Hacker News Reader

Comment by Ruud-v-A | Hacker News Reader

I think it is a mistake to license anything under a “or later version” license. If you do that, you licence your code under a license that you have never read, so how can you know that it serves your intentions? You are blindly licensing code under a licence that has yet to be written!

In the GPL case, the Free Software Foundation states that “The Free Software Foundation may publish revised and/or new versions of the GNU 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.” (Section 14 of the GPLv3.) Now of course I don’t expect the FSF to publish something radically different, but they _could_ do so. Even if you trust the FSF now, are you sure you trust the FSF twenty years from now?

View on HN

If you don't trust the FSF, the GPLv3 actually offers another alternative: "If the Program specifies that a proxy can decide which future versions of the GNU General Public License can be used, that proxy's public statement of acceptance of a version permanently authorizes you to choose that version for the Program."

So you could say "GPLv2, GPLv3, or any later version approved by ...". Which then means that a future relicense to allow GPLv4 (after reviewing it to your satisfaction) would not require contacting every single contributor.

Even in the worst case scenario, it remains available under the original license, and that's not going to change.

What is the worst case scenario, by the way? You could say it depends on who you are. One such scenario would be for developers who agree with the FSF's strong copyleft philosophy who are betrayed by a future hypothetical revision that removes all copyleft provisions to make it no stronger than, e.g., BSD. This affects only a subset of those choosing a license for their project. (I.e., if you don't care about copyleft, this purportedly horrific thing doesn't even affect you.) Funnily enough, the subset of those who would be affected are hardly ever the ones I see making the argument you are.

So what terrible things do you imagine the worst case scenario involving?

I would like to know that too, since all the worst case scenario I can think of is also defining aspects of permissive licensed projects.

For example, a community could fork a project and license it under a new'er version, and then the original founder could suddenly want to incorporate the new changes but at the same time not want to use the new license, and the new community refuses request for a license to the old version. I don't think it ever has happen with a gplv2+ project, and it require a quite hostile community to begin with, but as a worst case scenario that would likely be it.

If you are pro-patent protection, the FSF already released something radically different.

To be fair, most FLOSS license, including GPLv2 and BSD 2-clause and 3-clause, are (were?) usually considered to have an implicit patent grant, similar in spirit to the one in MPLv2 or GPLv3. I don't think that argument that one could use GPLv2 while thinking that software patent are good has that much ground.

On the "thinking that software patents are good" front, I think this bit from the GPLv2 preamble is pretty clear:

> Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all.

> they _could_ do so.

They already wrote legalese that says that they can't do so. This isn't merely an empty promise: it's already encoded into the text of the license. If a later version of the license is radically different and not in spirit of the GPL or imposes upon authors obligations that earlier versions of the GPL do not, then the current version of the GPL already says that later version would be invalid.

Would the anti-TiVoization wording in the GPLv3 count as "obligations that earlier versions of the GPL do not [impose]", and if so, does that mean that the GPLv3 is invalid?

The obligations are upon the authors and copyright holders. So, for example, a later version can't remove the no warranty clause, because that would burden the authors with providing warranties.

I trust the FSF twenty years from now more than I trust someone today who doesn't use "or later version".

So you don't trust Linus Torvalds et. al.?

Guy who thought bitkeeper was a great idea? Guy who doesn't do anything about nvidia's blob, even though he could? Yeah, I don't trust him.

> Guy who doesn't do anything about nvidia's blob, even though he could?

Well, other than flipping them off :)

Yeah, I'd like to know the legalities of this "we can rewrite the license in anyway we choose, and your code from 20 years ago will fall under its terms".

Once again the FSF political motives cast its license into dubious territory. Just stay away from the GPL.

Reply on news.ycombinator.com