Thread: Misplaced modifier in Postgresql license

Something very important was recently raised on the misc@openbsd.org
list.  Due to the current environment that SCO is fostering in the
open source community, it would be prudent for the PostgreSQL team to
consider this issue.

The website claims that "PostgreSQL is distributed under the flexible
BSD license".  A glance at the license appears to confirm this,
however, there is a misplaced modifier in the first paragraph
following the copyright notices:

"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."

I have used CAPS to highlight the apparent error.  There are two ways
to interpret this statement.  One interpretation is that permission is
given and no fee will be charged for the granting of that permission.
The other is that permission is given so long as by using the software
no fee is charged to others.  The result of this ambiguity is that the
latest CD release of OpenBSD (3.4) no longer includes Postgresql.  It
is on the ftp sites, but the OpenBSD CDs are distributed for a fee
because they are a profit generator for the project.  The project has
encountered problems in the past regarding ambiguous licenses, and as
a result the need to protect the porject outweighs the convenience of
distributing packages with ambiguous licenses.

I believe that this is merely a bug in the wording of the license, and
that it doesn't reflect the intention of the project.  I hope that my
words will be considered carefully, and that appropriate steps will be
taken to resolve this problem.

Thank you.

Breen Ouellette
OpenBSD & PostgreSQL user

> "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."

My personal interpretation isn't very ambiguous at all. If the license
were interpreted to require that you could not charge to provide someone
with a copy of postgres, that would imply that you aren't allowed to
have a written agreement with them either. That just doesn't make sense
to me.

I can see how a lawyer might tell someone to play it safe though. Also,
I suppose in any disagreement over the ambiguity of a text, the side
perceiving ambiguities is bound to win ;)

    regards,
        jeff davis

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


> "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."

> The result of this ambiguity is that the latest CD release of OpenBSD
> (3.4) no longer includes Postgresql.

It certainly would have been nice if they would have consulted with us
before taking such an action, or at least let us known about it themselves.
Nonetheless, thank you for taking the time to mail this list. Licensing
can be a touchy area, but I'll try not to shoot the messenger.

> I have used CAPS to highlight the apparent error.  There are two ways
> to interpret this statement.  One interpretation is that permission is
> given and no fee will be charged for the granting of that permission.
> The other is that permission is given so long as by using the software
> no fee is charged to others.

There are a number of issues related to this issue. Six of them, actually:

1. Is the license unclear due to that clause?
2. Should the 'misplaced' clause be of concern?
3. Should changes be made to the licensing text?
4. Can changes be made to the licensing text?
5. If so, how do we do it?
6. Who has copyright to the PostgreSQL source code?

Short answers as I see them: Yes,No,No,Yes,Easily,Individual contributors.

1. Is the license unclear due to that clause?

Yes, it could be construed both ways. I would not go so far as to label it
as "unclear" or "ambiguous", as the actual meaning has been obvious to people
who have been reading it for many, many years. However, it is technically
correct that it could be read as "cannot use if you charge for it."

2. Should the 'misplaced' clause be of concern?

Although it could be construed as "you cannot charge", there are compelling
reasons why this is not the case. A quick recap:

 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.

First and foremost, the fact that there is a "provided" clause strongly
suggests that the conditions, exceptions, and qualifiers are stated /after/
this clause. This is normally how things work: you lay everything out, and
then add in the restrictions. Second, if we were to assume that the
"without fee" applies to the "purpose", then we must also assume that the
license is stating that it cannot be distributed *without* a written
agreement. This seems very absurd and further weakens the "without fee"
argument. Third, the fact that their exists a revised version of the
license from the original authors (University of California) which clears
up the misplaced modifier is an indication that the "common" interpretation
of what the original writers were trying to say is the correct interpretation.

One thing that that seems to be overlooked in all of this, however, is that
this license has a long history of use, and this is a very, very important
thing in the law. Many companies have sold software (for a fee) that included
PostgreSQL, and nobody has raised this issue until now. Everyone associated
with the project has always understood (and stated) that PostgreSQL can be
sold for a fee. The long history of this 'contract', and the fact that nobody
has ever objected to the terms of it, makes the license bulletproof in court.
You cannot have a license that is interpreted one way for many years (and
among many parties) and then come along and try to throw in a unique
interpretation of it.

