Обсуждение: Copyright

This came down with my latest update and confused me quite a bit:

* Portions Copyright (c) 1996-2000, PostgreSQL, Inc

Since when do *they* own the copyright? Whatever happened to the good old
PostgreSQL Global Development Group? Assigning the copyright of an open
source project to a commercial entity is something unprecedented, and
quite frankly (without having anything at all against PostgreSQL, Inc) I
am not so excited about it.

-- 
Peter Eisentraut                  Sernanders väg 10:115
peter_e@gmx.net                   75262 Uppsala
http://yi.org/peter-e/            Sweden

[Charset ISO-8859-1 unsupported, filtering to ASCII...]
> This came down with my latest update and confused me quite a bit:
> 
> * Portions Copyright (c) 1996-2000, PostgreSQL, Inc
> 
> Since when do *they* own the copyright? Whatever happened to the good old
> PostgreSQL Global Development Group? Assigning the copyright of an open
> source project to a commercial entity is something unprecedented, and
> quite frankly (without having anything at all against PostgreSQL, Inc) I
> am not so excited about it.

That is what I was told by Marc.  If we want to discuss it, go ahead. 
Such a change was planned in late December, though there was no
discussion who gets the copyright.  I believe he used PostgreSQL Inc.
because it is a legal entity, vs. the development team, which is not.

--  Bruce Momjian                        |  http://www.op.net/~candle pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026
 

Bruce Momjian <pgman@candle.pha.pa.us> writes:
> ... I believe he used PostgreSQL Inc.
> because it is a legal entity, vs. the development team, which is not.

Right.  Although IANAL, I'm pretty sure it's pointless to slap a
copyright notice on something unless the copyright names an actual
legal entity (one which could go sue an infringer, if necessary).
The development team is not a person, corporation, or partnership,
so in the eyes of the law it doesn't exist.

I seem to recall some discussion of creating a separate legal entity
to hold the copyright, but offhand I don't see what it buys us
except more paperwork.  The same people (ie, the core developers)
would have the final say over what either that entity or PostgreSQL,
Inc does, so what's the difference?
        regards, tom lane

On Fri, Jan 28, 2000 at 05:40:22PM -0500, Tom Lane wrote:
> Bruce Momjian <pgman@candle.pha.pa.us> writes:
> > ... I believe he used PostgreSQL Inc.
> > because it is a legal entity, vs. the development team, which is not.
> 
> Right.  Although IANAL, I'm pretty sure it's pointless to slap a
> copyright notice on something unless the copyright names an actual
> legal entity (one which could go sue an infringer, if necessary).
> The development team is not a person, corporation, or partnership,
> so in the eyes of the law it doesn't exist.
> 
> I seem to recall some discussion of creating a separate legal entity
> to hold the copyright, but offhand I don't see what it buys us
> except more paperwork.  The same people (ie, the core developers)
> would have the final say over what either that entity or PostgreSQL,
> Inc does, so what's the difference?
> 

Ah, corporate assets? What if someone bought out PostgreSQL, Inc., for
a huge sum of money. They'd get control of anything PostgreSQL, Inc.
has control over. One can't buy out a non-profit organization.

Ross
-- 
Ross J. Reedstrom, Ph.D., <reedstrm@rice.edu> 
NSBRI Research Scientist/Programmer
Computer and Information Technology Institute
Rice University, 6100 S. Main St.,  Houston, TX 77005

At 17:40 28/01/00 -0500, Tom Lane wrote:
>Bruce Momjian <pgman@candle.pha.pa.us> writes:
>> ... I believe he used PostgreSQL Inc.
>> because it is a legal entity, vs. the development team, which is not.
>
>Right.  Although IANAL, I'm pretty sure it's pointless to slap a
>copyright notice on something unless the copyright names an actual
>legal entity (one which could go sue an infringer, if necessary).
>The development team is not a person, corporation, or partnership,
>so in the eyes of the law it doesn't exist.
>
>I seem to recall some discussion of creating a separate legal entity
>to hold the copyright, but offhand I don't see what it buys us
>except more paperwork.  The same people (ie, the core developers)
>would have the final say over what either that entity or PostgreSQL,
>Inc does, so what's the difference?

Just curious, but why was this route chosen at all? The copyright of
original code presumably resided with the original developers (who may have
assigned it somewhere), and the copyright for modifications would reside
with their authors, who also have to assign it to PostgreSQL, Inc (in
writing), if it is to be binding (at least where I come from).

Wouldn't a simpler solution be to release PostgreSQL on a GPL or LGPL or
similar which requires the authors who wish to contribute to do the same?

I must admit I am a little paranoid about what amounts to a concentration
of power. I have no idea who owns PostgreSQL, Inc, but let's suppose the
owners die and it is found that they are in debt. Then the debtors come
along and try to work out how much the copyright of PostgreSQL is worth. It
could get very messy.

Similary, let's just assume the owners get bored, move on to other things
etc (like the original developers did). New developers may be held back by
doing work for these, now, faceless people. Similaryl, the 'old guard' may
not see the new developers as worthy of the mantle, so might withold the
code (this *could* be good).

As I said, I'm paranoid, but it doesn't seem to be a good way to do an open
source project. There also seems to be a potential conflict of interest, if
PostgreSQL is a company that also does consulting and customization of
PostgreSQL.

Just my 0.02c, and I'm sure that the current owners have the best of
intentions, but most legal problems (eg. GIF/LZW/Unisys) are born out of
people changing their minds about something.


----------------------------------------------------------------
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   |/

Philip Warner wrote:
> 
> Just curious, but why was this route chosen at all? The copyright of
> original code presumably resided with the original developers (who may have
> assigned it somewhere), and the copyright for modifications would reside
> with their authors, who also have to assign it to PostgreSQL, Inc (in
> writing), if it is to be binding (at least where I come from).

I'm curious about this as well.  I have been under the impression that
the only barrier to someone taking postgresql and making a company out
of it, supporting and shipping postgresql, would be satisfying
whatever the original (Berkeley?) copyright terms were.  I thought the
"leverage" that the core group holds here is simply that nobody else
has the technical familiarity with the software, and thus nobody else
could support it as well.

Does the core group, or Postgresql, Inc., or anyone else for that
matter, have any legal ownership/licensing rights over postgresql
beyond UCB?

Cheers,
Ed Loehr

Unless there are agreements in place to transfer copyright, the
copyright to the PostgreSQL source code belongs to *ALL* the authors
of it. Putting a copyright notice with PostgreSQL, Inc. is OK because
PostgreSQL, Inc. is one of those copyright owners, but anyone else who
contributed is also entitled to put a copyright notice on the
software. The fairest course of action is to reference the complete
list of contributors when the copyright is mentioned because they are
the true copyright owners.

