|
|
Subscribe / Log in / New account

Planning for decades

Planning for decades

Posted Mar 29, 2012 3:39 UTC (Thu) by jzbiciak (guest, #5246)
In reply to: Planning for decades by dlang
Parent article: A turning point for GNU libc

That's pretty much my point. Thank you for providing additional examples.

The FSF is in a unique position of power with respect to the GPL (at least the "ver X or later" variant). And because they ask you assign copyrights for anything you contribute to the FSF, they're in a unique position of power over that. Both give the FSF various (and differing) unilateral powers over your code, should you elect to extend those powers to the FSF. Those powers can be used in ways you do or do not agree with.

We trust that FSF won't disappoint us. They've earned a lot of trust from the community over the years, but that vague yet pervasive sense of trust is not a legal construct. As you say, "They have the ability; the only question is 'Will they?'"

The most likely answer is "They will not." But, it _is_ a potential single point of failure if I ever saw one...


to post comments

Planning for decades

Posted Mar 29, 2012 4:08 UTC (Thu) by xtifr (guest, #143) [Link] (7 responses)

If you're really paranoid about the "or later" thing, you can always say something like "or any later version which has similar requirements to always provide full source code when distributing derivatives", etc. Heck, now that I think about it, I might actually change some of my code to say that.

Planning for decades

Posted Mar 29, 2012 4:17 UTC (Thu) by jzbiciak (guest, #5246) [Link]

I'm not really all that paranoid myself. But to deny that the FSF is in a unique position and seeks to retain that unique position belies the facts. That's the only point I'm trying to make.

Planning for decades

Posted Mar 29, 2012 11:06 UTC (Thu) by smurf (subscriber, #17840) [Link] (5 responses)

Legally, "any later version" might well be interpreted as "… which intends to convey essentially the same rights".

If I had intended to publish my code under a 3-BSD license, ten years ago, I would have done so; therefore the FSF has no right to publish a 3-BSD-equivalent and label it "GPLv4".

Again, this is about the stated intent of both the FSF (what does it stand for? what are their goals?) and me (did I agree with these, when I chose that license?). It's not about the text "or any later version" with an implied "… whatever that says, decided unilaterally by the FSF".

Whatever your opinion about legal beagles in general, give them some credit.

Planning for decades

Posted Mar 29, 2012 11:16 UTC (Thu) by dlang (guest, #313) [Link] (1 responses)

remember that in some people's eyes, GPLv3 breaks the "… which intends to convey essentially the same rights" criteria compared to v2

in the opinion of the FSF it doesn't, in the opinion of many others, it does.

Planning for decades

Posted Mar 29, 2012 12:25 UTC (Thu) by wookey (guest, #5501) [Link]

"in the opinion of the FSF and many others it doesn't, in the opinion of many others, it does."

Would be a more even statement of the situation. (Your original suggests that it is FSF's opinion versus 'many people'. Opinion is more evenly divided than that.)

Planning for decades

Posted Mar 29, 2012 12:50 UTC (Thu) by jzbiciak (guest, #5246) [Link]

3-BSD conveys fewer restrictions on the recipient of the code than GPL v2. So are you saying future GPL licenses can only be more restrictive?

Let's consider the spectrum of freedom from the perspective of the recipient of the code, not the author. Public domain code is the most free of all. 3-BSD is a step back from public domain, but not by much. GPL v2 is more restrictive than 3-BSD, with all its wording about derivative works and the like, source redistribution requirements, and so on. GPL v3 is more restrictive yet. And then there's the default state of "copyrighted work," where the recipient has very few default rights.

Are you suggesting that the only arc future GPL licenses can take is toward less freedom for the recipient?

Ok, so perhaps I was leading you down a rhetorical garden path with those questions. And, since you're well steeped in Free Software, you're not persuaded to follow. But, imagine arguments constructed along those lines in court. Such arguments could be much more persuasive to a less deeply invested judge and/or jury. If someone used your code with an imagined GPL v5 that let them make a proprietary package that you didn't approve of, good luck winning over the court if you try to sue.

I realize that the FSF is trying to maximize the freedom of the ecosystem as a whole. But realize that it does so by limiting the freedom of individuals to appropriate the code and take it private. It lays that all out in the preamble, saying that to protect your rights, it puts limits on what you (or anyone using the software) can do, and places on you certain responsibilities in order for you to use the licensed software. That subtle, tricky and almost counter-intuitive balance of "protecting your rights by limiting your actions" can be a lot for folks to wrap their heads around.

What if the FSF decides that it's in the greater interest of the ecosystem to place fewer restrictions on the recipients of the software, thinking that yes, there will be some that abuse the additional privilege, but it's entirely likely be more new contributors that will not. There are those who find the GPL legally risky and don't contribute, and maybe someday the FSF will try to meet them in the middle to bring them into the fold. (Probably not as long as St. Ignucious is still kicking, but hey, there's an awful lot of future ahead of us. And don't think that I'm advocating that the FSF should do this. I'm only arguing about the potential range of outcomes here.)

The GPL v2 states:

The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.

"Similar in spirit" is rather wide open to interpretation. The overriding theme is "freedom", and as we've all seen, there are many, many differing interpretations of what constitutes freedom. A license that shifts the balance of freedoms toward the recipient (ie. a move toward something more like 3-BSD, if not moving entirely to 3-BSD) would not necessarily be incompatible in spirit with the GPL in many peoples' eyes.

In any case, the only truly legally binding phrase in that clause is this one:

If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.

Sure, the preceding paragraph said "similar in spirit," but the FSF alone gets to define what their spirit is. After all, they wrote the license.

Planning for decades

Posted Mar 29, 2012 13:03 UTC (Thu) by jzbiciak (guest, #5246) [Link] (1 responses)

Here's a thought experiment for you: Suppose someone releases a software package under "GPL v3 or later", with all its new requirements to prevent tivoization. They do this because they like the additional guarantees their software will remain free, and they really don't want their software tivoized.

Now the FSF releases the GPL v4 that reverts the changes in GPL v3, and reads identical to GPL v2.

Theoretically, the GPL v3 was similar in spirit to the GPL v2. So, this imagined GPL v4 should also be similar in spirit to GPL v3. Now what?

Really, the "or later" clause cedes an awful lot of control to the FSF. That's been my point all along. The punch line is that if you care that much about the specific semantics of the license that covers your code, you really shouldn't put a wildcard in the license that you don't control.

Planning for decades

Posted Mar 29, 2012 16:45 UTC (Thu) by khim (subscriber, #9252) [Link]

The punch line is that if you care that much about the specific semantics of the license that covers your code, you really shouldn't put a wildcard in the license that you don't control.

Yup. The irony here is that it looks like FSF is the only organization which understands that.

MPL: You may distribute the Covered Software under the terms of the version of the License under which You originally received the Covered Software, or under the terms of any subsequent version published by the license steward.

CC-BY-SA: You may Distribute or Publicly Perform an Adaptation only under the terms of… (ii) a later version of this License with the same License Elements as this License.

GPL is the only popular copyleft license which gives program author the right to refuse delegation of “autoupgrade” rights to license “stewards”. Note that such liberty was given to authors by all versions of GPL, including the first one. All these other licenses were created after GPL was created - yet they cede “an awful lot of control” unconditionally.

It's funny that the very people who are bashing FSF for the “or later” clause are often advocating MPL or CC-BY-SA as an alternative despite the fact that they are obviously more problematic in this regard.

Planning for decades

Posted Mar 29, 2012 4:45 UTC (Thu) by slashdot (guest, #22014) [Link]

I guess that as long as Stallman is in charge, it's a reasonably safe bet that he won't release a non-copyleft GPL.

The more worrisome thing is what happens after him.


Copyright © 2025, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds