Top Mud Sites Forum - View Single Post

nhl · 04-29-2006, 08:08 PM

If litigation was as simple as you imply, IPR trials would be done as 30 minute drive-in litigation, instead of the lengthy processes they are today.

If DIKU would sue Medievia in the US, which has a legal system based on case law, from my layman perspective, here is roughly what would follow:
1. DIKU claims that Medievia is violating the copyright license by making profit of the DIKU teams work, and by not providing credit. To which Medivia counterclaims that:
a) the DIKU license is too restrictive, and therefore can not be legally enforced,
b) the DIKU licenses claims about non-profitability were based on the university's demand, to which Medivia is not a party,
c) that Medievia is making profit based on their own proprietary work, not that of DIKU. Further, Medievia would likely claim that their code is significantly and materially different from DIKU, and that therefore it can not be considered a derivative work.

I hope we can all somewhat agree that this is the most likely path, should it ever reach a trial?

What happens after this? For claims a) and b), the court would probably hear expert testimony from both sides. For c) the court would hear witness testimony from Medievia about their development model and the amount of work they have put in to the development. It is also likely that for c) the judge would order a code review to find significant overlaps between the DIKU codebase and Medievia.

Once all this has been done, the court would have somekind of idea to which direction the case is leaning. Then enters the case law. In the US, litigation is based on more than just the law. As part of any trial, lawyers and the judge references previous cases in the same field. If there are no such cases, then in order to pass verdict, the court takes into account not only the law, the intent of the law, but also its impact on society. For example, a court would not be likely to impose a sentence that severely limits commercial interests in that area, unless it can be shown that the commercial interests are causing direct damage to the other party.

For some legal cases that a court might see as precedents, please refer to Valens posts on the mudmagic boards from 2004, especially , and . Overall, Valens makes a pretty realistic argument from a litigation point of view - I think we, as programmers, need to remind ourselves that litigation is less black-and-white and less absolute than what we are used to.

I haven't claimed that I have hired a lawyer. I have said that I have spoken with one that specializes in IPR. There's a big difference between hiring one, and discussing out of general curiosity about the issue. Outside of general curiosity (and the feeling that this yearlong battle of personalities is silly), I do not have an interest in DIKU (I don't like the engine myself) or Medievia (never played it, I likely never will).

You claim many things too. But you, on the other hand, do have a direct interest in this case. You have publicly stated that Medievia has been part of a crime, that they have participated in theft. Were this case ever found in the favor of Medievia, they would have a quite strong case of libel against you.

I'm not lying (and I think it's a childish personal attack to claim that I am, something which has no place in a public debate), I am only reiterating my understanding of matters as explained to me by someone who should understand these things alot better than I personally do. Does it mean that I can't be wrong? Ofcourse not, the only one who seems to assert absolute authority on IPR matters (despite having no qualifications for doing so), is you. The legal system is complex, and as I've stated above, litigation takes into account many other factors than just the written law. Which is why you'll find that even legal specialists have split opinions on many IPR cases.

This case has nothing to do with LP drivers. I'm purely participating in this conversation out of personal curiosity. I'm especially curious about the "clean room" debate. Why? Because it has a big implication on software development in general, and if a "clean room" piecemal implementation would be required to avoid automatic copyright infringement, then how does that relate to open source? There's thousands of open source projects available - if a commercial venture were to write their own webserver that emulates the apache interface to PHP, would they automatically be infringing on the Apache copyrights unless they can prove the development was done in a clear room environment? I would certainly hope not, usually the burden of proof in legal cases has been on the plaintiff.

Another example of the "Clean room" problem. In the recent SCO versus (insert basically anybody), SCO claimed that parts of Linux (SMP, JFS, LVM etc) was based on SCO's copyrighted code. Linus Torvalds and OSI called for SCO to disclose which parts of Linux infringed on SCO's IPR so that those could be rewritten. As the sourcecode of these functions is openly available, it would (under your broad definition of derivative work) mean that Linux would be irrepairably tainted for practical purposes, if SCOs claim would hold up.

Legal arguments aside, I think it's unethical to strip out the credits from DIKU, even if one claims to have rewritten the whole thing. A few lines of "We thank XYZ for their work" is certainly not a big demand. My ethical position on the other issue, that of making profit from Medievia, is a bit more complex. Looking at a totally ethical perspective, the best solution would have been had Medievia and the DIKU team found some common ground and maybe even agreed on somekind of revenue sharing. Failing that, as I don't have access to Medievia's current sourcecode, I will have to give Vryce & co the benefit of a doubt that they have indeed rewritten the whole underlying driver as they claim to have, and that it's therefore not my thing to throw stones and sticks at them. I also argue, that Medivia and co make their profit from running Medivia (as in all the custom work they have done), not from running DIKU.

If this position somehow groups me with Matt and Medievia (I didn't know they were in the same group even), and means that my opinion and value-set come into the line of ethical fire (is this somekind of lame threat?), then so be it. To me, it only says that you are unable to deal with the fact that these issues are complex and that just because people disagree, doesn't make either party unethical.