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.

+----------------------------------+------------------------------------+
| Robert E. Bruccoleri, Ph.D.      | Phone: 609 737 6383                |
| President, Congenomics, Inc.     | Fax:   609 737 7528                |
| 114 W Franklin Ave, Suite K1,4,5 | email: bruc@acm.org                |
| P.O. Box 314                     | URL:   http://www.congen.com/~bruc |
| Pennington, NJ 08534             |                                    |
+----------------------------------+------------------------------------+

> > I seem to recall some discussion of creating a separate legal entity
> > to hold the copyright, but offhand I don't see what it buys us
> > except more paperwork.  The same people (ie, the core developers)
> > would have the final say over what either that entity or PostgreSQL,
> > Inc does, so what's the difference?
> > 
> 
> Ah, corporate assets? What if someone bought out PostgreSQL, Inc., for
> a huge sum of money. They'd get control of anything PostgreSQL, Inc.
> has control over. One can't buy out a non-profit organization.

I am worried about that too. Maybe we need an organization something
like the Apache Software Foundation to hold the copyright.
--
Tatsuo Ishii

On Fri, 28 Jan 2000, Ross J. Reedstrom wrote:

> On Fri, Jan 28, 2000 at 05:40:22PM -0500, Tom Lane wrote:
> > Bruce Momjian <pgman@candle.pha.pa.us> writes:
> > > ... I believe he used PostgreSQL Inc.
> > > because it is a legal entity, vs. the development team, which is not.
> > 
> > Right.  Although IANAL, I'm pretty sure it's pointless to slap a
> > copyright notice on something unless the copyright names an actual
> > legal entity (one which could go sue an infringer, if necessary).
> > The development team is not a person, corporation, or partnership,
> > so in the eyes of the law it doesn't exist.
> > 
> > I seem to recall some discussion of creating a separate legal entity
> > to hold the copyright, but offhand I don't see what it buys us
> > except more paperwork.  The same people (ie, the core developers)
> > would have the final say over what either that entity or PostgreSQL,
> > Inc does, so what's the difference?
> > 
> 
> Ah, corporate assets? What if someone bought out PostgreSQL, Inc., for
> a huge sum of money. They'd get control of anything PostgreSQL, Inc.
> has control over. One can't buy out a non-profit organization.

One can't buy out open source software either ... 

Marc G. Fournier                   ICQ#7615664               IRC Nick: Scrappy
Systems Administrator @ hub.org 
primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org 

At 12:53 29/01/00 +0900, Tatsuo Ishii wrote:
>> > I seem to recall some discussion of creating a separate legal entity
>> > to hold the copyright, but offhand I don't see what it buys us
>> > except more paperwork.  The same people (ie, the core developers)
>> > would have the final say over what either that entity or PostgreSQL,
>> > Inc does, so what's the difference?
>> > 
>> 
>> Ah, corporate assets? What if someone bought out PostgreSQL, Inc., for
>> a huge sum of money. They'd get control of anything PostgreSQL, Inc.
>> has control over. One can't buy out a non-profit organization.
>
>I am worried about that too. Maybe we need an organization something
>like the Apache Software Foundation to hold the copyright.

This sounds like a very good idea; I've sent some very general questions to
the Apache people asking about the issues (and costs) involved in setting
up the ASF - it'll be interesting to see what they say (if anything). If
anyone on the list, actually knows something about process, that would be
great too. 

Any US lawyers out there, by any chance?

I've often thought of donating funds for PGSQL, especially for features I
would like (like optimizer hints), but in the longer term, setting it up on
a good legal footing is probably a far more substantial contribution, and
far more needing of funds. Does anybody else feel the same way (monetarily
speaking)?




----------------------------------------------------------------
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   |/

At 01:02 29/01/00 -0400, The Hermit Hacker wrote:
>> 
>> Ah, corporate assets? What if someone bought out PostgreSQL, Inc., for
>> a huge sum of money. They'd get control of anything PostgreSQL, Inc.
>> has control over. One can't buy out a non-profit organization.
>
>One can't buy out open source software either ... 

This depends on the definition of open source; I think Netscape is
OpenSource these days, but I doubt if anyone would be allowed to take the
sources and create "Petscape: The Dog's Browser". But, as far as I know, I
could go away tomorrow with the PostgreSQL sources and create "Mostgress",
so long as I did not try to sell it. AFAIK, I might even get away with
selling it, but it's not really the point.

What matters is that it remains open, free, and modifyable by anyone, for
any purpose.

This does bring up the larger question of what happened to the original
copyrights? Is this covered in Bruce's upcoming book?


----------------------------------------------------------------
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   |/

> What matters is that it remains open, free, and modifiable by anyone, for
> any purpose.
> 
> This does bring up the larger question of what happened to the original
> copyrights? Is this covered in Bruce's upcoming book?

The confusing part is that the Berkeley copyright allows the code to be
integrated into any product, so you can call it Momgres and close the
source, and there is nothing we can do about it.  At least that is the
meaning of the Berkeley copyright.

Many people have taken Berkeley code, added it into their kernels, and
print Berkeley when the system boots up, and they have satisfied the
copyright.

What is quite confusing to me is that though we have copyrighted the
code now, we don't have a statement saying what that copyright allows
people to do with the code.  And with no statement, we are basically
doing an "all rights reserved" thing, I think, which is certainly not
what is intended.

In fact, there can be an argument that we were never open since 1996
because we didn't have any copyright covering years after 1996, and
there is an implied copyright to all works, even if not stated.

Now, I don't mean to suggest that this was intended, but the more I
think about it, the more I wonder.  We have never enforced the
copyright, so there may be no legal basis for it, but I believe there is
some muddy water here.

This all started because I asked about copyright issues around Christmas
(as someone reminded me), and we agreed to copyright the code via
PostgreSQL Inc.  Now that we have done that, it seems we may have a
little more to do.

--  Bruce Momjian                        |  http://www.op.net/~candle pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026
 

bruc@stone.congenomics.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.)

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.  *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.
        regards, tom lane

At 12:53 PM 1/29/00 +0900, Tatsuo Ishii wrote:
>> 
>> Ah, corporate assets? What if someone bought out PostgreSQL, Inc., for
>> a huge sum of money. They'd get control of anything PostgreSQL, Inc.
>> has control over. One can't buy out a non-profit organization.
>
>I am worried about that too. Maybe we need an organization something
>like the Apache Software Foundation to hold the copyright.

