Stories
Slash Boxes
Comments

News for nerds, stuff that matters

First Legal Test of the GPL

Posted by CmdrTaco on Wed May 23, 2001 01:19 PM
from the its-only-a-matter-of-time dept.
Trepidity writes "In stark contrast to the plethora of false alarms recently, there's a pretty clear-cut case that Vidomi, a DVD ripping product by SloMedia, is composed of a great deal of code from VirtuaDub, a GPLd product. As SloMedia have refused all requests to either release their source or stop using the code, the developer is planning to file suit with the aid of the Free Software Foundation, in what could be the first legal test of the GPL's enforceability."
This discussion has been archived. No new comments can be posted.
First Legal Test of the GPL | Log In/Create an Account | Top | 524 comments (Spill at 50!) | Index Only | Search Discussion
Display Options Threshold:
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1) | 2 | 3 | 4 | 5 | 6 | 7
  • Re:Get a life by Anonymous Coward (Score:1) Wednesday May 23 2001, @09:58AM
  • Re:Huh? by Anonymous Coward (Score:1) Wednesday May 23 2001, @10:18AM
  • Re:No Credit for Nuke Either by Anonymous Coward (Score:1) Wednesday May 23 2001, @10:28AM
  • Re:Huh? by Anonymous Coward (Score:1) Wednesday May 23 2001, @10:59AM
  • Re:Huh? by Anonymous Coward (Score:1) Wednesday May 23 2001, @11:16AM
  • Not So Clear (was Re:Grey area) by Anonymous Coward (Score:1) Wednesday May 23 2001, @03:53PM
  • Another possibility by Anonymous Coward (Score:2) Wednesday May 23 2001, @09:47AM
  • Re:Huh? by Anonymous Coward (Score:2) Wednesday May 23 2001, @10:18AM
  • In this DMCA age... by Anonymous Coward (Score:2) Wednesday May 23 2001, @10:22AM
  • I'm confused on a couple of points. by Anonymous Coward (Score:2) Wednesday May 23 2001, @11:05AM
  • Re:A true test of the GPL by Anonymous Coward (Score:2) Wednesday May 23 2001, @11:08AM
  • Re:Update from Vidomi by Anonymous Coward (Score:2) Wednesday May 23 2001, @03:28PM
  • Re:easy: (Score:3)

    by Anonymous Coward on Wednesday May 23 2001, @09:31AM (#202754)
    Hint: what you did in college may not work in a copyright infringement lawsuit.
  • Re:Vidomi's position by Yarn (Score:2) Thursday May 24 2001, @12:06AM
  • Re:A true test of the GPL by Mark J Tilford (Score:1) Wednesday May 23 2001, @11:03AM
  • FUD, FUD, FUD (Score:3)

    by oGMo (379) on Wednesday May 23 2001, @10:51AM (#202757)

    While IANAL, it's obvious you're going on mostly hearsay, popular opinion, and misunderstandings (whether intentional or not). Let's see:

    Consider that libraries are CREATED with the intent that someone else will use the API and header files.

    Yes, they are, and it's not too unreasonable that the person who creates the library may wish it to only be used for Free Software purposes. Like, say, TrollTech. This fits perfectly with their intent. It's their code, if you want your own nice widget set that's closed, write it yourself, or pay TrollTech their due. Next up:

    I have never agreed with RMS (and Trolltech's) assumptions that dynamic linking makes something a derived work.

    Your opinions on the matter don't affect the matter, of course. Save this thought for later. Going on:

    It would certainly make ALL third party software vendors beholden to OS vendors, for example.

    Hmm, better not give software vendors any bright ideas here.

    Every program ever written under linux would be a derivative work of the kernel, which is GPL. Imagine if someday the kernel developers decided that all of a sudden ALL works ever developed under linux were actually GPL'd (not that this is a likely scenario, but it is a possible one if RMS is right).

    OK, this is complete tripe. The standard C/C++ library (glibc) is under the LGPL, Linus has made special provision for allowing use of the kernel interface however you'd like (allowing binary drivers for instance). Let me reiterate: this is just pure FUD.

    Fortunately, RMS is wrong. GPL libraries do not make all dynamically linked programs derivative works, and I expect this lawsuit to uphold that point (and sanity in general).

    Um, and RMS is wrong why? Remember, your opinions don't make it so. You certainly haven't shown it legally to not be the case. The only real reasoning you've shown is FUD. This assertion, therefore, doesn't seem to have a whole lot of weight behind it.

    Remember, RMS has quite a legal team. He very well understands what the license means, and it's been analyzed by many other legal teams (IBM, who could find no holes; MS, who is running scared). Let's see what the real lawyers with informed opinions have to say.

  • Re:VirtualDub is GPL, not LGPL by Chris Johnson (Score:2) Wednesday May 23 2001, @10:12AM
  • Re:Vidomi's position by The Man (Score:1) Wednesday May 23 2001, @05:23PM
  • Re:GPL is NOT LGPL by Ian Bicking (Score:2) Wednesday May 23 2001, @08:28PM
  • Re:Huh? by Ian Bicking (Score:2) Wednesday May 23 2001, @08:50PM
  • by Ian Bicking (980) <ianbNO@SPAMcolorstudy.com> on Wednesday May 23 2001, @09:09PM (#202762) Homepage
    There's also a possible:
    4) Pay Avery Lee to license his work to SloMedia (assuming he holds sole copyright).

    It's pretty common that Free Software programmers write proprietary code for a living. And many of them would be willing to provide non-GPL licenses for money. I believe this has been a major source of revenue for Cygnus (don't know what it is now that Redhat owns them). And it's only fair.

  • Re:Vidomi's position by Danse (Score:1) Wednesday May 23 2001, @02:21PM
  • Re:Vidomi's position by Danse (Score:2) Wednesday May 23 2001, @10:22AM
  • by Danse (1026) on Wednesday May 23 2001, @11:20AM (#202765)

    Nope, you got it wrong. It is GPL proponents who are outraged at corporation for their EULAs and yet they are using the very same methods.

    Not even close. You can do whatever you want with a GPL'd program as long as it doesn't violate copyright law. You don't even have to agree to the license. However, if you'd like to use the GPL'd code in a program of your own, then you can do so by agreeing to the GPL which states that you must release the source code to the resulting derivative program as well.

    Contrast that with standard industry EULAs which routinely take away all sorts of rights such as publishing benchmarks, using the program in a way that criticizes the maker, making backup copies, running the software on more than one machine, etc. It becomes quite ridiculous. There is no real comparison between the GPL and EULAs. At least with GPL'd software you can use it without having to agree to such a ridiculous licensing agreement. If these corporations put this many restrictions on simply USING their software, imagine what kinds of restrictions they would put on using their source code!