04-29-2006, 08:08 PM	#35
nhl Member Join Date: Mar 2006 Posts: 37	If litigation was as simple as you imply, IPR trials would be done as 30 minute drive-in litigation, instead of the lengthy processes they are today. If DIKU would sue Medievia in the US, which has a legal system based on case law, from my layman perspective, here is roughly what would follow: 1. DIKU claims that Medievia is violating the copyright license by making profit of the DIKU teams work, and by not providing credit. To which Medivia counterclaims that: a) the DIKU license is too restrictive, and therefore can not be legally enforced, b) the DIKU licenses claims about non-profitability were based on the university's demand, to which Medivia is not a party, c) that Medievia is making profit based on their own proprietary work, not that of DIKU. Further, Medievia would likely claim that their code is significantly and materially different from DIKU, and that therefore it can not be considered a derivative work. I hope we can all somewhat agree that this is the most likely path, should it ever reach a trial? What happens after this? For claims a) and b), the court would probably hear expert testimony from both sides. For c) the court would hear witness testimony from Medievia about their development model and the amount of work they have put in to the development. It is also likely that for c) the judge would order a code review to find significant overlaps between the DIKU codebase and Medievia. Once all this has been done, the court would have somekind of idea to which direction the case is leaning. Then enters the case law. In the US, litigation is based on more than just the law. As part of any trial, lawyers and the judge references previous cases in the same field. If there are no such cases, then in order to pass verdict, the court takes into account not only the law, the intent of the law, but also its impact on society. For example, a court would not be likely to impose a sentence that severely limits commercial interests in that area, unless it can be shown that the commercial interests are causing direct damage to the other party. For some legal cases that a court might see as precedents, please refer to Valens posts on the mudmagic boards from 2004, especially , and . Overall, Valens makes a pretty realistic argument from a litigation point of view - I think we, as programmers, need to remind ourselves that litigation is less black-and-white and less absolute than what we are used to. I haven't claimed that I have hired a lawyer. I have said that I have spoken with one that specializes in IPR. There's a big difference between hiring one, and discussing out of general curiosity about the issue. Outside of general curiosity (and the feeling that this yearlong battle of personalities is silly), I do not have an interest in DIKU (I don't like the engine myself) or Medievia (never played it, I likely never will). You claim many things too. But you, on the other hand, do have a direct interest in this case. You have publicly stated that Medievia has been part of a crime, that they have participated in theft. Were this case ever found in the favor of Medievia, they would have a quite strong case of libel against you. I'm not lying (and I think it's a childish personal attack to claim that I am, something which has no place in a public debate), I am only reiterating my understanding of matters as explained to me by someone who should understand these things alot better than I personally do. Does it mean that I can't be wrong? Ofcourse not, the only one who seems to assert absolute authority on IPR matters (despite having no qualifications for doing so), is you. The legal system is complex, and as I've stated above, litigation takes into account many other factors than just the written law. Which is why you'll find that even legal specialists have split opinions on many IPR cases. This case has nothing to do with LP drivers. I'm purely participating in this conversation out of personal curiosity. I'm especially curious about the "clean room" debate. Why? Because it has a big implication on software development in general, and if a "clean room" piecemal implementation would be required to avoid automatic copyright infringement, then how does that relate to open source? There's thousands of open source projects available - if a commercial venture were to write their own webserver that emulates the apache interface to PHP, would they automatically be infringing on the Apache copyrights unless they can prove the development was done in a clear room environment? I would certainly hope not, usually the burden of proof in legal cases has been on the plaintiff. Another example of the "Clean room" problem. In the recent SCO versus (insert basically anybody), SCO claimed that parts of Linux (SMP, JFS, LVM etc) was based on SCO's copyrighted code. Linus Torvalds and OSI called for SCO to disclose which parts of Linux infringed on SCO's IPR so that those could be rewritten. As the sourcecode of these functions is openly available, it would (under your broad definition of derivative work) mean that Linux would be irrepairably tainted for practical purposes, if SCOs claim would hold up. Legal arguments aside, I think it's unethical to strip out the credits from DIKU, even if one claims to have rewritten the whole thing. A few lines of "We thank XYZ for their work" is certainly not a big demand. My ethical position on the other issue, that of making profit from Medievia, is a bit more complex. Looking at a totally ethical perspective, the best solution would have been had Medievia and the DIKU team found some common ground and maybe even agreed on somekind of revenue sharing. Failing that, as I don't have access to Medievia's current sourcecode, I will have to give Vryce & co the benefit of a doubt that they have indeed rewritten the whole underlying driver as they claim to have, and that it's therefore not my thing to throw stones and sticks at them. I also argue, that Medivia and co make their profit from running Medivia (as in all the custom work they have done), not from running DIKU. If this position somehow groups me with Matt and Medievia (I didn't know they were in the same group even), and means that my opinion and value-set come into the line of ethical fire (is this somekind of lame threat?), then so be it. To me, it only says that you are unable to deal with the fact that these issues are complex and that just because people disagree, doesn't make either party unethical.