[legal] Draft OpenID Intellectual Property RightsPolicy for Review

[Gabe Wachob]

> "Gabe Wachob" <gabe.wachob at amsoft.net> writes:
> 
> > Simon-
> > 	The licenses you cite are licenses around software. The licenses
> > discussed in the IPR policy are around specifications. Much of the
> language
> > in GPL simply doesn't make sense (e.g. GPL talks about "source" and
> "object"
> > - simply inapposite here).
> 
> The GPL is used by a few projects for documentation.  For example,
> Debian uses the GPL for their installation guide:
> http://www.us.debian.org/releases/stable/i386/

That's just odd. I don't understand how the license makes any sense in that
situation. 

But the Unicode one is more interesting I think for our purposes...

> 
> > 	Also, I don't think the Debian Free Software Guidelines apply in any
> > sensical way here either - I don't see how the required copyright
> license
> > here restricts using any implementation of OpenID in any field of
> endeavor.
> 
> Right.  But I believe it would restrict Debian from including the OpenID
> standard in their archives (they could distribute it in 'non-free'
> though).

OK, this wouldn't keep me up at night. But if its important to the
community, we could reconsider. 

> 
> The proposed rules would also prevent all OpenID implementations from
> re-using portions of the text in the standard, for documentation, which
> may be a typical real-world scenario.

Actually, I'm not sure that's the case - if you think this is an important
feature, we could certainly "enable it" explicitly. Again, this is not
something I would lose sleep over. 

> 
> > 	Creative Commons is *more* applicable - though even there, the
> > emphasis is on sharing and derivative works - the question is to what
> extent
> > derivatives of OpenID specifications need to be allowed by the copyright
> > policy. It would be a distinct departure from normal standards making
> specs
> > to create a license as liberal as any of the CC licenses.
> 
> Yeah, but I thought OpenID was intended to be more liberal than most
> other standards! :)

It is!

 
> Btw, the Unicode Consortium uses a liberal license for their standards,
> see exhibit 1 of <http://www.unicode.org/copyright.html>.

That one is interesting. However, if you read carefully, you'll see the
following language: 

" Any person is hereby authorized, without fee, to modify such documents and
files to create derivative works conforming to the UnicodeR Standard,
subject to Terms and Conditions herein."

So, you can create derivative works so long as they "conform to Unicode" -
well, I'm honestly not sure what that would mean for the OpenID specs...

> > 	So in short, I think we're actually enabling more openness by
> > explicitly allowing, *without any affirmative action on the part of the
> > original authors* the preparation of derivative works within the
> umbrella of
> > the OpenID community.
> 
> My point is that this is still far from a specification licensed under
> "the most liberal license available".

You could be right, there are other specs out there that allow more freedom
for use/modification/etc of the the *text* of the specs. The thrust of our
efforts in the IPR work has been in enabling *implementers* to implement the
spec. This is mostly an issue of trying to clear the field of patent
encumberances. Copyright in the underlying spec does not typically hold up
implementers. 

I'd ask the rest of the community if they feel the copyright issues Simon
brings up are important to them as well. 

	-Gabe