A couple of points (as an outsider)...

I have no doubt that the organization of PG, Inc and the copyright
statements which reflect that are done in good faith.

If the distribution license is explicit and "loose", the copyright
issue shouldn't really matter much, i.e. the distribution terms
under the existing license are such that you can do whatever you
want to the code including productize it.

These terms supercede copyright in the sense that the copyright
says "we assert ownership", and having said that, the license
says "we owners say that you can do what you want".

So...maybe the concerns shouldn't be focused so much on the 
copyright issue as in doing whatever's necessary to make sure
that the license terms are perpetuated?

The BSD license under which it is distributed would seem to
do that...each distribution today goes out under that, unless
that's changed which I don't think it has.

The questions about "What happens when the postgres, inc
owner(s) dies?" is valid, and maybe some sort of protection
can be built up that ensures that postgres is distributed
under the same loose license it is today.  Obviously the
U can make that claim for most of the code anyway, but
PG development is also an international phenomena so it
isn't clear how strong that claim is.  



- Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert
Serviceand other goodies at http://donb.photo.net.
 

At 00:42 29/01/00 -0500, Bruce Momjian wrote:
>> What matters is that it remains open, free, and modifiable by anyone, for
>> any purpose.
>> 
>> This does bring up the larger question of what happened to the original
>> copyrights? Is this covered in Bruce's upcoming book?
>
...etc...
>
>This all started because I asked about copyright issues around Christmas
>(as someone reminded me), and we agreed to copyright the code via
>PostgreSQL Inc.  Now that we have done that, it seems we may have a
>little more to do.

It's probably a little more than just putting the notices in the code. In
Australia, copyright rests with the writer unless it is explicitly granted
in writing to another entity by the author, or unless a singed contract of
service exists which grants copyright of works performed directly in the
course of performing those service.

If the US & Canada are similar, the upshot of this is that you'll need to
get the contributors (and Berkely) to sign their copyrights over to
PosgreSQL, Inc, before you can lawfully place those copyright notices in
the code. I would *guess* that there would be some resistance to this.

It's a hell of a lot simpler to leave the copyright with the authors, who
then grant usage under a GPL/LGPL or whatever. Or, set up an arms-length
entity designed to promote free distribution of PostgreSQL that authorsd
may feel happy about assigning their copyrights to.

Again, just my 0.02c, but copyright is a very thorny issue designed to
protect authors - it's a pity to take that away from them. I am very happy
that the copyright of the code I have contributed to, eg, the GD graphics
library remains with me, but that it can be freely used & distributed in
GD. I would not want to sign my rights to that code away, then be told by
the person that buys GD from Tom Boutell that I am no longer permitted to
use the code I wrote.



----------------------------------------------------------------
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   |/

At 12:42 AM 1/29/00 -0500, Bruce Momjian wrote:

>In fact, there can be an argument that we were never open since 1996
>because we didn't have any copyright covering years after 1996, and
>there is an implied copyright to all works, even if not stated.

Licenses compliment copyright, pure and simple.  In other words, as
a photographer (http://donb.photo.net) I defend my copyright, yet
I give permission (license) students and a wide variety of other
people to make free use of my work.  Yet I still charge other 
entities for this right.  

When I license my work for use, I in no way diminish my copyright.
This is true even if I release it into the public domain, i.e. if
my license says "you can use this for whatever you want!".  I still
don't waive my copyright (though I've shot my foot in terms of 
getting damages for use, because if you say folks can use your
photos for free you can't expect to get paid when they accept your
license terms for free use!)

License...copyright...not the same.

>Now, I don't mean to suggest that this was intended, but the more I
>think about it, the more I wonder.  We have never enforced the
>copyright, so there may be no legal basis for it, but I believe there is
>some muddy water here.

Last time I looked (which I admit was like 15 years ago) enforcement of
copyright wasn't actually an issue.  What's an issue is getting 
compensation for unapproved use.  This is where a registered copyright
comes in (you can collect punitive as well as actual damages and the
standard of proof of ownership is far, far lower where you've registered
the copyrighted entity - if you don't register, you can only collect
actual damages.  Since punitive damages can be 3x actual in the US,
you can see why lawyers are more interested in taking a case regarding
a registered copyright on contingency than one regarding an unregistered
copyright - plus the standard of proof of ownership is less).

Of course, it's hard to prove damages or even 3x punitive damages
when you GIVE YOUR STUFF AWAY FOR FREE! :)

So...you probably haven't given up any rights to sue someone for
"using my stuff without permission that I give away for free" but
your compensation for these extra copies that would be free anyway?

Not enough to cover your legal costs...

The above ignores licensing, the PG license makes it clear we don't
have to ask permission to use this stuff.  In essence, the PG 
license is stating that the copyright holders and otherwise code
owners are granting lots and lots of freedom to the recipient.

You can't legally win saying "I gave it away for free with no
restriction, but I still own the copyright and have changed my
mind, and now these folks have to pay!".

[you could for FUTURE distribution, but not past]

Despite all the above, things get cloudly when (say) the PG 
core gets together in Fiji for a big bash and coding party
and a Tsunami wipes everyone out at once and PG, Inc is left
without owners, only heirs of owners...

Worth worrying about?  I don't think so...as long as the latest
release was still out there with the current license, other
folks around the world could pick it up and any greedy PG, Inc
heirs would be left without a leg to stand on.

Not because of copyright, because of the license agreement under
which we are granted the use of the copyrighed material.

>This all started because I asked about copyright issues around Christmas
>(as someone reminded me), and we agreed to copyright the code via
>PostgreSQL Inc.  Now that we have done that, it seems we may have a
>little more to do.

Well...you can't just copyright someone's work without their permission.
When (say, Peter) makes a contribution you can't just say "OK, now it
is copyright PG, Inc" if Peter didn't agree to this arrangement.  Because
he has copyright in his code.  Of course, his code isn't worth much without
the rest of the 250,000 lines of PG but that's not the point.

I think the idea of looking at the Apache Foundation's sound, because
it is possible they've looked more deeply into this and have found
solutions.

I suspect that everyone here has the same general impression of what is
wanted, but getting the *bleeping* legal structure organized to recognize
it is difficult especially given the inheritance issue (with the original
UC license).  

Face it, the legal framework in the US isn't really set up to cater
to those who work hard and give their work freely to others...PG, Inc
etc needs to be creative.




- Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert
Serviceand other goodies at http://donb.photo.net.
 