  • Re:If this claim stands, then "Bye-bye Be!" by demon (Score:1) Wednesday May 23 2001, @07:14PM
  • Re:GPL, Distribution, Intent and Spirit by bhurt (Score:2) Wednesday May 23 2001, @12:39PM
  • Why Not Money? by Omega (Score:1) Wednesday May 23 2001, @10:21AM
  • GPL vs. L-GPL by Omega (Score:1) Wednesday May 23 2001, @10:35AM
  • Re:Vidomi depends on GPL'ed libraries by jd (Score:2) Wednesday May 23 2001, @10:14AM
  • Re:What's with the GNU icon? by jaraxle (Score:1) Wednesday May 23 2001, @11:57AM
  • The same company violates MS's copyright as well by Andy Tai (Score:1) Wednesday May 23 2001, @10:19AM
  • Re:This is just what the FSF has been waiting for. by Jason Earl (Score:2) Thursday May 24 2001, @08:28AM
  • by Jason Earl (1894) on Wednesday May 23 2001, @10:57AM (#202774) Homepage

    The FSF has been waiting patiently for a violation of the GPL by a company small enough that they have a good chance of actually winning the case. Unfortunately they aren't the owners of the copyrighted software, but apparently the author of VirtualDub is willing to let the FSF do most of the heavy lifting. Once they win a case then they have precedent supporting their view of the GPL.

    Hopefully this actually goes to court, and hopefully the FSF get the win that they have been waiting for. Otherwise we are in for some interesting times on the Free Software front.

  • Ahem... by slothbait (Score:2) Wednesday May 23 2001, @12:41PM
  • Re:Vidomi's position by cpt kangarooski (Score:1) Wednesday May 23 2001, @11:31AM
  • Re:Don't go to court! by cpt kangarooski (Score:1) Wednesday May 23 2001, @11:37AM
  • Re:Removing vs. Opening by spitzak (Score:2) Thursday May 24 2001, @08:43AM
  • Re:GPL can probably be circumvented by spitzak (Score:2) Thursday May 24 2001, @09:03AM
  • Re:I'm confused on a couple of points. by spitzak (Score:2) Thursday May 24 2001, @09:25AM
  • Re:Why Not Money? by Dastardly (Score:1) Wednesday May 23 2001, @12:35PM
  • Re:Grey area (Score:3)

    by Dastardly (4204) on Wednesday May 23 2001, @12:20PM (#202782)
    Actually dynamic linking to GPL code does make the linking program based on that code. There is no grey area when is come to the GPL. Linking in pretty much any form is a violation fo the GPL. Consider this... Let's say I take some GPL software break it down into various components, compile those as dynamically linked libraries, then write my own proprietary application that links to those libraries I created from GPL code. This has to be a GPL violation otherwise there is no GPL. It is possible that a court might decide this, but I would not consider that a good thing. Secondly, the existence of the LGPL implies that dynamic linking is a violation of the GPL, otherwise why would the LGPL exist. Dastardly
  • Funny thing about that by crisco (Score:2) Wednesday May 23 2001, @12:39PM
  • Re:VirtualDub is GPL, not LGPL by johnnyb (Score:2) Wednesday May 23 2001, @10:08AM
  • Re:VirtualDub is GPL, not LGPL by johnnyb (Score:2) Wednesday May 23 2001, @10:11AM
  • Re:OT: Slashdot Submission process... by Johann (Score:1) Wednesday May 23 2001, @09:49AM
  • More info... (Score:5)

    by Uruk (4907) on Wednesday May 23 2001, @09:28AM (#202787)
    You can find more information about this at Advogato [advogato.org] where one of the guys involved in this posts about his experiences.

    It also contains some technical evidence as to which functions were lifted and how they know.

  • Ha ha ha! by jimdesu (Score:2) Wednesday May 23 2001, @09:30AM
  • Legal vs. GPL? by TBone (Score:2) Wednesday May 23 2001, @10:09AM
  • Re:No harm intended/done? by TBone (Score:2) Wednesday May 23 2001, @10:21AM
  • by John Fulmer (5840) on Wednesday May 23 2001, @11:29AM (#202791)
    I really dislike extremely misleading articles like this. Avery Lee said that he is asking for HELP to enforce the GPL with Vidomi, not that he's filing suit. There are other forms of help that do not necessarily involve litigation.

    I've been a VirtualDub supporter/contributor (financially) for a couple of years. It's a great program, and Avery, what I know of him, is a pretty quality guy. Is he thinking of suing Vidomi? I don't know. And neither does anyone else, from the references posted in this article.

    I, and I assume others, submitted a MUCH less inflamitory, more accurate entry. I wonder why THEY weren't used instead? It's this sensationalism in article selection and lack of even basic fact checking that really makes me wonder about value of Slashdot as a 'news' portal.

    jf
  • Re:a GPL DLL issue I once pondered by ZxCv (Score:1) Wednesday May 23 2001, @10:46PM
  • I had a (somewhat) similar issue - it gets murky by Booker (Score:2) Wednesday May 23 2001, @10:21AM
  • Re:Vidomi's position by Zagadka (Score:1) Wednesday May 23 2001, @11:06AM
  • Re:Vidomi's position by Zagadka (Score:2) Wednesday May 23 2001, @02:10PM
  • Programs that use GPL'd DLLs/libs/modules/etc by Vic (Score:1) Wednesday May 23 2001, @10:13AM
  • Re:Oh look! More /. GPL alarmism by bstadil (Score:1) Wednesday May 23 2001, @09:41AM
  • Re:Grey area by mandolin (Score:1) Wednesday May 23 2001, @01:14PM
  • Re:Vidomi's position by Just Another Perl Ha (Score:1) Wednesday May 23 2001, @11:00AM
  • Huh? (Score:5)

    by mcelrath (8027) on Wednesday May 23 2001, @09:26AM (#202800) Homepage
    Looks like (relevant parts of) the source code is available right here [vidomi.com]. Could be a short case.

    --Bob

  • Re:No harm intended/done? by The Swedish Chef (Score:1) Thursday May 24 2001, @02:51AM
  • Re:DEADLY VIRUS! by ergo98 (Score:1) Wednesday May 23 2001, @11:21AM
  • Re:GPL code is proprietary by ergo98 (Score:1) Wednesday May 23 2001, @11:54AM
  • Re:DEADLY VIRUS! by ergo98 (Score:1) Wednesday May 23 2001, @12:38PM
  • Re:Microsoft must be drooling by ergo98 (Score:1) Wednesday May 23 2001, @12:58PM
  • Re:DEADLY VIRUS! by ergo98 (Score:1) Wednesday May 23 2001, @01:52PM
  • Re:Vidomi's position by ansible (Score:2) Wednesday May 23 2001, @10:18AM
  • by ansible (9585) on Wednesday May 23 2001, @10:32AM (#202808) Homepage Journal

    You might be able to get away with such a Perl script hack in college, but not in the real world.

    Changing variable names won't substantially change the binary code the code compiles into.

    Even switching compilers and optimization levels doesn't give you much obfuscation. Besides, how many viable compilers are there for any particular platform? It's not hard to try them all.

    To substantially change the generated code (in a hard to detect fashion), you need to change the fundamental structure. And then you need to be careful and not introduce new bugs. And you still have to test it a lot. You might as well write it from scratch.

