Re: Licensing and the client/server divide [was Re: License for mixminion dist]

[Zooko <zooko@zooko.com>]

 Nick wrote:
>
> Ah!  No, I envision it differently.  The server runs under a Python
> interpreter of its own, and communicates via sockets, disk spools, and
> other mechanisms with the applications that want to use it.

Oh -- okay.  So what's the "client" code then?  I think it's a thin piece 
of glue that provides a "function invocation API" to application code and 
translates the function invocations into RPC calls to the MM node.

So if the node code were under GPL and the little glue code were under LGPL 
then it should be legal to ship proprietary MM-using apps but not to ship 
proprietary MM-altering apps.

BTW, I just read the GPL again to try to verify if that were the case and 
I couldn't really tell -- it's frighteningly vague by talking about "a work 
based on the Program".  But I'm willing to ...  Ack!  No, dammit, now I'm 
confused again.

The GPL says that any "work based on the Program" must be offered under the 
terms of the GPL, so there's no reason just looking at the text of the GPL 
itself to believe that the above scheme would leave people free to ship 
proprietary MM-using applications.

What a headache.

Considering this kind of uncertainty of the meaning, and considering the 
widespread antipathy for everything associated with RMS and the FSF, I'm 
increasingly in favor of using MPL for everything instead, thus forgoing the 
attempt to prohibit people from shipping proprietary code which uses our core 
source code but alter the Core behavior.

http://zooko.com/license_quick_ref.html

Regards,

Zooko