At 00:52 29/01/00 -0500, Tom Lane wrote:
>bruc@stone.congenomics.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   |/

At 12:52 AM 1/29/00 -0500, Tom Lane wrote:

>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. 

I think this is the concern (not mine, raised by others) - what
guarantee is there that PG, Inc couldn't change the terms if 
(say) a new disease cropped up that killed all believers in Open
Source? :)  (yes intentionally silly).

In reality, only CHANGES could fall under more restrictive licensing,
but these changes might become intertwined in a way that made the
original distribution terms moot.  

This could be really simple to fix in legal terms.  With a not
for profit, distribution and development of a free and unencumbered
system could probably be incorporated into the bylaws and this may
be why the Apache Foundation was formed.  It's harder with companies,
which by definition are formed to be profitable and indeed in the
eyes of the IRS are supposed to strive for that goal (though there
is no need to succeed).

So...maybe "Inc" isn't the right long-term shape of the entity?

Personally, I don't think anyone should be terribly concerned about
this.  But I can see how some, especially folks who aren't US
citizens and perhaps don't know much about the realities of all this
in US law, could become concerned.



- Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert
Serviceand other goodies at http://donb.photo.net.
 

At 05:01 PM 1/29/00 +1100, Philip Warner wrote:

>It's probably a little more than just putting the notices in the code. In
>Australia, copyright rests with the writer unless it is explicitly granted
>in writing to another entity by the author, or unless a singed contract of
>service exists which grants copyright of works performed directly in the
>course of performing those service.

This could probably be argued here, too, though the weight of oral
and implied conditions weigh heavily here.  It would be hard for a
significant contributor, who by definition would spend a lot of 
time working with the code and being social with "the group", to claim
they were unaware that their code would be distributed.

And what would be the remedy?  The code contribution's worthless without
the database.  If there's a novel or patentable idea in the code snippet
you could still patent it and be nice and grant the PG distribution the
right to use it royalty free.  How does modern (screwed-up, IMO) software
patenting possibilities impact possible contributions, oh my, another
can of worms!

Anyway, anyone who contributes code to the repository is doing so with
the intent of its being distributed.  I don't think good lawyers would 
have much trouble with defending that part of the arrangement.

>If the US & Canada are similar, the upshot of this is that you'll need to
>get the contributors (and Berkely) to sign their copyrights over to
>PosgreSQL, Inc, before you can lawfully place those copyright notices in
>the code. I would *guess* that there would be some resistance to this.

So, what, you sign every "XXX" comment with a copyright?

I don't think so.  

Things are muddied by the fact that there's no legal relationship between
PG, Inc and the individuals (as opposed to your working and being paid
by a company, which sets up a relationship with implicit rules).

>It's a hell of a lot simpler to leave the copyright with the authors, who
>then grant usage under a GPL/LGPL or whatever. Or, set up an arms-length
>entity designed to promote free distribution of PostgreSQL that authorsd
>may feel happy about assigning their copyrights to.

Which again is probably like the Apache Foundation or whatever it is 
called.

>Again, just my 0.02c, but copyright is a very thorny issue designed to
>protect authors - it's a pity to take that away from them.

Do you have newspapers in Australia?  I seem to recall you do from my
one visit there :)

Do these newspapers have a "letters to the editor" column?  If so, do
the newspapers make each and every contributor sign over their rights
before publishing such letters?

I know in the US they don't.  

There's an explicit permission to publish when you mail in your letter
to the editor.  If you were to sue, the court would have a couple of
questions:

1. Did you send your letter in so it WOULDN'T be published?  When you  read the little blurb in the paper that said
"writeyour opinion  letters for consideration of publishment to xyz box" that they  would never publish it?
 

and importantly:

2. What was the financial harm done you by their publishing it.
  (non-registered copyrights are strictly a civil affair here)

> I am very happy
>that the copyright of the code I have contributed to, eg, the GD graphics
>library remains with me, but that it can be freely used & distributed in
>GD. I would not want to sign my rights to that code away, then be told by
>the person that buys GD from Tom Boutell that I am no longer permitted to
>use the code I wrote.

Yes, of course.

The legend, though, says "portions copyright 1996-2000 by PG, Inc."

It doesn't identify which portions.  Nor does the original statement
by UC.  The reality is that no one could necessarily claim that any
contribution you make is part of the "portions".



- Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert
Serviceand other goodies at http://donb.photo.net.
 

At 22:32 28/01/00 -0800, Don Baccus wrote:
>At 05:01 PM 1/29/00 +1100, Philip Warner wrote:
>
>>It's probably a little more than just putting the notices in the code. In
>>Australia, copyright rests with the writer unless it is explicitly granted
>>in writing to another entity by the author, or unless a singed contract of
>>service exists which grants copyright of works performed directly in the
>>course of performing those service.
>
>This could probably be argued here, too, though the weight of oral
>and implied conditions weigh heavily here.  It would be hard for a
>significant contributor, who by definition would spend a lot of 
>time working with the code and being social with "the group", to claim
>they were unaware that their code would be distributed.
>
>And what would be the remedy?  The code contribution's worthless without
>the database.  If there's a novel or patentable idea in the code snippet
>you could still patent it and be nice and grant the PG distribution the
>right to use it royalty free.  How does modern (screwed-up, IMO) software
>patenting possibilities impact possible contributions, oh my, another
>can of worms!
>
>Anyway, anyone who contributes code to the repository is doing so with
>the intent of its being distributed.  I don't think good lawyers would 
>have much trouble with defending that part of the arrangement.

As you say, there is an implied license to distribute. Similarly, if they
remove theire code, there is an implied revocation of the license. As you
said earlier, this is nothing to do with copyright. Copyright is the thing
that gives you the ability to write your own license.


>>If the US & Canada are similar, the upshot of this is that you'll need to
>>get the contributors (and Berkely) to sign their copyrights over to
>>PosgreSQL, Inc, before you can lawfully place those copyright notices in
>>the code. I would *guess* that there would be some resistance to this.
>
>So, what, you sign every "XXX" comment with a copyright?
>
>I don't think so.  

Well, somone told me that the GNU developers have to do just that - they
sign a piece of paper up front waiving copyright. I've only heard this from
one source, though.


>Things are muddied by the fact that there's no legal relationship between
>PG, Inc and the individuals (as opposed to your working and being paid
>by a company, which sets up a relationship with implicit rules).
>
>>It's a hell of a lot simpler to leave the copyright with the authors, who
>>then grant usage under a GPL/LGPL or whatever. Or, set up an arms-length
>>entity designed to promote free distribution of PostgreSQL that authorsd
>>may feel happy about assigning their copyrights to.
>
>Which again is probably like the Apache Foundation or whatever it is 
>called.