    Try running 'objdump --disassemble /bin/ls' on your Linux system for yucks.

  • Re:DLLs and COM (Score:3)

    by landley (9786) on Wednesday May 23 2001, @09:36PM (#202809) Homepage
    >What if I take GPL'ed code and create a COM
    >component out of it. I releast the full source.

    >3 months later, I decide to start writing a
    >closed-source app and I find the component would
    >be useful.

    >Should my application now be covered under the
    >GPL?

    You're the author. Your rights to use the code do not originate from a license granting you access to somebody else's copyright. You OWN the copyright.

    A license you grant delegating some of your rights to other people doesn't limit YOUR rights to your own work.

    Unless, of course, you sign over the copyright. But that's not what a license does. (If your employer paid you to create something, they may own the copyright due to their contract with you. And the FSF wants people to sign their copyrights over to them so they'll have a stronger position if they ever have to sue anybody. But that's a seperate issue.)

    Rob
  • Re:what if.... by DavidTC (Score:1) Wednesday May 23 2001, @10:46AM
  • Re:what if.... by DavidTC (Score:1) Thursday May 31 2001, @05:45AM
  • by Laxitive (10360) on Wednesday May 23 2001, @10:29AM (#202812) Journal

    I think you are wrong here, and fall into a common trap - the "technical" trap. You assume that any law is true to it's fullest technical interpretation, and that anything that can be interpreted as meeting those technical considerations, is affected by the law.

    But laws are human, and are meant to be interpreted by a human, and not a robot. A human (judge) can take a look at a situation, and make value judgements about the validity of the law or not.

    A good example is the Napster case - the judge decided that Napster was illegal because it was aiding in piracy. But all Napster is, is an online opt-in search engine/database. Does that mean that all search engines are now illegal, because they too can be used for piracy? No - because as humans, we can see the difference between Napster and search engines, even though someone can make a claim that Napster == Search engine.

    Likewise, it's kind of obvious that this product pretty much relies on GPL code to be useful. Perhaps it offers some token functionality without the GPL code, but, that functionality is pretty irrelavent. A human can take a look at it, and make a value judgement on the situation.

    Laws are not interpreted by turing machines (nor are they expressed in lambda calculus) - and that's a very good thing indeed.

    -Laxitive
  • Re:Vidomi's position by sab39 (Score:2) Wednesday May 23 2001, @10:01AM
  • Re:Crap. by hawkfan (Score:1) Wednesday May 23 2001, @01:36PM
  • Re:Huh? (Score:3)

    by jms (11418) on Wednesday May 23 2001, @11:21AM (#202815)
    Suppose you buy a piece of software and after installing and using it for a while, you decide to read the shrinkwrap agreement. You discover to your horror, that by running the software you agree to give your first-born son to the software company. They sue to get your son. Would the judge hand him over?

    That would depend on whether or not you live in a state that has passed the UCITA legislation.

  • Re:Huh? by grub (Score:2) Wednesday May 23 2001, @11:17AM
  • Re:A true test of the GPL by Omnifarious (Score:1) Wednesday May 23 2001, @11:18AM
  • Re:COM objects? by IntlHarvester (Score:2) Wednesday May 23 2001, @01:51PM
  • Re:A true test of the GPL by Outland Traveller (Score:2) Wednesday May 23 2001, @11:08AM
  • Re:Huh? by rhavyn (Score:1) Wednesday May 23 2001, @10:20AM
  • Re:Huh? by rhavyn (Score:1) Wednesday May 23 2001, @10:25AM
  • Re:Oh look! More /. GPL alarmism by rhavyn (Score:1) Wednesday May 23 2001, @10:42AM
  • Re:GPL, Distribution, Intent and Spirit by rhavyn (Score:1) Wednesday May 23 2001, @10:46AM
  • Re:A true test of the GPL by rhavyn (Score:1) Wednesday May 23 2001, @10:55AM
  • Re:Inquiry? by rhavyn (Score:1) Wednesday May 23 2001, @10:59AM
  • Re:Huh? by rhavyn (Score:2) Wednesday May 23 2001, @10:28AM
  • Re:VirtualDub and IP by Samrobb (Score:1) Wednesday May 23 2001, @12:43PM
  • Re:I had a (somewhat) similar issue - it gets murk by Samrobb (Score:2) Wednesday May 23 2001, @05:31PM
  • Re:It's funny... by MSG (Score:2) Wednesday May 23 2001, @10:22AM
  • Re:It's funny... by MSG (Score:2) Wednesday May 23 2001, @10:09PM
  • Re:It's funny... by MSG (Score:2) Wednesday May 23 2001, @10:15PM
  • Re:It's funny... (Score:3)

    by MSG (12810) on Wednesday May 23 2001, @10:47PM (#202832)
    does that mean ... you're bound by joe schmoe license that i invented c under??

    Yes. If you design a language, and place a restrictive license on it, then the people who use *your product* are bound by your license. (Although, some people may be legally allowed to reverse engineer the software) Additionally, if your language is an open specification, and you write documentation for the language that includes example code, then you may place a license on those examples. (A *lot* of docs are written this way, unfortunately) People who read your documentation are bound by the license of the example code.

    If i read the source code to... diablo 2... can blizzard bust my balls cause i figured out how they did something and put it into my game?
    Yes. Yes they can. That's what copyright protects. That's why proprietary software engineers are generally required to stay the hell away from Free Software code, and Free Software engineers don't read proprietary code.

    As an example, you won't find the Wine project accepting code from former Microsoft employees, because if those people violate the license used by their former employer, Wine as a whole may be judged to be in violation.
  • Re:Huh? by Erik Hensema (Score:1) Wednesday May 23 2001, @10:14AM
  • Re:Huh? by Erik Hensema (Score:1) Thursday May 24 2001, @02:10AM
  • Re:VirtualDub is GPL, not LGPL by um... Lucas (Score:1) Wednesday May 23 2001, @12:17PM
  • Re:VirtualDub is GPL, not LGPL by um... Lucas (Score:1) Wednesday May 23 2001, @04:14PM
  • Re:Get a life by um... Lucas (Score:1) Wednesday May 23 2001, @04:30PM
  • Re:Vidomi's position by dap24 (Score:1) Wednesday May 23 2001, @03:51PM
  • Re:Vidomi's position by dap24 (Score:1) Wednesday May 23 2001, @03:55PM
  • Re:GPL can probably be circumvented by ethereal (Score:1) Wednesday May 23 2001, @11:42AM
  • Re:Not to MLP, but... by SmittyTheBold (Score:1) Thursday May 24 2001, @12:19AM
  • Derivative Works (Score:3)

    by Sloppy (14984) on Wednesday May 23 2001, @04:20PM (#202842) Homepage Journal

    The header files describing the libraries had to be used to produce the application linked (dynamically) to it and so their application forms a derivative work.

    This is a horrible implication for Free Software (and non-monopoly commercial software too), as it could effectively allow someone to use IP laws to protect an interface. This must not be allowed to happen! If judges agree with you on this point, Linux dudes may win the battle but they will lose the war.

