Re: cc65 licensing
Ullrich von Bassewitz writes:
>> The usual presumption is that someone who distributes the code does so
>> (a) as the original author, (b) in accordance with some license, or
>> (c) because the code is in the public domain. If you distribute code,
>> it is your responsibility that licenses on it are clear. When you mix
>> code under two licenses, *you* -- and not re-distributors -- are
>> responsible for identifying which code falls under which license.
>
> That is definitely wrong. There is no difference between "distributing" and
> "re-distributing". Everyone distributing code needs to check and comply with
> the license that covers that code. Just because I'm distributing code does not
> make it safe for you to do that, too. You cannot just take code from
> somewhere, distribute it violating a license, and then come up with the excuse
> "I'm just re-distributing it".
There is an important difference between distributing and
redistributing: Only the author can be the original distributor, so he
may give the code to anyone he wishes. All redistributors have to
have license from the original author(s).
The code you (re-)distribute is partly under the new license and
partly under the old license, but the source files (for example, those
in the cc65-2.10.1/src/cc65 directory of cc65-sources-2.10.1.tar.bz2)
refer only to the new license. You clearly don't think that means
anything further. Benjamin Cutler thought that the files are governed
by the license described in them; I would probably make the same
conclusion.
It is even more confusing that the source files mention the new
license but your non-source-code references to the old license say you
use the old license for the compiler files. This confusion is
compounded by the more restrictive license being buried in files with
names like "doc/newvers.txt" or "src/cc65/copyleft.jrd" rather than
being clearly visible in the top level of the tarball.
At best, it is misleading to write your new license in the files
without mentioning that you consider them to also or instead be under
a different, considerably more restrictive, license. It is also
misleading (again at best) for you to claim copyright to those files
without mentioning that John Dunning owns the copyright on parts of
them.
Michael Poole
Reply to: