Jeff Teunissen wrote > Henning Makholm wrote: > > [EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > > > > > All the owner can take back is the promise as it applies to new > > > copies. > > > > That is bad enough as it is. It means that once the owner changes his > > mind, we lose the right to make and distribute new modifications: > > I might still have the right to make one modified copy of the work, > > but I don't have any right to copy that one copy further. > > Unless I miss my mark, it means that the assumptions we've made on the > legality of free software are correct. The author cannot change the > license on code you already have; if he does change the license on new > versions, you still have a free version and can fork a new project out of > it. And yes, that means you still have the ability to make copies of that > source, because the license you received the code under allowed you to > copy and modify it as many times as you want.
I agree. One question occurs to me though: what exactly changes for somebody with a previously made copy when the copyright is transferred? I suppose this means that the copyright on the copy has a different owner, but the licence does not change; the only thing that changes is that if he violates the (original) licence he can now be sued by the new copyright owner. This question came up because I am often forced to transfer copyright to journals when they publish articles, and they tend to place draconic restrictions on further distribution (particularly electronically) of the paper, even by myself. May I conclude that if I make sure to distribute the paper with a liberal licence (maybe even GPL) prior to transferring the copyright, they cannot enforce their measures, provided I make sure that everything publically distributed is based on copied made before the transfer? Marc van Leeuwen Universite de Poitiers http://wwwmathlabo.univ-poitiers.fr/~maavl/
