conducting a sane and efficient GPLv3, LGPLv3 Review

Luis Villa luis at
Tue Jul 31 23:33:02 UTC 2007

[Warning: possible tilting at windmills ahead.]

On 7/31/07, Wilson, Andrew <andrew.wilson at> wrote:
> Once you understand this, you see the compatibility problem.

I'd suggest that compatibility, while admittedly a pragmatic problem
and something we could discuss for many, many hours and many, many
words, is really not very useful for an OSI analysis of the v3:

* The v3 is already more widely adopted than some other OSI-approved,
but v2-incompatible licenses, so if a license is to be rejected on
proliferation grounds merely because it is incompatible, another
license should be retired instead.

* The OSI (to the best of my knowledge) does not maintain a formal
compatibility database, so a compatibility analysis does not help with

More generally, I'd suggest that this discussion should focus very,
very strictly on whether or not the v3 is OSD-compatible; other issues
are either obviously resolved or otherwise not relevant:

* OSI typically wants to know if the existence of a new license is
justified; I'd suggest that the extensive review process and very
quick adoption of the license indicates very strongly that this is the
case. We could discuss endlessly whether individual members of this
list feel that the license is justified, but that would be a waste of
bits while others go ahead and use the thing.

* the license has already received the most extensive public review of
any license in history, and is finalized, so discussion of potential
flaws in the license (except inasmuch as they may impact
OSD-compatibility) are not useful. Again, time, money, and bits could
be wasted discussing non-OSD-relevant flaws (like the ACT "analysis"),
but that seems fairly pointless.

So, yeah. We could spend literally hundreds of hours discussing
compatibility, 'need', and potential license flaws, and in the process
finish driving away from license-discuss anyone of goodwill but
limited time/energy, but I'd strongly suggest that instead we limit
GPL v3 discussion to the strict question of OSD compatibility and save
each other the time and energy for other pursuits.


More information about the License-discuss mailing list