For Approval: GPLv3
matthew.flaschen at gatech.edu
Fri Aug 31 07:13:18 UTC 2007
Chris Travers wrote:
> I think the GPL3 is very confusing here because on one hand it seems to
> prohibit such an agent negotiation role (sublicensor) but at the same time,
> try to put one in effect. I wonder whether the right to remove these
> permissions could be held unconscionable under contract law...
Maybe, instead of wondering, you could provide some actual evidence. I
don't see how something BSD and many other licenses allow you to do
could be considered unconscionable.
> This seems pretty sketchy to me. Part of the problem is that the license
> is a legal contract between the licensee and the licensor. Alteration on
> the part of a third party might seem problematic.
Not if the original licensor gives said third party permission to do so.
> Note that some other varients of the BSDL do not specifically permit
> sublicensing, which would be closer to your interpretation.
All BSD-like licenses allow sublicensing. It's irrelevant whether they
use the actual word.
> Part of the problem is that you are saying that the GPL 3 grants the right
> to misrepresent what someone says you can do under copyright in a *legal*
No, it absolutely does not. I have said repeatedly that legal notices
must be maintained intact. You can not misrepresent what the original
> True, but the fact remains that the GPL is a contractual agreement (if
> accepted by excersizing rights in it) which exchanges those rights it grants
> for certain responsibilities. Changing those responsibilities without
> adding copyrighted material to the affected portions seems dangerous.
Dangerous how? The GPL is a copyright license that gives permission to
remove extra permissions. How can something you are legally authorized
(in unmistakable terms) to do be dangerous?
(Note that the GPLv3 dropped all claims not to be a
> contract on the advice of lawyers,
Where exactly does GPLv2 say explicitly it is not a contract? The
situation hasn't changed at all in GPLv3. As far as I know, the FSF
stills views it as a license, but this isn't directly stated in either
> How can you read the ability to remove permissions other than a claim to
> restrict anyone?
If I remove permissions before distributing code to you, I'm simply
saying I'm not passing on these permissions. That doesn't mean you
can't get the permissions directly from the author for unmodified code.
More information about the License-discuss