    If interfacing to another piece of software causes your software to become a derived work of the other, then we're all in for a shitstorm of trouble, and this particular case will be small potatoes by comparison.


    ---
  • Re:Implications by Todd Knarr (Score:1) Wednesday May 23 2001, @11:02AM
  • Re:Vidomi's position by kinkie (Score:2) Wednesday May 23 2001, @01:15PM
  • Re:GPL, Distribution, Intent and Spirit by Royster (Score:1) Thursday May 24 2001, @04:59AM
  • by Royster (16042) on Wednesday May 23 2001, @10:01AM (#202846) Homepage
    Unfortunately this case, if it ever gets heard, will go up against one of the areas that the GPL is weakest -- dynamic linking.

    The question of what constitutes a "derivative work" in the case of software has not ben litigated to my knowledge. The GPL gets its strength from copyright law. Copyright law reserves several rights to the copyright holder among these are the creation of "derivative works".

    But Copyright law does not cover anything which is a "method of operation". It could be argued that an API exposed by a DLL or other dynamically linked object is a method of operation and not protected by copyright. Thus merely using a library routine might not be viewed as creating a derivative work.

    If this were not the case, we would be in a worse situation. Then MS, for example, could claim that any piece of software using services provided by a MS DLL is a derivative work. Similarly for any platform where routines are exposed through an API and where there isn't license text specifically allowing the use.
  • by Royster (16042) on Wednesday May 23 2001, @11:36AM (#202847) Homepage
    The header files describing the libraries had to be used to produce the application linked (dynamically) to it and so their application forms a derivative work.

    This is certainly the point of view in the GPL community, but it if far from clear whether the courts will endorse this view. Header files describe an API. An API is arguably a 'method of operation' in the context of copyright law and 'methods of operation' are not protected by copyright.
  • Is it independent? by bluestar (Score:1) Wednesday May 23 2001, @12:17PM
  • Microsoft must be drooling by bluestar (Score:1) Wednesday May 23 2001, @12:29PM
  • roughly how the GPL would sound in court by redd (Score:1) Wednesday May 23 2001, @01:27PM
  • by mattbee (17533) <matthew@bytemark.co.uk> on Wednesday May 23 2001, @10:43AM (#202851) Homepage
    Of course people are seizing onto the GPL's slippery definition of linking-- to add to the melting pot of opinions, the MySQL folk specify that you must pay them for a special GPL exception license if You have a commercial application that ONLY works with MySQL and ships the application with the MySQL server. This is because we view this as linking even if it is done over the network. e.g. any kind of reliance == linking. I'm not sure whether this makes the case weaker or stronger against dynamic linking as GPL-covered linking.
  • Um Something is really weird here by IceFox (Score:2) Wednesday May 23 2001, @10:46AM
  • Re:Um Something is really weird here by IceFox (Score:2) Wednesday May 23 2001, @01:33PM
  • Re:Um Something is really weird here by IceFox (Score:2) Wednesday May 23 2001, @04:07PM
  • Re:scenarios by Shadowlion (Score:1) Wednesday May 23 2001, @10:26AM
  • Re:A true test of the GPL by Shadowlion (Score:2) Thursday May 24 2001, @06:43AM
  • by Shadowlion (18254) <cdc@gis.net> on Wednesday May 23 2001, @10:34AM (#202857) Homepage
    Every program ever written under linux would be a derivative work of the kernel, which is GPL.

    *bzzzzt!*

    Programs under GNU/Linux are not linked against the kernel. Programs under GNU/Linux are linked against the GNU C library, which is LGPL'ed (Library GPL). The LGPL says that you can link non-GPL'ed software against the C libraries and be OK, which allows for commercial, closed-source applications to be developed for GNU/Linux.

    So even if the GPL is held to be enforceable against libraries, commercial apps are safe and can continue to co-exist because the LGPL'ed C library gives them the exception they need.

    --
  • Re:Lawsuit? Probably not. by dkixk (Score:1) Thursday May 24 2001, @01:39AM
  • Re:Microsoft must be drooling by B.D.Mills (Score:2) Wednesday May 23 2001, @03:01PM
  • Re:Vidomi's position by Anomie-ous Cow-ard (Score:1) Wednesday May 23 2001, @12:11PM
  • The $75,000,000 question by JabberWokky (Score:2) Wednesday May 23 2001, @11:56PM
  • Re:Huh? (Score:3)

    by JabberWokky (19442) <slashdot.com@timewarp.org> on Wednesday May 23 2001, @10:35AM (#202862) Homepage Journal
    Yes, but remember that linking to a dll occurs at runtime not at compile time, so the people linking to the dll are the end users, not Vidomi. I don't see how Vidomi can be held accountable for actions that end users might take with their program.

    Ah, but you fail to realize that lawyers are not programmers. Most of the time, this is a bad thing, but this may very well be a good thing in this case.

    They won't care if it's Corba, DCOM, or staticly compiled. Hell, they don't even know what that means. All the law cares about is: "Did this action violate the spirit of the license". Sure, they'll try and interpret the license every which way by pulling out each relevant sentence and dissecting the meaning of every word, but it won't be a technological analysis. Technical, yes, but not technological.

    They will look, see what the license says:

    2.b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

    (later in 2) These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.

    I boldfaced the important bits. Are they distributing a piece of software that is a work (as a whole) that is based on the GPLed software? If so, it must be GPLed, even if it's connecting to the code on a seperate server... as a higher level interpretation (which is my own, and not legal), even if the GPL code were to reside on a server: if the app requires the GPLed code as a basic part of its core functionality, then the GPL applies to the work as a whole.

    --
    Evan YMMV IANAL, but I used to work with them.

  • GPL is NOT LGPL (Score:3)

    by Hugonz (20064) <hugonz@@@gmail...com> on Wednesday May 23 2001, @10:09AM (#202863) Homepage
    I believe he FSF saw this coming from the very beginning, that's why they came up with the LGPL. Vidori claims that they have released all the changes they've made to the GPLed code itself (the DLL) but those are exactly the term of the LGPL, no the bare GPL.

  • by sweetooth (21075) on Wednesday May 23 2001, @09:33AM (#202864) Homepage
    posting this? http://www.vidomi.com/index.php?artid=6#news [vidomi.com]
    Also for those don't want to load the page they have three interesting news bits, the first is titled "Credit where Credit is due. They acknowledge the following sites
    http://hiroko.ee.ntu.edu.tw/
    http://www.freeamp.org/
    http://www.flaskmpeg.net
    http://www.virtualdub.org/
    http://www.mp3dev.org/mp3/

    Then they have a link to the source at this url http://www.vidomi.com/download.php?op=mydown&did=3 [vidomi.com]
    The next item reports that they will no longer allow download of the GPL'd items with the non GPL'd items, and that they will shortly release a GPL kit.
    Where's the problem?
  • Re:Get a life by Phill Hugo (Score:1) Wednesday May 23 2001, @09:50AM
  • Re:No harm intended/done? by Phill Hugo (Score:2) Wednesday May 23 2001, @09:55AM
  • Re:At least it's a Small(er) company by Phill Hugo (Score:2) Wednesday May 23 2001, @10:05AM
  • Re:GPL, Distribution, Intent and Spirit by Phill Hugo (Score:2) Wednesday May 23 2001, @12:30PM
  • Re:A true test of the GPL by Phill Hugo (Score:2) Tuesday May 29 2001, @10:43AM
  • Re:A true test of the GPL by Phill Hugo (Score:2) Tuesday May 29 2001, @10:49AM
  • by Phill Hugo (22705) on Wednesday May 23 2001, @09:40AM (#202871) Homepage
    It looks like Vidomi are trying to remove themselves from this while still allowing the code to be used with their closed source product. By decoupling the GPLd code into dynamically shared libraries and distribtuting them seperately, the GPL violation is less obvious but still quite present.

