[Fwd: [gnu.org #285277] Open Source Initiative Certification for GPL 3.0]

[Matthew Garrett]

On Mon, May 01, 2006 at 11:16:41AM -0400, Russ Nelson wrote:
> Matthew Garrett writes:
>  > b) nobody seems interested in discussing where boundaries should be 
>  > drawn when it comes to patent-related license termination
> 
> If nobody is interested in it, why bother talking about it?

Because there's clearly some degree of disagreement in the community 
(see Debian's rejection of the APSL and Gnome's removal of a dependency 
on APSLed code, for example), but absolutely no sign of leadership from 
any of the bodies that claim to represent the community (either the FSF 
or OSI). Having different standards of "Open Source" helps nobody, 
especially not the people who release code under open licenses and then 
wonder why people won't touch it.

Now, what actually happened in this case is that the community 
reimplemented the APSLed code from scratch (the Avahi project provides 
an LGPLed API-compatible reimplementation of Apple's reference Bonjour 
implementation, which was the code in question) and then moved on. If 
the community is unwilling to work with an OSI-approved license, then 
what is the point in marking it OSI-approved? The mark only means 
something if it agrees with community standards[1], otherwise it's just 
marketing. And in the area of patent-related termination clauses, it's 
not clear that the OSI does agree with community standards. How can we 
fix that?

[1] I accept that Debian is likely to be more hardline than the 
community in general, and that having a direct relationship between 
OSI-approval and DFSG-free licenses is unlikely to happen.
-- 
Matthew Garrett | mjg59 at srcf.ucam.org