3. Should changes be made to the licensing text?

Although there are different interpretations, I do not feel it that the
uncommon one is strong enough to warrant any concern, either by the
PostgreSQL community or by those who include PostgreSQL in products they sell
for a fee (e.g. the OpenBSD CD). There would be no harm in moving to a
clearer license, but neither would there be harm in keeping the current one.

4. Can changes be made to the licensing text?

That's a tough one (see #6) but the quick answer is "yes". If we were to
make a change that was entirely compatible with the existing license, I do
not foresee anybody having a problem with it. Probably the only people who
would have a standing to object would the original authors, and if we were
to change it to one of their newer wordings of the same license, there
should be no problems.

5. If so, how do we do it?

Cut and paste. :) Probably with a post to the general list asking if anyone
(especially past contributors) would have any objections, and then committing
the change to the COPYRIGHT file.

6. Who has copyright to the PostgreSQL source code?

An interesting question. Certainly the Regents of the University of California
owned the original copyright, but I suspect that the portion of the codebase
that has remained unchanged is very small indeed. Unlike some projects,
contributors are not asked or required to assign the copyright to a single
entity (e.g. The Apache Foundation), so one presumes that the copyright is
held by the original contributors. One could also argue that there is an
implicit assigning of the rights to the "PostgreSQL Global Development Group",
(PGDG) because no individual copyright notices exists anywhere (not entirely
true, but we'll skip over that), and thus the copyright really does belong
to the PGDG unless people state otherwise. (Of course, the PGDG is not a
real entity...)

I hope that the BSD people will reconsider their move and put PostgreSQL
back on the CD, without a license change. The existing one is more than
strong enough to survive any legal challenge based on the "misplaced modifier."

- --
Greg Sabino Mullane greg@turnstep.com
PGP Key: 0x14964AC8 200312011619

-----BEGIN PGP SIGNATURE-----

iD8DBQE/zAP7vJuQZxSWSsgRApLEAJ9Ddr3H7MYQkCaLXeIMMhrikH8QbQCglAyg
U/xxcWC0DQtC0Ao+BfAfCQ4=
=/Pxh
-----END PGP SIGNATURE-----

On Tue,  2 Dec 2003 03:22:34 -0000 greg@turnstep.com wrote:
> I hope that the BSD people will reconsider their move and put PostgreSQL
> back on the CD, without a license change.

this is unlikely. i think the current situation is effectively a standoff.

> The existing one is more than
> strong enough to survive any legal challenge based on the "misplaced modifier."

as i said in my own posting on this subject, Theo De Raadt consulted with
IP attorneys on this subject and was advised that the misplaced modifier
is a serious enough issue that he should pull PostgreSQL from the OpenBSD
distribution. given the reluctance to change the PostgreSQL license, i think
that the situation is what it is. i certainly have no plan to try to persuade
Theo to change his mind ("Bullwinkle, that trick never works!").

richard
--
Richard Welty                                         rwelty@averillpark.net
Averill Park Networking                                         518-573-7592
    Java, PHP, PostgreSQL, Unix, Linux, IP Network Engineering, Security

Hi all;

IANAL, but I feel that this discussion is missing some important things and
it might not hurt to get a legal opinion.  In fact, the purpose of this post
is to try to widen the legal questions in order to make sure that this is
meeting the community's (and OpenBSD's) needs.

I have a feeling that we are talking apples and oranges here.  There are two
issues here that may be confused:
1)  Does the misplaced modifier give the PostgreSQL contributors grounds to
sue anyone who sells the software for a fee?  The concensus here appears to
be "no" and that is what I am pulling from Greg's email.

2)  Does the license leave a redistributor open to the fear of a lawsuit,
however groundless?  I have a feeling that the OpenBSD standoff may be more
about this issue than the former one.

I think that the license should be changed.  Most lawyers I have ever
interacted with have given advice regarding how to play it safe.  So the
question I would ask an IP lawyer is:

If I were looking at redistributing this software (for a fee) under this
license, what advice would you give me?  Could I be at risk of being sued by
individual contributors?

Note this is a very different question than "could I win any legal
challenges on this basis?"  I suspect that Theo is asking the former while
we are asking the latter.

Also, I am a little confused by Tom's statement that we don't have the right
to modify the license.  If I can take the source code, package it up in a
binary-only release, forbid redistribution, charge per-seat licensing fees,
etc. is that not more drastic than changing to a more recent BSD-style
license, as long as the restrictions are the same?