    The header files describing the libraries had to be used to produce the application linked (dynamically) to it and so their application forms a derivative work. This is why GPL shared libraries are not linked against elsewhere and is why the LGPL was produced.

    Even if Vidomi produced an middle man DLL, that middle man DLL would be subject to the GPL for the same reason and anything linked against it would also be. However, it would be interesting as it would make Vidomi responsible for prosecuting any violation of their proxy DLL. I still think most courts would not favour their case after the situation was clearly explained.

    It is really a true shame that Vidomi have put their own interest in front of the Avery's, it shows them in a very bad light, although it would be interesting to see the GPL subject to court-time but I think the outcome is predictable.

    I can only suggest they save their time in court, GPL their application and get on with writing applications that people want to use, that is what its all about, right Vidomi?
  • by Phill Hugo (22705) on Wednesday May 23 2001, @10:15AM (#202872) Homepage
    > whether dynamic linking to a library makes something a derivative work in the copyright sense.

    I think it does since the libraries header files are used in the subsequent work. The courts will always take into account NON COURT precedent in the absence of previous similar findings and there they would run into the NeXT Object C compiler based on GCC (NeXT backed down, its GPL now!), the QT problem, the absence of linking to GPL libs on Linux/UNIX systems, the intent of the LGPL to specifically permit this (this alone should tell you the intent of the GPL is NOT to permit it)

    You statement about making 3rd party developers beholded to OS vendors is wrong. They already ARE beholden to them but most sucessful OS's permit free linking to their application without license propegation to the result. The LGPL allows this too and is why glibc and mesa are LGPLd.

    The interface between kernel and application is via an LGPL application and Linus specifically permits binary only drivers in Linux (basically making the header files required to use them effecticely LGPL)

    You are incorrect on a lot of counts here and you're not alone.
  • Re:Vidomi's position by DeathBunny (Score:1) Wednesday May 23 2001, @11:18AM
  • Re:Vidomi's position by DeathBunny (Score:2) Wednesday May 23 2001, @10:31AM
  • Re:Vidomi's position by DeathBunny (Score:2) Wednesday May 23 2001, @10:42AM
  • Re:Vidomi's position by DeathBunny (Score:2) Wednesday May 23 2001, @10:50AM
  • Re:Vidomi's position by DeathBunny (Score:2) Thursday May 24 2001, @05:45AM
  • by DeathBunny (24311) on Wednesday May 23 2001, @11:08AM (#202878) Homepage
    >It might seem tired and overused argument but >GPL really does force people into GPL

    That's just pure stupidity. No one is forcing you to use GPL code. As I've said before, anytime you use ANY copywrited material (outside of fair use) in a derived work, you have agreed to abide by the license given to you by the original copywrite holder. If you don't like that license, don't use THAT code.

    There are three easy ways Vidomi could have avoided this whole issue:
    1) Use VirtuaDub and release thier software under a GPL compatible license.
    2) Don't use VirtuaDub, write thier own code.
    3) Look for and use code whose license *is* compatible to Vidomi's license, use that.

    Seems simple enough.

    >license that is not much different than "evil" >they are trying to fight.

    That's ridiculous. The GPL prevents you from using GPL code in a proprietary application. That's not evil, that's *the* primary feature of the GPL.

    If you don't *want* that feature, don't license your code under the GPL.

    And as always, whatever license you use will have to be compatible to the licenses on any other code you have incorporated into your program. That's not specific to the GPL, that applies to everything that isn't Public Domain!
  • Re:Huh? by Darren.Moffat (Score:2) Wednesday May 23 2001, @10:21AM
  • Re:GPL is NOT LGPL by maw (Score:1) Wednesday May 23 2001, @06:32PM
  • Re:Vidomi's position by Fluffy the Cat (Score:2) Wednesday May 23 2001, @11:11AM
  • Re:Vidomi's position by Fluffy the Cat (Score:2) Wednesday May 23 2001, @12:22PM
  • Re:Vidomi's position by Fluffy the Cat (Score:2) Wednesday May 23 2001, @12:53PM
  • Re:Real Men read binaries all the time by jonathanclark (Score:1) Wednesday May 23 2001, @12:05PM
  • EXE versus DLL. (Score:3)

    by jonathanclark (29656) on Wednesday May 23 2001, @12:02PM (#202885) Homepage
    So what's the difference between linking with a DLL and forking a processes that is GPLed - you can achieve the same levels of functionality. In Windows, the DLL and EXE file format are exactly the same. So renaming a .EXE to .DLL would be a violation of GPL? Taking this argument to it's logical extreme, any program that can create a pipe to another process could be in violation of GPL.

  • Re:EXE versus DLL. (Score:3)

    by jonathanclark (29656) on Wednesday May 23 2001, @04:39PM (#202886) Homepage
    Ohh.. I can't resist a troll...


    executables have stripped relocation tables, you can only load them to address 0x400000


    That's not true. Most exe's have relocation information. You can strip it out, but it's pretty rare that EXEs have this information removed. Even if they did it just means the process that loads them needs to be loaded at a different base address. There is nothing special about 0x400000, many exes have a base address other than that. Notepad.exe uses 0x1000000 for example (under w2k). Use the program "rebase" that comes with W2k to change the base address to whatever you like. You can load an EXE without relocation information as DLL. Also DLLs don't have to have relocation information assuming they are not going to conflict.

    even if the relocation table would be present, the OS doesn't permit loading & dynamically linking against an executable. It was originally possible, but the changed because it can cause security problem.

