I have been having a spirited debate in the Einstein forum (this thread), sparked by the OP's throwaway remark about the development of a 'closed' version of BOINC, with extra facilities only available to members of one team, and restricted from distribution to the community at large. The question arises whether this is permitted under the GPL.

However, I find that I'm on rather shaky ground, because it seems to me that there are serious defects in the "official" distribution of BOINC from this website under GPL.

For instance, the 'BOINC software development' page claims that BOINC software is distributed under the lesser GPL, which turns out to be false - read the 'copying' file in the root of the source tarball.

Also, the distributed executable (Windows package only inspected) contains a copyright notice, but none of the other information resources required under GPL - limited warranty, notification of redistribution rights, and the text of the GPL itself.

Could anyone from the project please

Describe Berkeley's offical release policy under GPL
Comment on and review actual practice in relation to that policy

Also, the distributed executable (Windows package only inspected) contains a copyright notice, but none of the other information resources required under GPL - limited warranty, notification of redistribution rights, and the text of the GPL itself.

I believe that the click through part of the installer contains the required files, but I agree that this file should also be included in the installation directory.

Could anyone from the project please

Describe Berkeley's offical release policy under GPL
Comment on and review actual practice in relation to that policy

The current BOINC distribution policy is fairly shaky. Someone should modify the self extracting archive creation to make sure it includes the required license, disclaimer of warranty, and offer of source code. I will email David, Rom, and Charlie remind them of what is supposed to be provided with the binary.

Until recently the SETI@home distribution was also lacking, but currently the required files are distributed with the binary when downloaded by BOINC.

Because SETI@home links to a GPL library (FFTW), we try to follow the GPL to the letter. Discussion of the BOINC policy needs to come from David, not me....

There has also been discussion of what constitutes "distribution" under the GPL as opposed to "internal use" especially as it applies to BOINC teams. My understanding (IANAL) is that "internal use" would apply to a single legal entity (a business, household, corporation). In otherwords, in order to have an internal distribution, the distributing entity must be capable of being sued if the license is violated.

I don't think a BOINC team would qualify unless membership in the team is limited to members of a household or employees of a corporation. I don't think use by customers of a corporation would qualify as internal. So if a team that doesn't fall into the above category tries to distribute a closed source version to team members, the distributor would be in violation of the GPL, and the distributor of the binary could be sued. A team would have to incorporate, sell shares to the members, or pay them as employees in order for the distribution to be internal. It would also need to be willing to be held responsible for license violations by the members of the team.

This may not be the official "BOINC" policy, but it is how I would approach such cases for SETI@home. It would be a lot easier for people wanting to distribute SETI@home or BOINC to follow the license.

---

@ Richard: Message at EAH received. Thanks. Took a while to find my BOINC Alpha Account Key! :-)

I've been rebuilding all my personal systems and LAN. Sheesh, I've accumulated so much "stuff" it takes a while to get all the ducks back in a row! ;-)

Guess that's why I don't do it too often anymore! LOL

@ Eric:

Thanks for taking the time to elaborate some on the state of BOINC policy as it stands. This has provided some additional food for thought.

I do admire the decision to go GPL with the software. Not only is the idea of BOINC pioneering in and of itself, but by releasing it as GPL it provides a vehicle to explore the uncharted waters of what the License means in a hopefully more, shall we say scientific rather than high stakes confrontational manner than would occur in the strictly commercial world.

Alinator

Eric - many thanks for your prompt and helpful response. I'm sorry you had to register as a new user in order to be able to post!

And a warm welcome to my sparring partner from Einstein. But apart from that, my attempt to start a broader debate seems to have failed spectacularly.

Let's bump the thread a little.

IANAL, but...

Here in the UK, it's clear that a club or association is not recognised as a legal entity, unless it chooses to go through one of the formal processes of incorporation - usually as a company, but there are others.

Disregarding the confrontational example of sueing/being sued, the commonest practical implication relates to property (real estate). An unincorporated club cannot enter into a contract to buy or lease land or buildings.

The workround is for the members to appoint "Trustees" to hold the property on their behalf. These are not usually the executive or 'point' officers (Chair, Secretary, Webmaster) - partly because officerships often change by election each year, and it becomes expensive to transfer property every time. Also, the trustees act as legal long-stop to pick up the pieces if it all goes pear-shaped: members failing to elect a committee that a working majority have confidence in, for example. And it would be the trustees who sue and are sued on behalf of the members.

So I suppose it would be possible for individuals to hold a GPL licence "in trust" for the other members of the organisation, but it all sounds a bit tortuous, and presupposes the existence of a formal trust deed.

The club/team example is not the only place where the "You" in the GPL is ambiguous. Towards the end of David and Eric's recent interview with Lookers, David talks about plans to expand the BOINC user-base by promulgating it more widely within UCB. I'm sure that in practice he'll want to do that under GPL, but for argument's sake, how widely could you circulate a 'closed' UCB version under GPL?

Tenured academic staff? Of course
Computer Studies students on a distributed computing course? Almost certainly
Full-time enrolled students on campus? Probably
Part-time students on distance-learning courses, studying at home? er...
Day-release or sandwich course students, studying at work in their employers' time?
Non-tenured research associates?
Alumni?

It seems that this question has arisen just as we have a once-in-fifteen-years opportunity to seek some clarification. Version 3 of the GPL is at second discussion draft stage, shortly to go into 'last call' and ratification. At the moment, there is only one comment on the unchanged phrase "each licensee is addressed as 'you,'", and that's a minor semantic quibble.

The clause relating to distribution of modified versions, on the other hand, proposes to licence "anyone who comes into possession of a copy". That may change, however, as a couple of people have already objected that that literally gives licence to thieves and hackers!

Eric - in your official 'position statement' posting on optimised SETI applications (9 June 2006, 23:26:50 UTC), you mentioned that UC employs more 'Intellectual Property' lawyers than FSF ever will. Although you despise the term (and by extension, I detect a certain antipathy to the profession), are any of them human enough that you could prevail upon them to propose a form of words for inclusion in GPLv3 which would help to define who and what can and cannot be a licencee?

are any of them human enough that you could prevail upon them to propose a form of words for inclusion in GPLv3 which would help to define who and what can and cannot be a licencee?

I can suggest it. Whether they can be consise is another things....

Eric

---