Yes, I assume so. I hope they get back to me...


>>Again, just my 0.02c, but copyright is a very thorny issue designed to
>>protect authors - it's a pity to take that away from them.
>
>Do you have newspapers in Australia?  I seem to recall you do from my
>one visit there :)
>
>Do these newspapers have a "letters to the editor" column?  If so, do
>the newspapers make each and every contributor sign over their rights
>before publishing such letters?
>
>I know in the US they don't.  
>
>There's an explicit permission to publish when you mail in your letter
>to the editor.  If you were to sue, the court would have a couple of
>questions:
>

Again, there is an implied license for a one-off usage, but no loss of
copyright. They could not publish my collected letters in a book without
obtaining further permission.


>> I am very happy
>>that the copyright of the code I have contributed to, eg, the GD graphics
>>library remains with me, but that it can be freely used & distributed in
>>GD. I would not want to sign my rights to that code away, then be told by
>>the person that buys GD from Tom Boutell that I am no longer permitted to
>>use the code I wrote.
>
>Yes, of course.
>
>The legend, though, says "portions copyright 1996-2000 by PG, Inc."
>
>It doesn't identify which portions. Nor does the original statement
>by UC.  The reality is that no one could necessarily claim that any
>contribution you make is part of the "portions".

As you said earlier, we all agree that they are doing this in the best
interests on PostgreSQL. The question is, what are they trying to achieve,
and can we achieve a better result another way?

As the law stands here, the only portions the company would own would be
the portions written by their employees while working at the company's
direction (assuming of course their work agreement assigns copyright to
their employer).

I'm not trying to offend anyone - as you say, I think we all want the same
result.


----------------------------------------------------------------
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   |/

> This depends on the definition of open source; I think Netscape is
> OpenSource these days, but I doubt if anyone would be allowed to take the
> sources and create "Petscape: The Dog's Browser". But, as far as I know, I
> could go away tomorrow with the PostgreSQL sources and create "Mostgress",
> so long as I did not try to sell it. AFAIK, I might even get away with
> selling it, but it's not really the point.

Yes you could. The original Postgres developers (or at least some of
them) did just that in founding Illustra. They sold the company a
couple of years later for $50M US to Informix.

> What matters is that it remains open, free, and modifyable by anyone, for
> any purpose.

The original sources, and the modified sources *that we know about*
(someone could have and probably has taken the source code, modified
it, and not contributed back the changes) are always fair game to be
taken out of open source status. That is an artifact or benefit or
downside of the BSD license, depending on your PoV.

> This does bring up the larger question of what happened to the original
> copyrights? Is this covered in Bruce's upcoming book?

The original copyrights are still valid and travel with the code.
However, afaict they are designed to release UC Berkeley from
liability and to preserve some credit for the original work, not to
allow Berkeley to assert ownership control over derivative sources
(into which category I think the current PostgreSQL tree falls, so to
speak).

I think Marc is concerned that there be someone or something able to
represent the current code tree, and to prevent hijacking of the
PostgreSQL (and perhaps Postgres) names from this open source group.
He has been consistantly adamant about preserving the BSD copyright,
which leave the maximum flexibility for the use of the code tree
basically for any purpose by anyone. That isn't quite the same goal of
GPL, but check the archives for threads which cover these topics
exhaustively.
                   - Thomas

-- 
Thomas Lockhart                lockhart@alumni.caltech.edu
South Pasadena, California

Don Baccus <dhogaza@pacifier.com> writes:
> I think this is the concern (not mine, raised by others) - what
> guarantee is there that PG, Inc couldn't change the terms if 
> (say) a new disease cropped up that killed all believers in Open
> Source? :)  (yes intentionally silly).

Well, the critical point here is that neither PG, Inc nor anyone
else can *retroactively* change the terms of distribution.  The
copies that are out there are out there, with the terms of
distribution stated right in them.  Anyone can pick one up and
start doing their own thing with the code.  Neither UC Berkeley nor
any PostgreSQL contributor would have a leg to stand on to stop them;
you think the courts will look kindly on "oh, we didn't mean what we
said in the terms of distribution"?

The worst-case possible scenario is that Marc goes around the bend
and, five minutes before the release of version 7.42, announces that
7.42 will be distributed under new terms that everyone else thinks are
too tight.  Everyone else just flips him the bird, goes back to 7.41
and continues on with life.  (Furthermore, if anyone felt like suing,
such a last-minute switcheroo would never hold up in court.  Anyone
who had contributed code to 7.42 under the reasonable expectation that
it would be licensed just like 7.41 would have plenty of grounds to
say "wait a minute, where do you think you're going with my code?")

Same scenario applies if Marc gets run over by a truck and the new
owners of PG Inc try to do something unreasonable.  The fact is that
as long as the terms of distribution stay the same, PG Inc doesn't
have any real ability to hurt anyone.

If anyone wants to look into Apache and see how it's set up, I've got no
problem with taking a look.  I think it's real easy to make a mountain
out of a molehill in this area, however.  Look at FSF --- they actually
require signed paperwork (hard copy, not email) from any potential
contributor before they will accept code contributions.  Do we want to
get that anal-retentive?  I hope not.
        regards, tom lane

Philip Warner <pjw@rhyme.com.au> writes:
> When you refer to the license terms, do you mean the 'COPYRIGHT' file in
> the root directory, or is there something more?

Yeah.  I have seen more extensive statements of the standard UC Berkeley
license terms (didn't BSD come with a different statement?) but that's
the one we are working with.

> 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.

Not retroactively.  Someone could try to do *new* development under
more restrictive terms, or just plain do some traditional proprietary
development using Postgres as a base.  They've got just as much right
to use the code in that way as you or I do to use it as we want to.
But I doubt many of the current contributors would help.

> The big advantage of a separate entity is that it has a clear charter
> and no conflict of interest.

Only to the extent that the people sitting on its board of directors
have no conflict of interest --- otherwise they could easily vote to
make the separate entity do something unhappy-making.  AFAICS these
would be more or less the same people who control PostgreSQL Inc;
namely the core developers.  If you don't trust them when they're
wearing one hat, why trust them when wearing another?
        regards, tom lane

> What is quite confusing to me is that though we have copyrighted the
> code now, we don't have a statement saying what that copyright allows
> people to do with the code.  And with no statement, we are basically
> doing an "all rights reserved" thing, I think, which is certainly not
> what is intended.

Here is what we have now:

PostgreSQL is Copyright © 1996-9 by the PostgreSQL Global Development
Group, and is distributed under the terms of the Berkeley license. 

Postgres95 is Copyright © 1994-5 by the Regents of the University of
California. Permission to use, copy, modify, and distribute this
software and its documentation for any purpose,
without fee, and without a written agreement is hereby granted,
provided that the above copyright notice and this paragraph and the
following two paragraphs appear in all copies. 

In no event shall the University of California be liable to any party
for direct, indirect, special, incidental, or consequential damages,
including lost profits, arising out of the use of this
software and its documentation, even if the University of California
has been advised of the possibility of such damage. 

The University of California specifically disclaims any warranties,
including, but not limited to, the implied warranties of
merchantability and fitness for a particular purpose. The software
provided hereunder is on an "as-is" basis, and the University of
California has no obligations to provide maintainance, support,
updates, enhancements, or modifications. 


What we may need to do is be more explicit; perhaps we should restate
the text of the Berkeley copyright but substitute another group as the
current copyright holder, asking that credit be given to the group
when using or distributing the software.

afaict that is a perfectly reasonable request to make of users of the
code, and it gives all of us open-source contributors a way to see
where our work is being used in the future. That's part of what makes
this fun...
                   - Thomas

-- 
Thomas Lockhart                lockhart@alumni.caltech.edu
South Pasadena, California

Thomas Lockhart <lockhart@alumni.caltech.edu> writes:
> The original sources, and the modified sources *that we know about*
> (someone could have and probably has taken the source code, modified
> it, and not contributed back the changes) are always fair game to be
> taken out of open source status.

The above could easily be misinterpreted.  I believe the point Thomas
meant to make is that anyone is free to make a derivative version that
they choose not to release as open source.  It does *not* mean that
someone can take away your right to use existing code that was already
released with a Berkeley-style license.

> The original copyrights are still valid and travel with the code.
> However, afaict they are designed to release UC Berkeley from
> liability and to preserve some credit for the original work, not to
> allow Berkeley to assert ownership control over derivative sources
> (into which category I think the current PostgreSQL tree falls, so to
> speak).

Right, and I think that it's past time that the Postgres group (in
the person of PG Inc, or some other entity if that's what a majority
want) explicitly make the same statements that UC Berkeley has made.
Anyone here want to be on the hook for liability when some big
company's database crashes?  Not me...

> I think Marc is concerned that there be someone or something able to
> represent the current code tree, and to prevent hijacking of the
> PostgreSQL (and perhaps Postgres) names from this open source group.

That's actually quite a separate issue.  Trademarking the name
"PostgreSQL" might be a good idea to prevent some random bozo from
claiming ownership of it.  (In reality, I think any attempt by
someone else to register that name as a database trademark at this
point could easily be shot down, but it would be far cheaper to
register the mark pre-emptively than to file suit against someone's
predatory registration.)  Yet ... on the other hand, if it were a
trademark then someone *could* buy it off PG Inc.  If that happened,
we (the community) would still have the rights to use the Postgres code,
but we'd have to find another name for it ;-).
        regards, tom lane

On Fri, Jan 28, 2000 at 05:07:43PM -0600, Ross J. Reedstrom wrote:
> Ah, corporate assets? What if someone bought out PostgreSQL, Inc., for
> a huge sum of money. They'd get control of anything PostgreSQL, Inc.
> has control over. One can't buy out a non-profit organization.

Which brings up the question whether PostgreSQL Inc. is a profit or a
non-profit organizationn.

AFARC that's the way Debian solved this problem, by forming a non-profit
organization name Software in the Public Interest (SPI) Inc.

Michael-- 
Michael Meskes                         | Go SF 49ers!
Th.-Heuss-Str. 61, D-41812 Erkelenz    | Go Rhein Fire!
Tel.: (+49) 2431/72651                 | Use Debian GNU/Linux!
Email: Michael@Fam-Meskes.De           | Use PostgreSQL!

On Sat, Jan 29, 2000 at 04:31:55PM +1100, Philip Warner wrote:
> This depends on the definition of open source; I think Netscape is
> OpenSource these days, but I doubt if anyone would be allowed to take the

The new version 5 code yes. The old version 4 is just given away for free.

> sources and create "Petscape: The Dog's Browser". But, as far as I know, I
> could go away tomorrow with the PostgreSQL sources and create "Mostgress",
> so long as I did not try to sell it. AFAIK, I might even get away with
> selling it, but it's not really the point.

Yes, you can sell Mostgress. That's one of the problems with a BSD type
license. That's why may use GPL for there open source projects.

> What matters is that it remains open, free, and modifyable by anyone, for
> any purpose.

The source as it is now will remain open, free and modifyable. If someone
forsk its own version that's a different matter. But ours remains open.

Michael

-- 
Michael Meskes                         | Go SF 49ers!
Th.-Heuss-Str. 61, D-41812 Erkelenz    | Go Rhein Fire!
Tel.: (+49) 2431/72651                 | Use Debian GNU/Linux!
Email: Michael@Fam-Meskes.De           | Use PostgreSQL!

Michael Meskes wrote:

> On Sat, Jan 29, 2000 at 04:31:55PM +1100, Philip Warner wrote:
> > This depends on the definition of open source; I think Netscape is
> > OpenSource these days, but I doubt if anyone would be allowed to take the
>
> The new version 5 code yes. The old version 4 is just given away for free.
>
> > sources and create "Petscape: The Dog's Browser". But, as far as I know, I
> > could go away tomorrow with the PostgreSQL sources and create "Mostgress",
> > so long as I did not try to sell it. AFAIK, I might even get away with
> > selling it, but it's not really the point.
>
> Yes, you can sell Mostgress. That's one of the problems with a BSD type
> license. That's why may use GPL for there open source projects.

Some of us view this the other way around....using anything GPL can "infect"
any product which may be putting the food on the programmer's table....What do
we care if someone else out there is making money off of, say,
"Mostgress"...good for them. Letting people do whatever the devil they want, it
seems to me, is the as free as free can get. :-)

Mike Mascari

At 02:10 AM 1/29/00 -0500, Tom Lane wrote:

>If anyone wants to look into Apache and see how it's set up, I've got no
>problem with taking a look.  I think it's real easy to make a mountain
>out of a molehill in this area, however.  Look at FSF --- they actually
>require signed paperwork (hard copy, not email) from any potential
>contributor before they will accept code contributions.  Do we want to
>get that anal-retentive?  I hope not.

