Re: [HACKERS] Re: Copyright

At 00:52 29/01/00 -0500, Tom Lane wrote:
>bruc(at)stone(dot)congenomics(dot)com (Robert E. Bruccoleri) writes:
>> However, the real issue with PostgreSQL is not the copyright, but
>> rather the permissions granted to everyone. As long as all the
>> contributors are happy with the permission notice, then all is OK.
>
>I think that's an excellent point that bears underlining. All the
>original code bore a UC Berkeley copyright --- but that didn't make
>anyone unhappy, or stop any of us from doing what we wanted to do
>with the code, because Berkeley's license terms are loose enough
>not to pose any problems.
>
>The license terms are not going to change. Someone suggested switching
>to GPL or LGPL terms, but we cannot do that without (a) violating the
>Berkeley terms, which we are still bound by, and (b) losing many
>contributors who work in commercial settings and would not find a
>GPL'd database usable for their purposes. (Berkeley terms are not
>a problem for someone who wants to use code as a component of a
>larger proprietary system --- but GPL terms are.)

LGPL allows some commercial use, but I agree, they are both inappropriate
agreements, especially if you want to allow people to write a commercial DB
based on PostgreSQL, which is a very good thing to allow.

When you refer to the license terms, do you mean the 'COPYRIGHT' file in
the root directory, or is there something more?

>As long as those terms don't change, adding PostgreSQL Inc (or
>PostgreSQL Nonprofit Copyright Holding Corporation, or anything else)
>to the copyright notices doesn't really change anything, except for
>adding one more line to the boilerplate notice that people aren't
>supposed to strip out of their copies. PostgreSQL Inc can't sell the
>rights to Postgres, because it hasn't got any rights that anyone else
>hasn't got.

Now I'm confused. I thought there was a notice stating that PostgreSQL, Inc
had copy right? If that is the case, then they *can* sell it, and the buyer
could revise the license terms. This is a sort of inverse-Unisys trick:

1. Start with a free open source product
2. Shift the copyright of new code to yourself. The old code remains as it
was.
3. Sponsor and encourage some very nice changes to optimizers, OO support
and RI (naturally shifting copyright of the new stuff your organization)
4. When it's all stable, get taken over by someone who enforces the
copyright on the recent changes, and prevents further distribution.
5. Open source PosgreSQL development falls back to 3 year old sources, with
possible further copyright arguments.

I agree that this is *very* unlikely, but if such a scenario can be
prevented by some simple changes (like a non-prfit organization that has a
clear charter), then it may be worth doing.

> *Anyone* could take the code and start developing it
>independently, just as the current set of developers did with Berkeley's
>code. And if PostgreSQL Inc did something that any significant number
>of developers were unhappy with, that's exactly what those developers
>would do.
>
>So, while I don't have anything against forming a nonprofit organization
>to hold the copyright on behalf of the development team, I really doubt
>that it makes any difference. The thing to keep your eye on and guard
>jealously is the license/terms-of-distribution. If anyone proposes
>mucking with those, THAT is the time to start hollering.

As I said above, I'd like to know where they are, if you mean something
more than the COPYRIGHT file. The big advantage of a separate entity is
that it has a clear charter and no conflict of interest. Disadvantage is
that, I presume, it costs $$$.

----------------------------------------------------------------
Philip Warner | __---_____
Albatross Consulting Pty. Ltd. |----/ - \
(A.C.N. 008 659 498) | /(@) ______---_
Tel: +61-03-5367 7422 | _________ \
Fax: +61-03-5367 7430 | ___________ |
Http://www.rhyme.com.au | / \|
| --________--
PGP key available upon request, | /
and from pgp5.ai.mit.edu:11371 |/