    Try it yourself :

    HMODULE m=LoadLibrary("c:\\winnt\\system32\\notepad.exe");

    Amazing!! :)

  • Re:Automatic modification of GPL'd code possible? by Dr. Manhattan (Score:1) Wednesday May 23 2001, @11:06AM
  • Re:Inquiry? by scruffy (Score:2) Wednesday May 23 2001, @04:46PM
  • Re:Huh? by bdowne01 (Score:1) Wednesday May 23 2001, @09:43AM
  • Almost... by schon (Score:1) Wednesday May 23 2001, @11:30AM
  • Re:US Copyright Law 102b by rking (Score:1) Wednesday May 23 2001, @04:09PM
  • Re:OT: Slashdot Submission process... by Covener (Score:1) Thursday May 24 2001, @05:31PM
  • Re:VirtualDub and IP by SpinyNorman (Score:1) Wednesday May 23 2001, @11:17AM
  • Re:GPL v. LGPL by SpinyNorman (Score:1) Wednesday May 23 2001, @11:21AM
  • Re:The GPL and Linking by SpinyNorman (Score:2) Wednesday May 23 2001, @10:18AM
  • Re:GPL v. LGPL by SpinyNorman (Score:2) Wednesday May 23 2001, @10:21AM
  • Re:VirtualDub and IP by SpinyNorman (Score:2) Wednesday May 23 2001, @10:39AM
  • Re:VirtualDub and IP by SpinyNorman (Score:2) Thursday May 24 2001, @04:01AM
  • Re:A true test of the GPL by Platinum Dragon (Score:2) Wednesday May 23 2001, @10:49AM
  • Re:Vidomi's position by Platinum Dragon (Score:2) Wednesday May 23 2001, @11:07AM
  • Re:GPL code is proprietary by Zurk (Score:1) Wednesday May 23 2001, @12:47PM
  • Re:Microsoft must be drooling by Zurk (Score:1) Wednesday May 23 2001, @01:11PM
  • A Question of Intent by ttfkam (Score:1) Wednesday May 23 2001, @12:00PM
  • GPLed headers aren't in the main program by ttfkam (Score:1) Wednesday May 23 2001, @12:07PM
  • Expectation of abilities by ttfkam (Score:1) Wednesday May 23 2001, @12:26PM
  • Re:Oh look! More /. GPL alarmism by ttfkam (Score:1) Wednesday May 23 2001, @12:37PM
  • Re:A true test of the GPL by ttfkam (Score:1) Wednesday May 23 2001, @12:51PM
  • Re:It's funny... by ADRA (Score:1) Wednesday May 23 2001, @10:19AM
  • Re:Huh? by ADRA (Score:1) Wednesday May 23 2001, @10:29AM
  • Re:GPL, Distribution, Intent and Spirit by csbruce (Score:1) Wednesday May 23 2001, @11:08AM
  • Re:GPL, Distribution, Intent and Spirit by csbruce (Score:1) Wednesday May 23 2001, @12:55PM
  • No Credit for Nuke Either by hndrcks (Score:2) Wednesday May 23 2001, @09:54AM
  • MOD PARENT UP. by Malcontent (Score:2) Wednesday May 23 2001, @09:01PM
  • Re:Inquiry? by firewood (Score:1) Wednesday May 23 2001, @12:38PM
  • Re:VirtualDub and IP by prizog (Score:2) Wednesday May 23 2001, @11:04AM
  • Re:Inquiry? by throx (Score:1) Wednesday May 23 2001, @11:31AM
  • GPL fails when dealing with components? by throx (Score:1) Thursday May 24 2001, @08:29AM
  • Inquiry? (Score:5)

    by EasyTarget (43516) on Wednesday May 23 2001, @10:10AM (#202918) Homepage Journal
    What, when it comes to it, is the difference between a Library, and a complete executable called from within another application. Does the GPL make a clear distinction?

    I ask since the product I do support for uses perl and gzip as part of it operation. We ship -unmodified- copies of these in binary form and supply the source (simply a copy of the relevent release's source from the Gnu sites) on demand. We call gzip from within some of our code to compress data 'on the fly', and we use perl everywhere, from install scripts through to cron jobs, and a whole bunch of perl utilities we ship that complement the primary product.

    My understanding (both from our legal bods, and from stuff I have seen in slashdot discussions) is this is quite legal, and the GPL actually is written to accomodate this sort of use (we use the GNU stuff in it's entirity, without extending it's functionality).

    But I fail to see a real difference between calling these utilities as standalone executable, and calling something within a library, surely the net effect is the same?


    EZ
  • Re:Vidomi's position by wct (Score:1) Wednesday May 23 2001, @04:55PM
  • Re:Huh? by Pedersen (Score:1) Wednesday May 23 2001, @09:39AM
  • Re:Get a life by Pedersen (Score:1) Wednesday May 23 2001, @09:48AM
  • Re:GPL v. LGPL by hattig (Score:1) Wednesday May 23 2001, @11:51AM
  • Re:Dangerous argument by hattig (Score:1) Wednesday May 23 2001, @11:56AM
  • Re:Huh? by hattig (Score:1) Wednesday May 23 2001, @12:05PM
  • How do you classify GPL plugins then? by hattig (Score:2) Wednesday May 23 2001, @10:22AM
  • Re:Huh? by hattig (Score:2) Wednesday May 23 2001, @11:13AM
  • Re:Huh? by hattig (Score:2) Wednesday May 23 2001, @11:16AM
  • Re:Programs that use GPL'd DLLs/libs/modules/etc by hattig (Score:2) Wednesday May 23 2001, @11:44AM
  • Grey area by Hard_Code (Score:2) Wednesday May 23 2001, @12:04PM
  • Re:Lately? by QuantumG (Score:1) Wednesday May 23 2001, @10:17AM
  • Re:Lately? by QuantumG (Score:1) Wednesday May 23 2001, @10:46AM
  • Re:The GPL and Linking by QuantumG (Score:1) Wednesday May 23 2001, @11:32AM
  • too bad you're not Vidomi by QuantumG (Score:1) Wednesday May 23 2001, @11:40AM
  • Re:GPL, Distribution, Intent and Spirit by QuantumG (Score:1) Wednesday May 23 2001, @12:02PM
  • Re:What did they do? by QuantumG (Score:1) Wednesday May 23 2001, @05:00PM
  • Re:OT: Slashdot Submission process... by QuantumG (Score:1) Wednesday May 23 2001, @05:23PM
  • Re:(OT)You would like Kuro5hin by QuantumG (Score:1) Wednesday May 23 2001, @05:27PM
  • Re:Good for the free software foundation.... by QuantumG (Score:1) Wednesday May 23 2001, @05:31PM
  • it's like a car accident.. by QuantumG (Score:1) Wednesday May 23 2001, @05:36PM
  • Re:We should thank these ... people by QuantumG (Score:1) Wednesday May 23 2001, @05:57PM
  • Re:Programs that use GPL'd DLLs/libs/modules/etc by QuantumG (Score:1) Wednesday May 23 2001, @06:01PM
  • Re:Ha ha ha! by QuantumG (Score:1) Wednesday May 23 2001, @06:11PM
  • Re:Lawsuit? Probably not. by QuantumG (Score:1) Wednesday May 23 2001, @06:22PM
  • Re:Update from Vidomi by QuantumG (Score:2) Wednesday May 23 2001, @04:37PM
  • Re:Not to MLP, but... by Kwil (Score:1) Wednesday May 23 2001, @07:07PM
  • Re:GPL can probably be circumvented by Keeper (Score:2) Wednesday May 23 2001, @07:41PM
  • Re:Huh? by Velox_SwiftFox (Score:2) Wednesday May 23 2001, @01:11PM
  • Re:A true test of the GPL by steveholden (Score:1) Thursday May 24 2001, @08:58AM
  • Re:A true test of the GPL by steveholden (Score:1) Thursday May 24 2001, @09:08AM
  • Re:Update from Vidomi by Johan Veenstra (Score:1) Wednesday May 23 2001, @03:45PM
  • Re:Get a life by CaptSwifty (Score:1) Wednesday May 23 2001, @09:58AM
  • Re:Huh? by ffatTony (Score:1) Wednesday May 23 2001, @11:26AM
  • Re:Vidomi's position by PurpleBob (Score:1) Wednesday May 23 2001, @11:06AM
  • Useful insight.. by Convergence (Score:2) Wednesday May 23 2001, @02:52PM
  • Re:A true test of the GPL by sconeu (Score:2) Wednesday May 23 2001, @11:44AM
  • Re:Inquiry? by etrnl (Score:1) Wednesday May 23 2001, @10:43AM
  • Re:scenarios by etrnl (Score:1) Wednesday May 23 2001, @11:13AM
  • Re:In this DMCA age... by etrnl (Score:2) Wednesday May 23 2001, @10:57AM
  • by Ted V (67691) on Wednesday May 23 2001, @09:29AM (#202959) Homepage
    At least it's a company like SloMedia, not Sony or Microsoft. I doubt SloMedia has the same legal resources as Cisco, for example. IANAL;TINLA (I am not a lawyer; This is not legal advice), but it seems like a common legal technique is testing your suits in easy cases to set a tentative precident, which gives you more power in later cases. So in the grand scheme of things, this is probably good for the GPL.