Yes, in spirit I'm certainly with you on this one.  I presume some
laywer started hacking code in his spare time then started hacking
the FSF as well :)



- Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert
Serviceand other goodies at http://donb.photo.net.
 

At 05:50 PM 1/29/00 +1100, Philip Warner wrote:
>>There's an explicit permission to publish when you mail in your letter
>>to the editor.  If you were to sue, the court would have a couple of
>>questions:

>Again, there is an implied license for a one-off usage, but no loss of
>copyright. They could not publish my collected letters in a book without
>obtaining further permission.

It's debatable, at least here in the US.  You could look into recent
fights among photographers and publishers over web redistribution of
articles originally published with ink on paper, for instance.

The issue isn't loss of copyright, but license scope...



- Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert
Serviceand other goodies at http://donb.photo.net.
 

On 2000-01-28, Bruce Momjian mentioned:

> [Charset ISO-8859-1 unsupported, filtering to ASCII...]
> > This came down with my latest update and confused me quite a bit:
> > 
> > * Portions Copyright (c) 1996-2000, PostgreSQL, Inc
> > 
> > Since when do *they* own the copyright? Whatever happened to the good old
> > PostgreSQL Global Development Group? Assigning the copyright of an open
> > source project to a commercial entity is something unprecedented, and
> > quite frankly (without having anything at all against PostgreSQL, Inc) I
> > am not so excited about it.
> 
> That is what I was told by Marc.  If we want to discuss it, go ahead. 
> Such a change was planned in late December, though there was no
> discussion who gets the copyright.  I believe he used PostgreSQL Inc.
> because it is a legal entity, vs. the development team, which is not.

In any case, a better way to search for copyright notices is

egrep -i '\(c\)|copyright'

Just

egrep '\(C\)'

will miss some. ;)

-- 
Peter Eisentraut                  Sernanders väg 10:115
peter_e@gmx.net                   75262 Uppsala
http://yi.org/peter-e/            Sweden

On Fri, 28 Jan 2000, Don Baccus wrote:

> At 12:52 AM 1/29/00 -0500, Tom Lane wrote:
> 
> >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. 
> 
> I think this is the concern (not mine, raised by others) - what
> guarantee is there that PG, Inc couldn't change the terms if 
> (say) a new disease cropped up that killed all believers in Open
> Source? :)  (yes intentionally silly).

Simple answer: the developers all go off with source code at the time that
PostgreSQL, Inc changes those terms and continues off where they left off
... IMHO, that is the nice thing about Open Source ... look at the
FreeBSD/OpenBSD/NetBSD camps ... developers A became unhappy with politics
in camp A so formed camp B and went on their merry way ... all three are
equally successful...

The same thing applies to Linux ... how many different distributions and
philosophies are out there now?  

You are right, PostgreSQL, Inc could go out, change the license on the
code "from this day forth", but that doesn't stop developers to branch off
a new code base based upon the license of "the day before" and totally
leaving PostgreSQL, Inc in the dust ...

> This could be really simple to fix in legal terms.  With a not
> for profit, distribution and development of a free and unencumbered
> system could probably be incorporated into the bylaws and this may
> be why the Apache Foundation was formed.  It's harder with companies,
> which by definition are formed to be profitable and indeed in the
> eyes of the IRS are supposed to strive for that goal (though there
> is no need to succeed).

> So...maybe "Inc" isn't the right long-term shape of the entity?
> 
> Personally, I don't think anyone should be terribly concerned about
> this.  But I can see how some, especially folks who aren't US
> citizens and perhaps don't know much about the realities of all this
> in US law, could become concerned.

Ummm...PostgreSQL, Inc isn't a US company...*technically*, like OpenBSD,
PostgreSQL is a Canadian Open Source Project, as its development is based
in Canada ... which has this neat little advantage that OpenBSD has taken
advantage of, but we haven't yet: we can add stuff like SSL encryption
directly into the source code and distribute it legally...

Marc G. Fournier                   ICQ#7615664               IRC Nick: Scrappy
Systems Administrator @ hub.org 
primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org 

On Fri, 28 Jan 2000, Don Baccus wrote:

> It doesn't identify which portions.  Nor does the original statement
> by UC.  The reality is that no one could necessarily claim that any
> contribution you make is part of the "portions".

And, as the wu-ftpd distribution has always done, there is *nothing* to
stop someone working in 'heapam.c' from adding a personal 'Portions
Copyright(c) 1999 by Bruce Momjian' ... we put a 'blanket copyright on' to
provide a 'general cover', but there is nothing from stopping ppl from
adding more specific ones for the work they do:

===========================================
/****************************************************************************    Copyright (c) 1999 WU-FTPD Development
Group.  All rights reserved.  Portions Copyright (c) 1980, 1985, 1988, 1989, 1990, 1991, 1993, 1994    The Regents of
theUniversity of California. Portions Copyright (c) 1993, 1994 Washington University in Saint Louis.  Portions
Copyright(c) 1996, 1998 Berkeley Software Design, Inc.  Portions Copyright (c) 1989 Massachusetts Institute of
Technology. Portions Copyright (c) 1998 Sendmail, Inc.  Portions Copyright (c) 1983, 1995, 1996, 1997 Eric P.  Allman.
PortionsCopyright (c) 1997 by Stan Barber.  Portions Copyright (c) 1997 by Kent Landfield.  Portions Copyright (c)
1991,1992, 1993, 1994, 1995, 1996, 1997    Free Software Foundation, Inc.   
 
===========================================

Marc G. Fournier                   ICQ#7615664               IRC Nick: Scrappy
Systems Administrator @ hub.org 
primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org 

On Sat, 29 Jan 2000, Tom Lane wrote:

> The worst-case possible scenario is that Marc goes around the bend
> and, five minutes before the release of version 7.42, announces that
> 7.42 will be distributed under new terms that everyone else thinks are
> too tight.  Everyone else just flips him the bird, goes back to 7.41
> and continues on with life.  (Furthermore, if anyone felt like suing,
> such a last-minute switcheroo would never hold up in court.  Anyone
> who had contributed code to 7.42 under the reasonable expectation that
> it would be licensed just like 7.41 would have plenty of grounds to
> say "wait a minute, where do you think you're going with my code?")

Actually, given that scenario, it wouldn't even a matter of going back to
7.41 ... everything in CVS, so you'd go back to the source tree as of the
date before the license changed ...

IMHO, the worst thing *anyone* organization could do is change the license
in such a way as to alienate the only thing of value ... the developers
*shrug*

At 12:51 PM 1/29/00 -0400, The Hermit Hacker wrote:

>The same thing applies to Linux ... how many different distributions and
>philosophies are out there now?  

