Re: GPL code reuse questions

[Arne Babenhauserheide]

On Wednesday 04 August 2010 20:59:50 olafBuddenhagen@gmx.net wrote:
> Hi,
>

> On Wed, Jul 21, 2010 at 12:08:45AM +0200, Samuel Thibault wrote:
> > No
need to care about license stuff: fold that into a separate
> > process, and
voilà :)
> 
> It's not that simple. What constitutes a derived work cannot
be decided
> based on process boundaries alone. After all, processes (i.e.
adress
> spaces) are just a technical detail. What really matters is whether
the
> components interact "at arms length" (as the GPL FAQ calls it) -- it
>
would be ridiculous to treat each Hurd server as an independent work.

The
term in the GPL is „intimate control flow“. I can give an example of my RPG
work for that: 

Our ruleset is GPLv3. Everything which has intimate
interaction with the rules (for example uses them to create new rules) is
likely to have to be GPLv3, too (not lawyer checked, though). Something
which only passes data to the rules (for example character stats or world
description) doesn’t have to be free. 

This is for stuff which has no
connection at all to computer processing, since it uses highly intelligent
agents for interpreting its very highlevel instructions, using one six sided
random number generator per agent :)

Since the GPL can be interpreted for
that and has meanings beyond the realm of computers, the exact interaction
methodology shouldn’t matter too much. 

If you really want to write an
unfree server for the Hurd, put all logic which interacts with free
components into one server, write a second server with all your unfree stuff
and design them in such a way that they only exchange simple data.


(caution: not lawyer-checked)

Interesting question: what about RPC
interfaces in webservices? 

Best wishes, 
Arne
--
Ein Würfel System -
einfach saubere Regeln: 

- http://1w6.org

signature.asc
Description: This is a digitally signed message part.