Public Domain Python

Laurence Tratt tratt at dcs.kcl.ac.uk
Sat Oct 28 14:55:07 EDT 2000


lvirden at cas.org wrote:

[Courageous]
> :I disagree; once you have granted someone the right to create a derived
> :work, and that derived work exists, you have no legal ability to withdraw
> :their ability to continue to own/sell their derived work.
> 
> If that were the case, then a product could never have a license change -
> but they change all the time.  Most license that I've read in fact address
> the issue of future changes to the license; if GPL/OpenSource/BSD/etc. do
> not, then they probably should.
> 
> What seems to make sense logically is that one can chance the license on
> anything that has not yet been distributed, which would include a new
> release of a product (even if said new release had, as the only change,
> the new license).

I think the generally accepted take on this is that essentially a license -
unless it has a "this can be revoked at any time" type clause - applies in
perpetuity to a specific version of a program. Later versions may have
different licences but the old versions licenses still apply to those old
versions. This has allowed eg the OpenSSH project to get hold of an old
version of SSH which had an open license and make their own version forked
from that whilst newer versions of SSH are commercial.


Laurie



More information about the Python-list mailing list