4 ms·
The license text as found in the Shen and KLambda sources lacks the "Comment on..." section, i.e. it's boilerplate 3-clause BSD. Apparently, Dr. Tarver intends
by michaelsbradley 12y ago
The license text as found in the Shen and KLambda sources lacks the "Comment on..." section, i.e. it's boilerplate 3-clause BSD.
Apparently, Dr. Tarver intends the license.pdf file to serve as a "licensing guide" for would-be port authors, while not intending to add terms or conditions to the BSD license by way of the Comments section. I think that intention could be made a bit more explicit, but I'm satisfied that the license as given in the sources is plain vanilla 3-clause BSD.
Perhaps a better solution would be to have a license.pdf file which contains only the BSD text and a separate PDF file which gives commentary on the license. Someone mentioned that possibility on the mailing list yesterday; if there's enough hubbub about it, perhaps Dr. Tarver will implement the suggestion.
I am doubtful that a BSD license excludes the possibility of a port author licensing the port author's own original code under a GPL license, though that seems to me to be Dr. Tarver's conclusion. At most, I think Dr. Tarver could choose to withhold certification from and refuse to list on the download page any port that includes GPL'd code.
- hga 12y agoHowever the pdf file is part of the source package, and presumably future ports. This is at the level were lawyers would have to be consulted to be sure this interpretation is correct (and that's a per country thing...), and that brings us back to negating the #1 advantage to Shen being "BSD licensed", you don't have to run it by "legal". His purpose in this is to prevent any one of a number of GPL related stunts, such as explicit re-licensing as was attempted on a painfully reverse engineered OpenBSD driver, or declaring a GPL license on an entire Shen distribution. Of course no part of the standard library will have a GPL or other viral license. He's OK with e.g. a Scientific Shen distribution incorporating the GPLed GNU Scientific Library, just as long as the, err, virility in both practice and claims is controlled. If he were to accept such a distribution on the downloads page, it would be very clearly marked. And addressing your final point while emphasizing mine, I don't believe he's trying to "[exclude] the possibility of a port author licensing the port author's own original code under a GPL license". He's just saying, or trying to say, you couldn't then claim GPL over the whole distribution including his modified BSD licensed code. Given that we've come to different conclusions about this (granted, I've been involved in this BSD effort for almost a month, and entirely approve of his ends, just not his means), I'm even more confident of my position. And thus Shen will continue to be kneecapped in trying to gain market share, something he's declared to be entirely comfortable with.
- michaelsbradley 12y agoI think we're agreed that the present approach is not optimal. I share your concerns about having to "run it by legal". I would prefer the Comments section to be preceded by a highly explicit statement of intention: "End of the license text. The following statements are the copyright holder's comments on the license and in no way add any terms or conditions to those given above." Or something like that. Better yet, the comments could be moved to a separate file; but now I'm repeating myself... I also share your interpretation of the motives. I read carefully all of the material linked from Dr. Tarver's recent license-related posts to the Shen mailing list, and your replies as well. The sentence under Comments that is causing confusion reads: "This work may be placed under any license of choice except GPL..." The subject of the previous sentence is "original code", which leads me to understand "[t]his work" as referring to the same. But perhaps you're correct and it's referring back to Shen itself.
- hga 12y ago"End of the license text...." Just about exactly that was proposed by someone, and implicitly rejected in the redraft :-(. Erk, I think you're interpretation is "right", as in that's the clearest way to read those sentences. I was going by one reading of the intent from the historical discussion, which matters less than the actual legal language. His intent might be exactly this, a Shen distribution cannot include GPLed code you write because of what he believes about the virality of the GPL, and the very bad behavior he's noted from RMS/FSF/SFLC (that I personally attested to/added from the '80s). As I recall, he also was unhappy with how GPLing Qi V1 worked out, or didn't, but you don't really need all that to treat the GPL like garlic and holy water, e.g. I've noted that the most successful implementations of niche languages have less severe licences. Yep, "run it by legal", which just isn't going to work when there are so many competitive languages, e.g. other Lisps including the very tasty Clojure, and non-Lispy functional languages which Shen is trying to make Lisp competitive with.