(by the way, text in caps surrounded by *'s is meant to indicate vocal stress,
not volume)
On Sunday 02 September 2007 22:01:18 David Schwartz wrote:
<snip>
However, this is not what is happening here. Jiri has made changes that he has
licensed solely under the GPLv2. This means that he now becomes the
licensor - not Mr. Floeter or Mr. Leffler. *BUT* *ONLY* of the version of the
code containing his changes.
Mr. Floeter *CAN* request that his code be removed from said fork - his code
is solely licensed (AFAICT and IIRC) under the BSD/ISC license and was only
covered by the dual-license because it was integrated into a work that
carried said dual-license. (I'm not sure how well such a revocation would
work in reality, but it is Mr. Floeters right.)...
(Sam Leffler could do the same - but I'm not sure how well that would carry.)
Correct. Doesn't apply in the case of the code in question (unless the changes
that were made are so tiny as to not be copyrightable).
In this case the code is question is a modified version, which means that the
right to distribute said modified version now originates with the person
holding the copyright on the modifications. (Though their right to distribute
the code, in such a situation, is lessened quite a bit by the text of the
license they received the code under)
No misunderstanding, really.
Alan seems to have given a bad example that doesn't apply to the situation
that is being discussed.
Agreed. When re-distributing an un-modified copy of a work. When distributing
a modified work, the "work" has the license that the person who made the
modifications places on it. But individual files and pieces of code will
still retain their original license - this is how it works.
DRH
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
-