On Thu, Mar 29, 2012 at 1:00 PM, David Jeske <davidj@gmail.com> wrote:

I started to talk to one of them about installing CSLA (or MHonArc, or anything really), and realized I should see if you folks are interested in a great bundled archiver,

I'm personally interested, but that's not going to be the main focus until the core gets a little less alpha, I would guess.

I admit that even with a pretty good knowledge of these many licenses, I'm not familiar with the intracacies of FSF copyright assignment and non-GPL free licenses.

The bottom line is that if you assign it to the FSF, they can change the license of future copies to something you don't like without your permission, or even consulting you. This has been done in the past (eg, libreadline). I don't know whether they would bother with a non-GNU project, though.

However, copies of the version you originally released under a different license remain under that license, and there's always a grant-back clause in the assignment contract that allows you to make derivatives of your contributions available under any license you like.

The ClearsilverArchiver code (written by me and two others) is released under the "Simplified BSD" license and "totally free". It's important to me that any code I release be similarly free-and-unrestricted (i.e. BSD/Python/Artistic/PublicDomain), not free under certain conditions (i.e. GPL/LGPL). It's not possible to assert GPL restrictions on totally-free code, because it's already totally free.

That's not the way copyright works, though. It certainly is possible to assert GPL (or any other) restrictions on given *copies* of permissively-licensed code. If you've got a copy of the old (say, early '90s) O'Reilly "X Window System" series kicking around, check out the copyright notice in those books. (Preventing that is precisely why copyleft advocates advocate copyleft.) Even on a verbatim copy, lawyerly FUD means that even if there is no actual legal issue, practically it may be infeasible for just plain folks to redistribute. Cf. the DMCA takedowns.

FSF says S-BSD is GPL-Compatible, which I believe means they are saying they have no problem with GPL code depending on and being combined with (i.e. linked with) S-BSD code, because the S-BSD code is fully open-source and does not put restrictions on the use of the GPL code.

No. What they mean is that it is Borg-able. You can assimilate S-BSD code into a GPL project, and that copy is distributed under the GPL. Perhaps in legal theory they cannot prevent you from making copies of the S-BSD portions and doing anything the S-BSD permits, but the S-BSD (like other permissive licenses) does not require them to tell you what parts are S-BSD, only that some parts are, and who wrote those (unspecified) parts. (The wording is generally to the effect of "This file is part of the FOO Program, which is licensed to you under the GPL. It contains software by J. Random Hacker, with the following permissions notice...".) The burden will be on the user to determine which parts can and cannot be copied, and if it comes to a court case, the user will have to prove that the parts they've copied are not actually GPL.

I would say you should try to retain copyright, and have the Mailman project distribute it with the S-BSD license under the "mere aggregation" clause of the GPL. This would entail certain restrictions on interface. Eg, you can't put the whole thing in a pipeline Handler, and you would need to have a separate webapp for summarizing/indexing/searching/retrieving the archived posts. I advocate those restrictions anyway. :-) Some small glue parts that need to be tightly integrated with core Mailman might need to be done under GPL.

It's also my understanding that the primary reason for FSF copyright assignment is to provide a coherent entity to enforce the terms of the GPL by challenging violators who don't redistribute source.... something which is not necessary for S-BSD. (Though I suppose they could enforce that folks include the S-BSD copyright notices.)

The FSF's reason is so that they have control over the license, which allows them to make it GPL if that seems like a good idea to them. (Mostly they are way too busy to go looking for opportunities, though, and it's a labor-intensive process for a project of any size.) In return, they will enforce license provisions.

Projects may also wish to do this so that they have the legal right to offer other licenses (the FSF is not a good assignee for this purpose!), or change the primary license. If no single entity owns the whole copyright, then you have to get agreement of all owners, some of whom may be in retreat in a Tibetan monastery or the heirs to someone who lost an argument with a bus, etc. (MIT specifically allows sublicensing, as does Larry Rosen's AFT; but S-BSD does not.)

Is Mailman-team is interested in having a better built-in archiver that is included in the distribution, but licensed under the less-restrictive S-BSD terms?

I certainly would be interested! (I believe that freedom can only be preserved by the blood and legal fees of patriots, while GPL isn't much help any more, and increasingly frequently a PITA. :-) Other folks prefer to license their work under copyleft, specifically the GPL. But there's explicit permission in the GPL for such distribution (the "mere aggregation" clause), as long as communication between the archiver and core Mailman is done by open protocols such as *MTP and maildir. (Ie, the included Handler and IArchive integration code will probably be GPL, but there's no need for anything in a separate process, including indexing and web UI, to be GPL.)