[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: turck-mmcache license violation?



On Thu, 11 Nov 2004 20:12:42 -0200, cascardo wrote:

> One could claim that php4 is part of the operating system, just like
> they do with OpenSSL. That is nuts!

One could claim anything.  However, most people don't consider php4 part
of the OS.

> 
> Sorry for introducing a reason for a flame.
> I'd just like to say that I think the if one line is to be crossed, that
> should mean that we should ask for the author permission, which would
> make the license non-free. Since turck-mmcache (I've just checked) is a
> PHP optimizer, seems reasonable that the author had the intention to
> allow it to be linked to php4, since I guess the program would not work
> without it.
>

I'm not sure I understand what you're saying; how would asking
permission make the license non-free?  And, while the author may have
intended the program to be linked to php4, he has not given explicit
permission for us to do so.  Quite the opposite, in fact; he chose a
license that dictates a set of strict rules for redistribution in binary
form.  So, while we can certain redistribute in source form, and require
the user to compile/link against php4, we cannot (currently)
compile/link against php4 and distribute those binaries until we get
some sort of (legally binding) permission from the author.

> Unfortunately, that is not implied when some one writes a patch to a GPL
> application and another person writes a SSL interface to it. That
> happened to me once and I replied to the message in the development
> mailing list saying that they should not include my patches if they
> wanted to distribute it.
> 

That's correct.  All copyright holders must agree when changing the
license on a piece of software.  That includes significant patch
submissions.  For the turck people to either change the license to LGPL,
or add some sort of exception clause for php (and everything php links
to), it would require the author to get the permission of everyone who
submitted a significant amount of code.  If it's just one person hacking
it, that shouldn't be too difficult.  If it's a team, w/ a lot of people
committing changes, it becomes more difficult.





Reply to: