W3C home > Mailing lists > Public > www-dom-ts@w3.org > June 2001

Re: GPL (was: Re: Early XSLT's)

From: David Brownell <david-b@pacbell.net>
Date: Tue, 19 Jun 2001 16:20:27 -0700
To: "Joseph M. Reagle Jr." <reagle@w3.org>
Cc: www-dom-ts@w3.org, plh@w3.org
Message-id: <020d01c0f932$5eb25380$6800000a@brownell.org>
> >Actually if I were contributing, I'd be bothered by the text
> >that Phillipe insisted be signed.
> 
> Phillipe and I have spoken further and new draft text follows below:

Better ...

> Good point; as you can see, the text now specifies the W3C Software License.
> http://www.w3.org/Consortium/Legal/copyright-software-19980720

... except that it doesn't specify a particular version (such as that one),
so again, there's a loophole.  It shouldn't be mandatory to permit W3C
to relicense other peoples' contributions at will, as I commented before.


> If you make a derivative work, you now have two works. ...

I'm curious though what level of change distinguishes a "derivative work"
as opposed to two versions of a single work.  There are Japanese temples
that have been carefully maintained, and look largely the same today as they
did hundreds of years ago -- even though no single board is the same.  Is
today's version the same one?  Is it a derivative?  If such issues truly matter,
I'd hope to see better answers than "litigate it to get the answer".   If they
don't matter, I'm not sure why such language is desirable.


>    All it does is permit the W3C to avoid 
> confused claims of conflicted or joint ownership. (Most software licenses 
> don't even speak of ownership and that whole issue remains rather murky in 
> my reading of them. 

Introducing such new issues -- beyond issues of "who holds the copyrights"
and "what license applies" seems to create even more problems.  What
exactly is "ownership", as distinct from "copyright"??


>    Consequently, I believe the explicitness is useful after 
> seeing work at the IETF come to a stop when a contributor was upset at a 
> decision and tried to yank his contribution from something the whole 
> community had worked on.)

Hmm, that presumes the original licence grant was in fact revocable.

If it weren't revocable, "yanking" (of that type :) would not be possible.
The essence of an Open Source license is that such revocation is not
possible:  everyone has the right to get, run, modify (etc) in perpetuity.


> The intent is not an exchange of ownership on the original work, only for 
> the W3C to own the work it derived (and license them under the same W3C 
> Software License.)

I guess I still don't see why that'd be perceived as an issue.  When the
original non-revocable license was granted, W3C (like every other
source licensee) received sufficient rights to evolve the software as it
sees fit, in perpetuity.  Why does W3C want to be "more equal"?


> >p.s. Separately, yes I would certainly prefer to see tests be under
> >     GPL, or at least LGPL.  The notion of making it easier for anyone
> >     to create proprietary variants of "standards", by subverting such
> >     community efforts, is bothersome.  This seems one of those cases
> >     of giving up liberty, a little bit at a time.
> 
> I hope this is moot now, but I honestly don't understand what the LGPL means 
> in this context.

That's a reasonable question, and things would have to be defined so that
could have an answer before LGPL could reasonably be applied.  Such
issues are one reason to avoid, in general, the "Lesser" GPL.  Though I
suppose a reasonable definitiion of a test suite is as a "library of test cases".

All the more reason to prefer GPL ... despite the current Microsoft offensive
against licenses which don't facilitate such subversion!  :)

- Dave
Received on Tuesday, 19 June 2001 22:41:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 6 April 2009 12:58:44 GMT