Re: [Oopic-compiler-devel] open source 'license' terms

[Neil D. <nd...@us...>]

Brian wrote:
> For this project, I'm not partial.  However, we might want to check
> with Scott first to see if there is any information we are receiving
> from him (code, etc.) that we do *not* have the right to release under
> an open-source license (i.e., there are no patent issues, IP rights
> etc.).

Definitely agreeed!!  From Scott's attitude so far he has seemed pretty
open and flexible.  However it would be only courteous to check with him,
and having a comment from him agreeing to our licence terms ought to shield
us from hassles in the future.

> Also, note that the BSD license allows 3rd parties to modify the code
> base without having to release modified source code (i.e., release
> under a closed-source license, much like Microsoft has done with major
> portions of BSD-licensed network/TCP/IP stack code).   Some folks
> favor this appropach in that the BSD license doesn't "taint" future
> forks by requiring release under the same open licensing scheme.  Of
> course, the GPL purists believe this simply encourages people to steal
> the hard work of others without requiring them to "give back" to the
> open source community.

I tend to be cautious about BSD-style licenses.  I think I would feel
aggrieved if I contributed a load of code, free, to the open-source world,
and then some company subsequently took my work and released it without the
code, and sold it.  If my understanding of the BSD licence is correct, it
permits that - am I right?

> A license worth looking at is the Mozilla Public License...this one
> has a provision that in the event of patent-based challenges, the
> offending code can be removed from under the provisions of the MPL.
> With the GPL, it's pretty much an all-or-nothing deal:  If any part of
> your code is found to be infringing, the entire code base is at risk
> (just check out http://www.groklaw.com to see this attack on the GPL
> in action!).

That does sound worth considering.  I can imagine parts of the code
generation phase being tied up with IP or patents etc (but probably not),
whereas the parser side would be entirely unconnected. Preserving the
ability to compartmentalise what we do could be valuable.

Neil
-- 
Neil Durant
<nd...@us...>