Re: Bob asks: Then why don't YOU come up with a rule 2?
Posted: Thu May 21, 2015 12:25 pm
The FSF was (and still is) the owner of Fruit, as Letouzey had indicated (both publicly, and to the ICGA). The Rybka case involved someone entering a non-original work based on Fruit into an ICGA event. Given these data, I'm not sure why the ICGA shouldn't have forwarded the report to the FSF?! Should the ICCA not have contacted Hegener & Glaser if one of their products had been involved in an investigation (eg, Langer case)? If the ICGA had forwarded their report to the SFLC (the legal arm of the FSF), perhaps as a trumping of the FSF doing an investigation themselves, then I would agree it would be unwarranted.Rebel wrote:Harvey - The ICGA have forwarded the report to the FSF.
But the FSF tries to negotiate, they care more about making the whole world "open source" more than anything else (they call it "education"), and only resort to the SFLC when the circumstances require.
The FSF reviewed the analysis of the similarities between Fruit and Rybka, and found it pretty compelling, with it being something that we could absolutely pursue as a compliance case. They then asked for the contact details of the Rybka authors, and things proceeded from there.FSF wrote:When the FSF pursues GPL compliance cases, our end goal is to have the party continue distributing whatever software they're currently distributing, but to comply with the GPL when they do so -- by making sure that all of the software is available under the GPL's terms, that they provide source code through one of the mechanisms listed in the license, and so on. Our goal is not to simply have the violator cease their activity, or to seek monetary damages (although we do seek compensation for the time we spend pursuing the case).
We have a standard form letter that we use to start these discussions, and try to engage the violator in a productive conversation about what steps they need to take to come into compliance with the license. If the violator is unresponsive or uncooperative, we get the Software Freedom Law Center involved.
The Programmer's Open Letter does not mention the FSF, only that Fruit was publicly released open-source under the GPL license. Letouzey's open letter does not mention either the FSF or the GPL. I don't know where all this "intimidation" came from, maybe some echo chamber on the Rybka forum (who were constantly worried about others "peeking at Vas' source code").Rebel wrote: with Fabien's FSF intimidation (GPL breach)
It would sort of defeat the purpose of the GPL if one did not complain when it was thought to be broken. Moreover, given that ownership had been transferred to the FSF, it would become even more proper to do so.Rebel wrote: and later the complaint to the FSF itself
Most people believe their conjectures to some degree. Else they would not make them in the first place. Whether Letouzey thought it was 51% likely, or 80% likely, or 95%+ likely, I don't know. He might also have taken the above actions merely if he believed that Rybka had just "rewritten Fruit with different words" (in the sense of a translation). See LOOP for instance.Rebel wrote:[from the above actions] isn't it clear that [Letouzey] really believed his conjecture [that Rajlich copied code, not just ideas]?
This is what the FSF does too (either when they are themselves the owners, or on behalf of the owners [though they admit they have much less legal sway in such cases]). Then, after some rounds of negotiating, if you are still not willing to comply with the GPL, they pass the case on to the SFLC, who then initiate action with the distributors. It is only in extreme cases that they go public with any details.Rebel wrote:1. You first try to solve things in private.
The same person who wrote the above on the ICGA page (Coulom), was also on the ICGA Board that found Rybka to be "based on the source code of Fruit" in the sense indicated (Rule #2 originality). Similarly the phrase "this kind of derivative work" refers to Rule #2, and not to "code copying" as you mean it. One might write that Strelka (or IPPOLIT) "is based upon a disassembly of Rybka", and I would guess most people would agree with the statement (even if VR now says they are OK if they re-typed everything themselves). You should really ask Schaeffer or Björnsson regarding whether merely "code copying" or a wider sense of non-originality was the principal determination with LION++ (note that Björnsson was again on the ICGA Board that found Rajlich guilty of plagiarism and breaking Rule #2).Rémi Coulom (on his ICGA pages) wrote:This program (LION++) is based on the source code of Fruit, and was disqualified because this kind of derivative work is not allowed by the rules of the ICGA tournaments.