Best Wishes,
Chris Travers

Chris Travers writes:

> Also, I am a little confused by Tom's statement that we don't have the right
> to modify the license.  If I can take the source code, package it up in a
> binary-only release, forbid redistribution, charge per-seat licensing fees,
> etc. is that not more drastic than changing to a more recent BSD-style
> license, as long as the restrictions are the same?

You can add your own conditions, but the original license must also appear
verbatim.  Thus, we cannot change it.

--
Peter Eisentraut   peter_e@gmx.net

"Chris Travers" <chris@travelamericas.com> writes:
> Also, I am a little confused by Tom's statement that we don't have the right
> to modify the license.

I don't see what's confusing about it.  Our implicit contract with
contributors (past and present) is that we'd distribute their work under
the terms of the same license they saw when they contributed.  Altering
the license without their agreement is breaking faith.

All of the arguments about license changes have been gone over in great
detail in the archives (I think the last major go-round on the topic was
in the summer of 2000).  No one who has been around long enough to
remember those flame wars is interested in re-opening the topic.  Not
even just to move a comma.

            regards, tom lane

Tom Lane wrote:
> "Chris Travers" <chris@travelamericas.com> writes:
> > Also, I am a little confused by Tom's statement that we don't have the right
> > to modify the license.
>
> I don't see what's confusing about it.  Our implicit contract with
> contributors (past and present) is that we'd distribute their work under
> the terms of the same license they saw when they contributed.  Altering
> the license without their agreement is breaking faith.
>
> All of the arguments about license changes have been gone over in great
> detail in the archives (I think the last major go-round on the topic was
> in the summer of 2000).  No one who has been around long enough to
> remember those flame wars is interested in re-opening the topic.  Not
> even just to move a comma.

What we could do is add a blurb on our web site or in the FAQ clarifying
this issue.

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

greg@turnstep.com wrote:

>I hope that the BSD people will reconsider their move and put PostgreSQL
>back on the CD, without a license change. The existing one is more than
>
>
>
Make that _Open_BSD people.

The NetBSD DESCR file puts it simply:  "PostgreSQL is free and the
complete source is available."

On Wed, 2003-12-03 at 17:06, Bruce Momjian wrote:
> Tom Lane wrote:
> > "Chris Travers" <chris@travelamericas.com> writes:
> > > Also, I am a little confused by Tom's statement that we don't have the right
> > > to modify the license.
> >
> > I don't see what's confusing about it.  Our implicit contract with
> > contributors (past and present) is that we'd distribute their work under
> > the terms of the same license they saw when they contributed.  Altering
> > the license without their agreement is breaking faith.
> >
> > All of the arguments about license changes have been gone over in great
> > detail in the archives (I think the last major go-round on the topic was
> > in the summer of 2000).  No one who has been around long enough to
> > remember those flame wars is interested in re-opening the topic.  Not
> > even just to move a comma.
>
> What we could do is add a blurb on our web site or in the FAQ clarifying
> this issue.
>

Oh, you mean like this paragraph

"The above is the BSD license, the classic open-source license. It has
no restrictions on how the source code may be used. We like it and have
no intention of changing it."

Which you added to the FAQ damn near two years ago !?!

http://developer.postgresql.org/cvsweb.cgi/pgsql-server/doc/FAQ.diff?r1=1.139&r2=1.140


Quite frankly I think the lawyer they spoke with was... well, this is a
family newsgroup so let's just say I don't feel his opinion is very
credible.  It would have been very easy for them to indemnify themselves
had they wanted to, to not even try to sort this out speaks of some
other agenda on their part IMHO.

Robert Treat
--
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL

Robert Treat wrote:
> > > All of the arguments about license changes have been gone over in great
> > > detail in the archives (I think the last major go-round on the topic was
> > > in the summer of 2000).  No one who has been around long enough to
> > > remember those flame wars is interested in re-opening the topic.  Not
> > > even just to move a comma.
> >
> > What we could do is add a blurb on our web site or in the FAQ clarifying
> > this issue.
> >
>
> Oh, you mean like this paragraph
>
> "The above is the BSD license, the classic open-source license. It has
> no restrictions on how the source code may be used. We like it and have
> no intention of changing it."
>
> Which you added to the FAQ damn near two years ago !?!
>
> http://developer.postgresql.org/cvsweb.cgi/pgsql-server/doc/FAQ.diff?r1=1.139&r2=1.140

Oh, that is in there. Good.  Seems we don't need anything else on our
end.  Someone can clearly point to that if they have concerns about our
wording being misinterpreted.

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073