POLL...Do you want a 64 bit BOINC client & BOINC MANAGER for Windows XP64 ?

Quote:

And therein lies the problem. You are quoting a commentary, or gloss, on the licence, just as I did. If you go back to the source text of the licence itself, the word 'organisation' is nowhere mentioned - not even as 'organization'!

In the actual licence, 'Each licensee is addressed as "you"', and the critical sentence is:

You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program.

Notice that the word is "recipient", as I quoted from Wikipedia.

So who is the "You" who is the original licensee in this case, on whom this obligation is placed? Did Berkeley licence the original to SETI-USA, or to an individual downloading programmer?

Agreed, but there is a crucial difference between the two interpretation sections of the license clause in question, which is the FAQ's on the GNU GPL were written to clarify the license text with the assistance of the GNU legal team. One has to consider it to be the more definitive interpretation of the intent of the license. The final arbiter of the intent, of course, would be a Court of Law in the jurisdiction where a case of license violation was filed.

With regard to the definition of "you" in the license text, at least here in the US, organizations are granted many of the rights and privileges of the individual.

Thus in the example of SETI-USA, the team is the licensee and has the right to make a binary only distribution of modified code to its members under the terms of the license. Members of the team are *not* recipients, they are users of the software in this context and are obligated to *not* distribute the binary outside of the team since they are not a licensee. This addresses roadrunner's comment in that the distribution of the modified binary to team members *only* does not constitute a release to the public. I would suspect (but have not researched it) that to be fully compliant with the GPL, SETI-USA would have to include some sort of EULA which stated any software provided was copyrighted by them and others, and use and/or distribution to non-team members was prohibited.

Regarding your last comment, there are also GNU FAQ's pertaining to the relationship between a developer and the organization sponsoring work done on their behalf under the GPL regarding transfer of rights and license obligations. Admittedly things get a little murky then, since you would have to be privy to the details of any terms drawn up between those two parties, but it is clear arrangements about non-disclosure can be reached which are in compliance with the GPL.

Again in the case of SETI-USA, the fact that Crunch3r (or others) is/are developing the software for them doesn't change the team's right to privately distribute a binary based on it. On the other hand, if they release a stripped down version of it to the public at large, then they would have to make the source code available for that version.

I should say that while I myself am not a lawyer, I have been involved with several projects which incorporated Open Source/GPL software for internal organizational use, and the advice from counsel was essentially what I stated above regarding the use of and license compliance with the software.

Alinator

Quote:

Hmm. Typically the word "you" is addressed to a person or to a legal person. A legal person is an individual or a corporation or in some cases a club. But the organisation would have to have some kind of articles of association, etc etc etc. Key question is if SETI-USA does the wrong thing, who do you sue? To me it is pretty clear that SETI-USA or any other BOINC team for that matter does not constitute a legal person and you would have to seek redress from individuals. In that sense, a team member who modifies GPL code is acting as an individual and not as the team. (heck, the team does not have any kind of governing body to stop him/her doing it!)

If you accept that, then if the team member then gave the software out to other members of the team, then he/she would definitely be "distributing" it because the transfer would be from one legal person to another.

Taking another look at it, suppose I was able to declare my team as a single "you" for purposes of GPL, then I could modify GPL code and then sell it freely without distributing source by simply:

a) Forming a boinc team
b) charging a membership fee to join, payable to me.
c) sending all members a binary copy.

It's asinine.

--Mark

You raise some good points here.

Your observation on whether a loose association, such as a BOINC team, constitutes an organiztion in the legal sense is certainly valid. I would argue that it does for the following reasons:

1.) You need to join/register to obtain full access to the services on the site/club.

2.) It has a founder, and by definition there is someone responsible for setting up and maintaining the website, though not necessarily the same person as the founder.

3.) It has mechanisms for the dissemination and communication of information, as well as collaboration and sharing of resources between members.

4.) There are well defined private areas for members only.

Already this is far beyond what is possible for a "team" in the strict BOINC sense of the word, and would seem to me to have all of the essential elements of an organization.

The question of whether they need to have a charter and/or more formal structure plays more of a role in defining whether they would enjoy limited liability for the "officers" in case of any disputes. In the absense of one, I would suspect the "point people" would carry more of this burden.

Regarding your question as to liability in case of disputes. I would suspect that, disclaimers aside, a Court of Law would hold the Webmaster responsible for any content presented on the site in the absense of another verifiable entity. Whether he would have culpable liability is another question, and most likely would only be required to remove the disputed material in a case such as this.

Pertaining to who is responsible for the release of and who is the copyright holder for GPL'ed software in this scenario, I disagree with your analysis. As Mr. Haselgrove and I were discussing, it is clear independant developer/client relationships are permitted under the GPL (as highlighted in the GNU GPL FAQ's), and the transfer of copyright, non-disclosure, and other private matters between those parties are allowed as long as they don't restrict the rights a third party would have under the license *depending* on what kind of a release was being made. I think it is fairly clear that if the private release is being made under the auspices of the team, then the team is the responsible party in all actions related to that release.

To address the issues you make specfically, whereas the team has no power to stop a member from excercising their right's under the GPL to modify the original Berkeley source code, they most certainly have the power to prevent *anyone* from releasing the modifications under their name, governing body or not.

I believe your extrapolation from there to individual members swapping modifcations between themselves is faulty, in that as long as the exchanges were happening in the members only sections of the website, then these would be considered to be private releases (or possibly collaboration depending on the circumstances) and not constitute a public release requiring source disclosure.

Your final scenario is closely related to another issue covered on the GNU GPL site. It is stated there that "free" in the sense of zero cost is not the intent or purpose of the GPL. From my read of it, there would be nothing wrong with doing exactly what you suggest (assuming you could get people to pay cash money for your service), as long as you provide a EULA stipulating the software is copyrighted and distribution outside of the confines of the team is prohibited. It is further stated there, the fact the software in this case is used to process data from a different public third party does not mean the software itself was released publicly, even though it originated from that public third party.

The key here is when you are acting as part/on behalf of a team you are not the individual in that context. If you run the binary under the terms of your membership, you are bound by the terms of that membership.

Finally, I see nothing asinine in any of the conversation so far, sir, unless you are saying the GPL specifically allowing private releases is. However, that would be an entirely different matter.

Alinator

Perhaps at this point further discussion would be more appropriate over at BOINC Alpha, since this really isn't directly related to EAH.

See link in Richard's previous post.

Alinator

Agreed, exactly.

There's no problem in a developer making two (or several) different sets of modifications. There's no problem in distributing different versions to different sets of people. There's no problem in making an extra-special version, and keeping it for your own use.

But if we assume, for the sake of argument, that BOINC is licenced under GPL, and giving a copy to a team-mate is distribution, then I have a major problem with anyone attempting to take away that team-mate's freedom to make and pass on further copies.