Netapp license approval retraction with apologies

Karsten M. Self kmself at
Wed Feb 20 09:22:12 UTC 2002

on Tue, Feb 19, 2002, Matthew C. Weigel (weigel at wrote:
> On Tuesday, February 19, 2002, at 08:29 p, Topper, Anthony wrote:
> >We understand the desire to have one "Common Public License" with 
> >no messy
> >variations. We think it is an excellent license, hence our desire 
> >for our own small variation of it. The issue that had our 
> >attorney's concerned was IBM as the sole steward of change and NY 
> >as the applicable law.
> Maybe I'm crazy, but it seems like the OSI should encourage or 
> require licenses that have such specific language to be submitted 
> in template form, or to at the least 'approve' the template - and 
> require that derivatives of the template, modified as templates are 
> intended to be modified, not be restricted.
> Otherwise, there is the 'one license per package' problem that the 
> OSD already attempts to prevent.

I've certainly encouraged "templatized" licenses over the past several
years.  There are several licenses which seem to be well suited to this
model, including SD/MIT, NPL, and the revised, generic, IBM PSL.

The OSI's purview is approving licenses which conform to the OSD.  As
such, encouraging or requiring specific forms of licenses may not fall
neatly into its lap.  But providing a guideline (template template?) for
templateizing a license  might itself go a long ways toward this goal.

Again, I've suggested that a "reference + implementation" model of
licenses might work.  The reference is the standard, jointly held
version of a license.  One of the major concerns in licensing is ceding
revision authority to another entity.  To get around this,
implementations of a reference license would reserve this right for the
specific party issueing the license.  Future revisions would, ideally,
be worked out among the parties using implementations of the reference
license.  A revised reference would be issued, and each party, acting on
its own revision clause, could issue an updated implementation of the
reference license.

This would allow for the option of forking licensing terms (if this was
felt necessary), preserving the right to independent evision, while at
the same time creating a standard licenseing base from which to work.


Karsten M. Self <kmself at>
 What part of "Gestalt" don't you understand?       There is no K5 cabal

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <>

More information about the License-discuss mailing list