BSD-like licenses and the OSI approval process

[Matthew Flaschen]

Chris Travers wrote:
> On 10/16/07, Matthew Flaschen <matthew.flaschen at gatech.edu> wrote:
> 
>> Doesn't that risk the eventual possibility of a court having to decide
>> whether something is OSD-compliant (to settle a trademark lawsuit)?...
> 
> What makes you think that the OSI could even justify or win such a
> lawsuit?

I didn't say either of those things.  I said OSI /might/ consider a
lawsuit, if e.g. a e.g. FooCompany insisted the clearly proprietary
FooLicense in fact "OSI-Certified" or met "OSI License Class X" (to
borrow your phrase).  OSI-Certified is clearly a trademark, and I
believe OSI License Class X would be enforceable as a trademark too.

> To be clear, it is not a lawsuit I am worried about but
> rather public pressure on the part of a relatively powerful
> organization in the community.

When has OSI ever put pressure on vendors of permissive software with
unapproved (but compliant licenses)?

> Does the OSI have any sort of trademark rights to the term "open
> source?"

Maybe, maybe not.

  If not, then this isn't at all a problem.  (I am not sure
> that even a registered trademark would be valid as it seems to have
> widespread generic use predating the trademark, but IANAL.)

You are incorrect.  There was little use to refer to software or source
code before OSI.

> In short approval of a license class could be done in such a way as to
> say "We, at the OSI, have no problem with people calling these
> licenses open source?

How do you define "these licenses" (i.e the licenses in the class)?
This is circular reasoning.

> So the alternative in your mind is to certify every reasonable BSDL
> variant individually?

The ideal choice is for them to settle on one of the approved licenses.
 It's frustrating to hear, "We know it's legally the same as
BSD/MIT/Fair/etc., but all the same we want to use our own license"

> Or do we insist on PostgreSQL not being "Open
> Source?"

Neither.  There are three choices:

1. PostgreSQL adopts the approved BSD/MIT/etc. license.
2. PostgreSQL says in good faith "We think we're open source", and OSI
doesn't object because it looks (from a distance) like they are.
3. PostgreSQL (or any other interested party) chooses to submit their
license for an up-or-down vote.

OSI is not going to suddenly condemn Postgre after 10 years (for both
OSI and Postgre!).  You are inventing an issue where there is none.

>>  People
>> can already abuse the trademark, but it's obvious from the license list
>> they're not approved and thus it's unlikely to get to court.
> 
> Which trademark?

OSI-Certified.

> More along the lines of "Meets the standards of OSI License Class X."

That's still an important trademark, which OSI shouldn't open to abuse.

Matt Flaschen