Single-company free software
Single-company free software
Posted Oct 11, 2005 17:12 UTC (Tue) by Ross (guest, #4065)In reply to: Single-company free software by lutchann
Parent article: Single-company free software
I like Wikipedia as much as the next person, but just because it says something doesn't make it true. I have not actually read the article. If I did I'd feel obligated to fix it which I just don't have the time to do ATM. Most licenses have not been tested in court in any jurisdiction. There just isn't any credible reason to believe in the ability for a pure license to be revoked outside of any terms embodied in that license (the GPL in fact automatically revokes itself in certain conditions). The GPL is a promise from the copyright holder to the recipient of the software. If they tell you you have a perpetual license to distribute modified copies of the work, there is no legal way for them to take back that promise in the future.
What they can do is release it under a different license and make all future modifications only under that license. That does not remove the copies of the work under the old license nor take away the rights granted to people holding those copies, including the right to distribute them to others under the same terms.
Posted Oct 11, 2005 18:49 UTC (Tue)
by lutchann (subscriber, #8872)
[Link] (3 responses)
The reason I linked to it, though, was because I'd not previously heard the argument it suggested regarding revocability of the GPL, which I will summarize here for you and others who would prefer not to read the article. I'm not saying the argument is correct or well-founded, merely that it is interesting to consider.
The main point is that the GPL is (was, maybe) unique in that permits redistribution not by allowing licensees to sublicense to the recipients to whom they distribute the software, but by licensing the work from the original licensor to the recipients. This might be interpreted to mean (IANAL!) that if the original licensor somehow loses the copyright, the program could no longer be distributed to new recipients.
Anyway, as I pointed out in a post a couple of minutes ago, the reason it is particularly important for the conditions of the GPL to be shaped by the courts is because the license is effectively set in stone for a huge amount of software, so there really is no other way to find further clarification on the grey areas that have surfaced over the last decade.
Posted Oct 11, 2005 23:22 UTC (Tue)
by Ross (guest, #4065)
[Link] (2 responses)
As for the point about licensing: licenses have to work that way. The only person who can license a copyrighted work is the copyright holder. They may give permission for another to sublicense, but that only works because they are acting under the copyright holder's authority. A copyright isn't ever transferred through licensing, GPL or otherwise. The point is not that the copyright holder doesn't have absolute authority over distribution with the GPL, but that they do have such authority and chose to use it to grant sublicensing rights with only minimal restrictions. They can't "ungrant" those later, as there was no provision in the original license for that to happen. IANAL either.
Posted Oct 12, 2005 18:02 UTC (Wed)
by MathFox (guest, #6104)
[Link]
If you succeeded in taking this notification hurdle, the real legal fun will start. You will be asked to pay damages for revocation of distribution rights, people will object to revocation of their rights to use the code, etc. I can not tell beforehand how the lawsuits will work out in all relevant jurisdictions, I don't expect that you'll win the ~100 lawsuits easily.
Posted Oct 13, 2005 6:50 UTC (Thu)
by fyodor (guest, #3481)
[Link]
I was not arguing (nor do I believe) that the GPL can be revoked in the general case. The SCO issue was based on specific language within the GPL clause 5: "You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License." To the best of my knowlege, SCO has ceased distribution of Nmap in accordance with my demand (If anyone catches them still distributing it, let me know).
As for the Nessus issue, I have already sent out a response for the Nmap Security Scanner Project. We aren't planning to follow suit. Nmap has been GPL since its release more than 8 years ago and I am happy with that license.
Posted Oct 21, 2005 20:49 UTC (Fri)
by pimlott (guest, #1535)
[Link]
I figured that merely pointing out the article was from Wikipedia would be enough of a hint for everybody to add that pinch of salt when reading it.Single-company free software
Discussion about revokability of the GPL came up when the nmap author forbid SCO from using his software due to their claiming that the GPL was invalid. Also with DJB's software with it's lack of real license.Single-company free software
I am not a lawyer either, but IMO revocation of a written offer requires at least a written revocation. Thanks to the GPL every owner of a copy of the sourcecode has the right to redistribute and relicense. It won't be trivial to send out revocation notices to all redistributors of your code. It is more difficult than the RIAA and MPAA task to track P2P users, a John Doe lawsuit shouldn't work against someone who legimately distributes your code.Single-company free software
Discussion about revokability of the GPL came up when the nmap author forbid SCO from using his software due to their claiming that the GPL was invalid.
Nmap Non-Revocation
Single-company free software
There just isn't any credible reason to believe in the ability for a pure license to be revoked outside of any terms embodied in that license
Yes there is. And if you google "GPL revocable", you will find additional discussion which makes it clear that the question is far from settled.