So, there’s a Gentoo/FreeBSD project, that attempts to graft the two systems together. The lead developer in that project misread the old 4-clause BSD license on some older files, and paniced repeatedly. (It even made it to the howling wasteland.) Of course, the problem is not actually a problem – it’s caused by worry about a clause that was removed years ago. Dru Lavigne has a nice writeup, and Wes Peters summed it up best in comments: “Now everybody get back to work. This is a 7-year-old nonissue…”
5 Replies to “License inanity”
Comments are closed.
Doesn’t the Dru Lavigne article miss the point mentioned in Flameeye’s second post?
Berkeley released all their code from 4 clause to 3 clause, sure. But other developers have used 4 clause licenses in BSD as well (I seem to remember the USB stack used by the BSDs has this, for instance). Berkeley have no authority over code they didn’t originate, so unless those individuals *also* agreed to relicensing then there is still 4 clause code in force in the tree.
So the question is have they *all* done so? If not, then I think Flameeyes is right in what he’s saying about the GPL conflict here.
These clauses apply only if the specific code is used in advertising for the product, so if Gentoo/BSD put up a big banner on the project’s page saying (for example) “Now using version 1.23 of Somebody Someone’s /src/dev/sys/usb_rev.c!”, then it would apply.
A lot of those clauses are there because nobody happened to remove them yet, as Dru found – it’s getting cleaned out of the tree. So, even in the rare circumstance where it would apply, it’s probably not supposed to be there any more.
There’s a difference between “there’s a vague chance there could be a problem with an obscure file in the source tree that hasn’t changed in years/decades” and “I must pull all copies of the software now!”.
In practive, it works out like this: to my knowledge, nobody (since the USL suit) has been sued over a BSD file license. There’s been a number of GPL lawsuits, though.
Please read http://leaf.dragonflybsd.org/~tgen/4clausebsd.log . It’s a transcript from a discussion that went on in the IRC channel regarding this. I hope people will realize after reading it, that this isn’t really this shocking, does not ‘[destroy] the very foundation FreeBSD is based off’, nor is it a ‘debacle’.
Regarding the USB stack, it originated in NetBSD, so yes, figures…
P.S.:
Both quotes are from the ‘flameeyes’ ‘blog’ post on 08/01/2007, including the edit.
Thing is, as Mark says up there, changing license statements on code acquired outside some sort of arrangement or contract allowing same isn’t kosher. Is FreeBSD core doing it automatically any better than that “Dingo” (?) fellow doing a global replace on his OMG ‘Free/Open/Net/Darwin merged’ wunderdistro a while back? (MicroBSD, apparently taken over by some saner minds in Bulgaria for the three months it took to discover Net and Open filled the niche already.)
GNU Worlders live in fear that their code will be copied into closed trees against their wishes (or in hope that Microsoft will be caught doing so), and the SCO thing has made some projects rightfully concerned about keeping good records of where the code committers commit originated (and thus who exactly holds copyright).
Here in BSD-land we have a less mind-bending license and fewer absurd threats — the license contains no ‘don’t be a jackass’ clause, so people are free to be jackasses without having to attempt to overturn the open projects or the license — but
The advertising clause isn’t really a big deal (except for the guy who has to explain the concept of telling the truth to management or marketing), and in the absolute worst case for noncompliance a court could enforce revocation under the license terms until the offender complies, maybe with some very piddly token damages — how do you calculate revenue lost for an unpaid “ad” that didn’t run? I’m having a hard time working out what worm-can a misclaimed authorship could open legally (fraud?), but rewriting a license to suit your mood (outside of a framework of consent — preferrably a provable, paper-trail-ish one — allowing such) would be a breach of contract with all the potential fun that entails.
—
I guess what I’m trying to say is: The advertising clause is not a time bomb, but as far as I can tell, *ripping it out* should be done with the caution you’d take to defuse one. Technically, for example, there’s the slim risk that someone could come forward who managed to touch a file after the CSRG disbanded with whatever (loose, but still arguable) legal framework it had to apply its license to submitted code, but before the Regents of UC gave their permission to drop the clause. This wouldn’t be the end of the world, but it’d be an opportunity to start shit and run up some legal bills — the sort of thing corporate monstrosities try to avoid when they’re aware of the possibility, since there’s always the (infinitesimal) possibility of everyone with a claim coming out of the woodwork at once.
Well, corporate monstrosities with foresight, anyway. Other corporate monstrosities find this distracts them from their golf game, and isn’t that what contracting with Microsoft is for?
Gah, try a whole paragraph:
Here in BSD-land we have a less mind-bending license and fewer absurd threats — the license contains no ‘don’t be a jackass’ clause, so people are free to be jackasses without having to attempt to overturn the open projects or the license — but that’s no excuse to get complacent or avoid considering the diligence required to keep posteriors covered if a dispute ever arises.
[Also, I have no idea how FreeBSD is approaching the fourth-clause removal (and I hope Open and Net and everyone else are equally careful when making such changes), so please don’t read that as specific criticism… I’m just pointing out that the MicroBSD guy got run out of town, while if a ‘respectable’ group proposed something equally bozotic it might slip through for not being a bikeshed. (Everyone can understand a bikeshed…)]