    -Ted
  • Re:Ahem... by sreeram (Score:1) Wednesday May 23 2001, @06:35PM
  • GPL Question by RSwan (Score:2) Thursday May 24 2001, @03:38PM
  • Re:Vidomi's position by jmccay (Score:1) Wednesday May 23 2001, @01:35PM
  • Removing vs. Opening by lowLark (Score:2) Wednesday May 23 2001, @09:28AM
  • Re:Huh? by dimator (Score:2) Wednesday May 23 2001, @10:08AM
  • What did they do? by scharkalvin (Score:1) Wednesday May 23 2001, @10:09AM
  • Re:Vidomi depends on GPL'ed libraries by OmniGeek (Score:1) Wednesday May 23 2001, @11:31AM
  • Vidomi depends on GPL'ed libraries by OmniGeek (Score:2) Wednesday May 23 2001, @09:53AM
  • by Vecna! (74330) on Wednesday May 23 2001, @10:04AM (#202968)
    Unfortunately for the principles at stake in this case, I tend to agree with the publisher of the software using the GPL'd code in DLLs that they're not violating the terms of the GPL license.

    The GPL is, at heart, a copyright license. It controls the right of a person to use, copy, modify and distribute a work that is copyright. As such, the scope of it's effectiveness is bounded by the scope of the copyright laws.

    The crux of the GPL's leverage is it's application in derivative works. If a portion of sourceode licensed using the GPL is combined with other material into a whole, the resulting work will be a derivative work, and the GPL's terms will apply to that work. I seriously doubt that a court will apply any of the computer-science analysis to the code, meaning that terms like "execution thread" and "shared memory" will be ignored. The court will ask "is the GPL'd code combined into a file with the non-GPL'd file", and the answer will be "no". From that point onward, I think the court is going to look very skeptically at the idea that run-time linking can create a derivative work.

    In the case of DLLs loaded at runtime, there is never a time when the GPL'd code is combined with the rest of the software to form a derivative work. Interaction with the DLL is in the form of interfaces where the operating system mediates the exchange of data between the calling program and the DLL, and the return of data from the DLL to the calling program. The two pieces of code are never combined into one whole program.

    Even if the author of the code in the DLL were to claim that the header files used to construct the DLL calls represented a copyright, and if used in the sourcecode for the calling program represented the creation of a derivative work, the company could simply replace the header files with ordinal values and value types in the description of the external function and call the functions in the DLL without even using the names of the functions or the names of the variables being passed, thus eliminating that source of copyright infringement.

    Let me give you another example. Using Windows, printer drivers are stored in DLLs. They are accessed by software at run-time to set printer settings, to rasterize output, and to peform various other kinds of error checking.

    If the GPL is held to apply to software that links at runtime to DLLs, it would mean that every Windows printer driver in the world would be required to conform to the GPL in order for GPL'd software to use them.

    It has been suggested that the exemption in the GPL for code distributed with the operating system covers printer drivers, but unfortunately many drivers are installed by users and are not a part of the operating system distribution.

    Here's an even more extreme example. If the model of "run time calling" is found to create derivative works comprised of the calling software and the target of the call, then anything using an RPC mechanism is also going to have to follow the rules of the GPL. Including the results of RPC calls made using HTTP. Meaning that GPL'd web browsers could only legally connect to GPL'd web servers.

    My opinion is that a court will not find the combination of a program using GPL'd code through run-time calls to a GPL'd library to form a derivative work. If it does not form a derivative work, the GPL cannot apply to the calling code. If it does not apply, then distributing a closed-source program that relies on run-time calls to GPL'd DLLs is going to be a legal method of circumventing the copyleft of the GPL.

    If the court does hold in that direction, it will essentially mean that the LGPL is a pointless license; anyone will probably be able to encapsulate GPL'd code in DLLs, and as long as they distribute that code in compliance with the GPL, distribute proprietary code that calls those DLLs in the same package. If that happens, the only benefit to using the LGPL will be to allow static linking at compile time to LGPL'd code, something so trivial to replace with a run-time link that the LGPL will become irrelevant.

    I am actually glad to see a court test of this issue, because it will help clarify the scope of the GPL and provide the first real road map for how copyleft licenses can be applied to derivative works. It's a valuable lesson to be learned, regardless of the copyleft in question.
  • Re:A true test of the GPL by selectspec (Score:2) Wednesday May 23 2001, @11:10AM
  • Re:Vidomi's position by selectspec (Score:2) Wednesday May 23 2001, @12:05PM
  • Re:COM objects? by selectspec (Score:2) Wednesday May 23 2001, @12:51PM
  • by selectspec (74651) on Wednesday May 23 2001, @10:27AM (#202972)
    From the GPL [gnu.org]:

    These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.

    What idiots. Clearly if you write non-derived modules and distribute them individually, such pieces are not under the GPL. Once you package it all up for distribution everything falls under the GPL. What an idiot. Time to get an attorney people or fork over the code!