Either too many, or not enough, I can never decide :)

>Ummm...PostgreSQL, Inc isn't a US company...*technically*, like OpenBSD,
>PostgreSQL is a Canadian Open Source Project, as its development is based
>in Canada

I hadn't realized that (not that I've cared!).

> ... which has this neat little advantage that OpenBSD has taken
>advantage of, but we haven't yet: we can add stuff like SSL encryption
>directly into the source code and distribute it legally...

Sigh...it's always painful to have my country's stupidity flaunted
in public :)




- Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert
Serviceand other goodies at http://donb.photo.net.
 

On Sat, Jan 29, 2000 at 12:07:45AM -0500, Mike Mascari wrote:
> > Yes, you can sell Mostgress. That's one of the problems with a BSD type
> > license. That's why may use GPL for there open source projects.                       ^^^

Should have been 'many'. :-)

> Some of us view this the other way around....using anything GPL can "infect"
> any product which may be putting the food on the programmer's table....What do
> we care if someone else out there is making money off of, say,
> "Mostgress"...good for them. Letting people do whatever the devil they want, it
> seems to me, is the as free as free can get. :-)

I can understand and tolerate both reasonings. That's why I do some of my
open source work under BSD (ecpg) and other under GPL (watchdog).

Michael
-- 
Michael Meskes                         | Go SF 49ers!
Th.-Heuss-Str. 61, D-41812 Erkelenz    | Go Rhein Fire!
Tel.: (+49) 2431/72651                 | Use Debian GNU/Linux!
Email: Michael@Fam-Meskes.De           | Use PostgreSQL!

Thomas Lockhart wrote:
> 
> > This depends on the definition of open source; I think Netscape is
> > OpenSource these days, but I doubt if anyone would be allowed to take the
> > sources and create "Petscape: The Dog's Browser". But, as far as I know, I
> > could go away tomorrow with the PostgreSQL sources and create "Mostgress",
> > so long as I did not try to sell it. AFAIK, I might even get away with
> > selling it, but it's not really the point.
> 
> Yes you could. The original Postgres developers (or at least some of
> them) did just that in founding Illustra. They sold the company a
> couple of years later for $50M US to Informix.

Actually Postgres was _not_ distributed under the BSD (do-whatever-you-want
but 
give credit) license but a much more restrictive license that required a
special 
permission from UBC to to anything non-educational.

IIRC Stonebraker et al founded Illustra after getting special permissions from
UCB

The shift to BSD license (that enabled the current blooming of PostgreSQL ;)
was oftained sometime during the development of postgres 95, with some
considerable backing from Stonebraker.

> 
-------------
Hannu

On Sat, 29 Jan 2000, Don Baccus wrote:

> At 12:51 PM 1/29/00 -0400, The Hermit Hacker wrote:
> 
> >The same thing applies to Linux ... how many different distributions and
> >philosophies are out there now?  
> 
> Either too many, or not enough, I can never decide :)
> 
> >Ummm...PostgreSQL, Inc isn't a US company...*technically*, like OpenBSD,
> >PostgreSQL is a Canadian Open Source Project, as its development is based
> >in Canada
> 
> I hadn't realized that (not that I've cared!).
> 
> > ... which has this neat little advantage that OpenBSD has taken
> >advantage of, but we haven't yet: we can add stuff like SSL encryption
> >directly into the source code and distribute it legally...
> 
> Sigh...it's always painful to have my country's stupidity flaunted
> in public :)

When those laws were created, there was a reason for them ... in this day
and age, there isn't, but, unfortunately, I think its easier to put *in* a
law then it is to repeal it, no? :(  Give it time ... just thank god for
loopholes :)

Marc G. Fournier                   ICQ#7615664               IRC Nick: Scrappy
Systems Administrator @ hub.org 
primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org 

> Thomas Lockhart wrote:
> > 
> > > This depends on the definition of open source; I think Netscape is
> > > OpenSource these days, but I doubt if anyone would be allowed to take the
> > > sources and create "Petscape: The Dog's Browser". But, as far as I know, I
> > > could go away tomorrow with the PostgreSQL sources and create "Mostgress",
> > > so long as I did not try to sell it. AFAIK, I might even get away with
> > > selling it, but it's not really the point.
> > 
> > Yes you could. The original Postgres developers (or at least some of
> > them) did just that in founding Illustra. They sold the company a
> > couple of years later for $50M US to Informix.
> 
> Actually Postgres was _not_ distributed under the BSD (do-whatever-you-want
> but 
> give credit) license but a much more restrictive license that required a
> special 
> permission from UBC to to anything non-educational.
> 
> IIRC Stonebraker et al founded Illustra after getting special permissions from
> UCB
> 
> The shift to BSD license (that enabled the current blooming of PostgreSQL ;)
> was oftained sometime during the development of postgres 95, with some
> considerable backing from Stonebraker.
> 

Well, that's interesting.

--  Bruce Momjian                        |  http://www.op.net/~candle pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026
 

On Sat, Jan 29, 2000 at 07:07:34AM -0800, Don Baccus wrote:
> At 02:10 AM 1/29/00 -0500, Tom Lane wrote:
> 
> >If anyone wants to look into Apache and see how it's set up, I've got no
> >problem with taking a look.  I think it's real easy to make a mountain
> >out of a molehill in this area, however.  Look at FSF --- they actually
> >require signed paperwork (hard copy, not email) from any potential
> >contributor before they will accept code contributions.  Do we want to
> >get that anal-retentive?  I hope not.
> 
> Yes, in spirit I'm certainly with you on this one.  I presume some
> laywer started hacking code in his spare time then started hacking
> the FSF as well :)

Nah, I think it's more like how the rules are always more strict for
first born children: parents don't want to make a mistake, and don't
know what might _be_ a mistake, so set early curfews, etc. Once some more
kids come along (read Free Software projects) they start to loosen up...

Ross
-- 
Ross J. Reedstrom, Ph.D., <reedstrm@rice.edu> 
NSBRI Research Scientist/Programmer
Computer and Information Technology Institute
Rice University, 6100 S. Main St.,  Houston, TX 77005

> > Right.  Although IANAL, I'm pretty sure it's pointless to slap a
> > copyright notice on something unless the copyright names an actual> 
> > legal entity (one which could go sue an infringer, if necessary).
> > The development team is not a person, corporation, or partnership,
> > so in the eyes of the law it doesn't exist.

Since the licence lets anybody do pretty much anything, who are you 
going to sue?? It seems like the only reason to have the copyright
is to establish that the code is not stolen.