Re: Congratulations and Some Thoughts - Mailing list pgsql-general

From Tom Lane
Subject Re: Congratulations and Some Thoughts
Date
Msg-id 15379.957909087@sss.pgh.pa.us
Whole thread Raw
In response to Congratulations and Some Thoughts  (Daniel Freedman <freedman@ccmr.cornell.edu>)
List pgsql-general
Daniel Freedman <freedman@ccmr.cornell.edu> writes:
> On a final note, I would like to suggest a discussion concerning the legal
> framework of the PostgreSQL development effort.  Tom mentioned that
> upcoming discussions would focus on the Postgres license, and I agree with
> him that such discussions can be delayed.  However, I think it would be
> fruitful to consider at the same time the tangential creation of a legal
> corporation in The PostgreSQL Foundation, to be modelled (to some degree,
> with developer and community input, of course) after The Apache Software
> Foundation, or a similar open-source consortium, such as Debian governance
> or KDE foundation.

Yes, this is something that we intend to discuss in parallel with
license-tweaking.  The idea has been mentioned before, but previously
the main argument for having a foundation was "to hold the copyright".
We now realize it's doubtful that that would actually mean anything
legally, unless we require all contributors to sign written copyright
assignments a la Free Software Foundation practices.  (In core's
discussions of this, we've guessed that that approach is too
bureaucratic to be likely to go over well with the Postgres community,
although of course we're open to being told otherwise.)  Still, no one
feels quite comfortable with having "copyright PostgreSQL Inc"
plastered on the code, since ultimately PostgreSQL Inc is a for-profit
corporation.  "Copyright Nonprofit PostgreSQL Foundation" would probably
make everyone happier, even if it wouldn't mean much should push come to
shove.

Better (IMHO) reasons for having a foundation would be as a means of
raising and spending money in direct support of the open-source effort:
say, to fund developer meetings, help individual developers purchase
hardware or books or those bloody-expensive spec documents, etc etc.
(Yet that could in itself create unhappiness, if money is being spread
around and none of it comes your way.  So it's not entirely a win.)

This is all very much blue sky at this point but we want to start some
focused discussions soon.

            regards, tom lane

pgsql-general by date:

Previous
From: Andrew Sullivan
Date:
Subject: Re: USMARC and postgresql?
Next
From: Bill Barnes
Date:
Subject: RE: problems with copy command