  • Re:Vidomi's position by townmouse (Score:1) Thursday May 24 2001, @09:34AM
  • Lee's position by townmouse (Score:1) Thursday May 24 2001, @09:41AM
  • once it is released that's it right? by willij3 (Score:1) Wednesday May 23 2001, @01:49PM
  • Re:Did you bother to look at vidomi.com before.... by Eric Seppanen (Score:2) Wednesday May 23 2001, @09:46AM
  • A request for clarification regarding library use by Vicegrip (Score:2) Wednesday May 23 2001, @01:11PM
  • it's copyRIGHT by kevin805 (Score:2) Thursday May 24 2001, @03:33AM
  • Re:Update from Vidomi by gss (Score:1) Wednesday May 23 2001, @06:23PM
  • Re:GPL, Distribution, Intent and Spirit by Amokscience (Score:1) Wednesday May 23 2001, @10:13AM
  • Re:US Copyright Law 102b by RGRistroph (Score:1) Wednesday May 23 2001, @11:49AM
  • Re:I'm confused on a couple of points. by RGRistroph (Score:1) Wednesday May 23 2001, @12:35PM
  • Re:Implications by RGRistroph (Score:1) Wednesday May 23 2001, @12:48PM
  • Re:I'm confused on a couple of points. by RGRistroph (Score:1) Wednesday May 23 2001, @09:15PM
  • Re:US Copyright Law 102b by RGRistroph (Score:1) Thursday May 24 2001, @09:20AM
  • Re:I'm confused on a couple of points. by RGRistroph (Score:1) Thursday May 24 2001, @10:23AM
  • Re:Automatic modification of GPL'd code possible? by mOdQuArK! (Score:2) Wednesday May 23 2001, @11:39AM
  • Re:Your compiler violates the terms of my code! by rschwa (Score:1) Thursday May 24 2001, @10:08AM
  • Your compiler violates the terms of my code! by rschwa (Score:2) Wednesday May 23 2001, @11:19AM
  • Re:VirtualDub and IP by Ded Mike (Score:1) Thursday May 24 2001, @06:41AM
  • by Ded Mike (89353) on Wednesday May 23 2001, @11:11AM (#202991) Homepage
    RE: I'm not sure what law prevents you from reverse engineering a file format!
    It's called UCITA [ucitaonline.com] and it's very, very bad [badsoftware.com]. The CPSR, 26 States Attorneys-General, the ACM, and many others [cpsr.org] oppose it vigorously. You should, too.

    Because state reptilesentatives are more worried about their bribes^H^H^H^H^H^H...errr...campaign contributions and the falling tax-base, more than they are about your rights either as consumers or developers, and because the supporters of this unconstitutional law have such deep pockets, UCITA is slowly gaining ground against its opponents.

    From Richard Stallman's updated ' Why We MUST Fight UCITA [linuxtoday.com]'
    UCITA has another indirect consequence that would hamstring free software development in the long term -- it gives proprietary software developers the power to prohibit reverse engineering. This would make it easy for them to establish secret file formats and protocols, which there would be no lawful way for us to figure out.
    FIGHT THE POWER!!!! [linuxtoday.com]

  • I'll return that: Huh? by rakslice (Score:1) Wednesday May 23 2001, @08:09PM
  • Re:GPL can probably be circumvented by Dirtside (Score:2) Wednesday May 23 2001, @02:52PM
  • Re:Another interesting take on linking by magi (Score:1) Wednesday May 23 2001, @12:02PM
  • Re:A true test of the GPL by blakestah (Score:2) Wednesday May 23 2001, @10:51AM
  • Re:A true test of the GPL by blakestah (Score:2) Wednesday May 23 2001, @10:53AM
  • Re:A true test of the GPL by blakestah (Score:2) Wednesday May 23 2001, @10:58AM
  • Re:"Let's hope Vidomi is correct" ?? by blakestah (Score:2) Wednesday May 23 2001, @11:03AM
  • Re:A true test of the GPL by blakestah (Score:2) Wednesday May 23 2001, @11:10AM
  • US Copyright Law 102b by blakestah (Score:2) Wednesday May 23 2001, @11:29AM
  • Re:A true test of the GPL by blakestah (Score:2) Wednesday May 23 2001, @11:44AM
  • Re:US Copyright Law 102b by blakestah (Score:2) Wednesday May 23 2001, @01:11PM
  • Re:A true test of the GPL by blakestah (Score:2) Wednesday May 23 2001, @01:55PM
  • Re:Wrapping GPL code with CORBA by blakestah (Score:2) Thursday May 24 2001, @12:56PM
  • by blakestah (91866) <blakestah@gmail.com> on Wednesday May 23 2001, @09:57AM (#203005) Homepage
    This will be a test of the weakest portion of the GPL that is assumed to be defensible by the FSF - whether dynamic linking to a library makes something a derivative work in the copyright sense.

    For EVERYONE's sake, let's hope that Vidomi is correct and that dynamic linking does NOT make something a derivative work. Consider that libraries are CREATED with the intent that someone else will use the API and header files.

    I have never agreed with RMS (and Trolltech's) assumptions that dynamic linking makes something a derived work. It would certainly make ALL third party software vendors beholden to OS vendors, for example. Every program ever written under linux would be a derivative work of the kernel, which is GPL. Imagine if someday the kernel developers decided that all of a sudden ALL works ever developed under linux were actually GPL'd (not that this is a likely scenario, but it is a possible one if RMS is right).

    Fortunately, RMS is wrong. GPL libraries do not make all dynamically linked programs derivative works, and I expect this lawsuit to uphold that point (and sanity in general).
  • by blakestah (91866) <blakestah@gmail.com> on Wednesday May 23 2001, @10:34AM (#203006) Homepage
    > whether dynamic linking to a library makes something a derivative work in the copyright sense.

    I think it does since the libraries header files are used in the subsequent work. The courts will always take into account NON COURT precedent in the absence of previous similar findings and there they would run into the NeXT Object C compiler based on GCC (NeXT backed down, its GPL now!), the QT problem, the absence of linking to GPL libs on Linux/UNIX systems, the intent of the LGPL to specifically permit this (this alone should tell you the intent of the GPL is NOT to permit it)

    Companies backing down do not set legal precedent. That is the entire point of backing down.

    The QT problem has NEVER been addressed in court. They make the same assumptions RMS does - that dynamic linking makes something a derivative. In fact, I think RMS convinced them of it.

    The interface between kernel and application is via an LGPL application and Linus specifically permits binary only drivers in Linux (basically making the header files required to use them effecticely LGPL)

    When I create a binary and use system calls, I am dynamically linking to the kernel in exactly the same sense as I am linking to a dynamic library when I make a call to a function in that library. I have no necessity to use gcc to make the binary - there are other compilers available, and I can write the program in assembly and in theory create the same executable bit for bit. I don't NEED to include the header file - gcc may require it, but I can certainly code around it.

    This has never made sense to me - that ANY dynamic link library makes ALL linking programs derivative works. It seems to me that the entire purpose of creating a dynamic link library is exporting a PUBLIC API and header file that allows someone to create reasonably separate programs that can access a function.

    Functions, as even a cursory glance at copyright law will tell you, are not copyrightable. Only the source is. I have no need to use the source to use the API, with the exception of the header file.

    If the FSF wins, it would be an almost unprecedented increase in copyright coverage.
  • You are so wrong... by Arker (Score:2) Thursday May 24 2001, @01:03PM