentially put it on GitHub.
    3) Somebody took the code and re-licensed it under the GPL.

    So basically samsung wrote a exFAT driver for us? Cause the GPL clearly sais the code has to be GPLed, too. If they sue Linux for using the code under the terms of the GPL Linux can simply say: "Wrong! You used GPL code! When this code breaks some agreements you have with MS it's your problem, go fix that."

    If all that is true could somebody please A) report to the FSF that the first report (against the GPLed driver) was invalid and B) report the original driver (from samsung) to the FSF as a GPL violation?
    I literally explained how that isn't true on the previous page. Just because you violated the GPL doesn't mean you have to release the code. A GPL violation in most cases (and such is the accused case here) means that someone modified GPL code, compiled it, then distributed the binary without distributing the modified code as well, still licensed under the GPL. If it turns out that Samsung did that, then they *do not have to relicense their code and distribute it*. They can stop distributing the binary and pay a fine instead, for example. If they released the code under a different license, that doesn't mean it's suddenly released under the GPL, it means they violated the GPL. Relicensing under the GPL is not their only option, and unless/until they do, what rzrx did is illegal.

  10. #40
    Join Date
    Sep 2007
    Location
    Connecticut,USA
    Posts
    962

    Default

    Stuff like this gives FOSS a black eye when its perceived that closed proprietary code accidentally got leaked and some "developer" claims it is his and slaps a GPL license on it. Good for Samsung to go after him for that. Imagine if it was nVidia's proprietary driver code that got leaked instead...nVidia would go thermonuclear over that!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •