On Tue, Apr 23, 2013 at 09:27:16AM -0400, Stephen Gallagher wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Ping. I'd like to get an answer on this soon, if possible.
The answer is: yes, this arrangement is acceptable.
On Thu 04 Apr 2013 01:07:53 PM EDT, Stephen Gallagher wrote:
> On 04/04/2013 11:41 AM, Stephen Gallagher wrote:
>> We have some concerns about whether our licensing will be in
>> conflict in the OpenLMI project. I'll attempt to describe the
>> topography. Tomas, please correct me if I have misrepresented
>> We have three components in question here:
>> 1) The OpenPegasus CIMOM. This is a daemon licensed under MIT.
>> It has a plugin architecture that enables it to dlopen() plugin
>> modules and use them. (Dynamic linking by dlopen)
>> 2) The SBLIM providers. These are a set of CIMOM plugins
>> licensed under the MPL that provides some common, standard
>> interfaces through OpenPegasus.
> Sorry, important distinction. This should have been EPL (as I
> correctly mention below).
>> 3) The OpenLMI providers. These are plugins licensed under
>> LGPLv2+ that consume some of the functionality provided by the
>> SBLIM providers, but which do so by means of local communication
>> through the OpenPegasus CIMOM.
>> Our question here is whether this is an acceptable arrangement.
>> I see two potential pitfalls that I would like (hopefully) to
>> have dispelled. A) I know it is acceptable for an MIT application
>> to link separately to an EPL library and an LGPL library.
>> However, is the resulting linked chimera also permissable, given
>> that the two libraries themselves do not have a direct link? B)
>> Is the indirect link between the OpenLMI providers (LGPL) and the
>> SBLIM providers (EPL) acceptable, given that it does so only
>> through interfaces provided by OpenPegasus (MIT) and not through
>> direct function calls?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.13 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
-----END PGP SIGNATURE-----