Re: [HACKERS] Copyright

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 |/