[Brad Hards <bhards@xxxxxxxxxxxxxx>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Fri, 22 Nov 2002 06:58, Ronnie Sahlberg wrote:
<snip>
> After thinking it over I like the current licence and can currently not see
> any real
> benefits at all from changing it. I can only see drawbacks.
<snip>
> A much simpler solution that probably works well for everyone:
> Companies that can not or wants not release decoders for ethereal have
> several other, for them, much better options they might want to consider.
> They can develop them as plugins for tools like Sniffer, GOSS etc and
> distribute it to those tools instead.

I guess Ethereal without Ronnie wouldn't be good, and my alternative of using 
a "restricted patent license" wasn't possible, so I have another idea.

- - - - -

Alternative Proposal - Another "plug-in" interface.

Instead of changing the license to allow exceptions for the current plug-in 
interface, add a _second_ interface that allows decoders to run in a seperate 
address space, with communications through a defined, checked interface. This 
might be a socket() type API, although that isn't necessary.

This avoids the need for licensing changes (consider the normal client server 
type architectures - using a socket() to a machine running a GPL'd server 
doesn't make the client under the GPL).

This avoids the need for a "click-through", since Ethereal will accept 
anything over that API, but _everything_ will be checked. (Again, think 
client-server).

The downside is that such an interface needs to be documented and build 
(ideally in that order:).

- - - - -

Comments, thoughts, flames?
 
Brad
- -- 
http://linux.conf.au. 22-25Jan2003. Perth, Aust. I'm registered. Are you?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE93TsRW6pHgIdAuOMRAn9jAJ9/Bt8Mv22obVE3EdVh6PjlXaHgugCfWtko
9gi2ODsNwyB311CqXeEQPQE=
=7/7D
-----END PGP SIGNATURE-----