The definition of PGDG

IMHO it would be appropriate to provide better links to Slony from
within the Postgres docs.

The main reason is that Slony is Copyrighted PGDG, so we own the code
and it is of course BSD licenced.

Now that this has been highlighted to me, I can't see a reason for the
previous balanced approach.

--
Simon Riggs
2ndQuadrant http://www.2ndQuadrant.com

Am Donnerstag, 8. November 2007 schrieb Simon Riggs:
> The main reason is that Slony is Copyrighted PGDG, so we own the code
> and it is of course BSD licenced.

Why is that a reason for mentioning it more prominently? Is "code ownership"
a relevant property?

--
Peter Eisentraut
http://developer.postgresql.org/~petere/

On Thu, 2007-11-08 at 10:10 +0100, Peter Eisentraut wrote:
> Am Donnerstag, 8. November 2007 schrieb Simon Riggs:
> > The main reason is that Slony is Copyrighted PGDG, so we own the code
> > and it is of course BSD licenced.
>
> Why is that a reason for mentioning it more prominently?

It's not, I'm assuming you'd actually like to see it more prominent.

My understanding was that we were trying to show equal favour to all of
the various solutions. This was a reason not to do that.

--
Simon Riggs
2ndQuadrant http://www.2ndQuadrant.com

Simon Riggs <simon(at)2ndquadrant(dot)com> writes:
> My understanding was that we were trying to show equal favour to all of
> the various solutions. This was a reason not to do that.

The reason for taking a "balanced approach" is that no one solution
fits everyone's needs. I don't think the core docs should be pushing
Slony more than other solutions.

regards, tom lane

Tom Lane wrote:
> Simon Riggs <simon(at)2ndquadrant(dot)com> writes:
> > My understanding was that we were trying to show equal favour to all of
> > the various solutions. This was a reason not to do that.
>
> The reason for taking a "balanced approach" is that no one solution
> fits everyone's needs. I don't think the core docs should be pushing
> Slony more than other solutions.

We do mention Slony for in-place upgrades because if its capabilities to
work across Postgres versions.

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://postgres.enterprisedb.com

+ If your life is a hard drive, Christ can be your backup. +

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160

Simon Riggs wrote:
> The main reason is that Slony is Copyrighted PGDG, so we
> own the code and it is of course BSD licenced.

As an aside, how can copyright be assigned to a non-defined
group (a concept really, as near as I can tell). Is the PGDG
actually defined anywhere yet? If not, anyone want to take
a stab at it?

IMHO, we need to get this resolved at some point - either have
the code owned by their respective contributors (e.g. Linux)
or by a legal entity (e.g. Apache Foundation). The former may
be what we actually have anyway.

Copying to advocacy as someone there may have the answer.

- --
Greg Sabino Mullane greg(at)turnstep(dot)com
PGP Key: 0x14964AC8 200711081016
http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8

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

iD8DBQFHMyqrvJuQZxSWSsgRAw0hAJ9DD2gwr4nlmeoPNPeifXTloWip6ACgwv9z
WQTV1ccmRQ0EBbomxQUxeak=
=zng7
-----END PGP SIGNATURE-----

On Thursday 08 November 2007 10:38, Greg Sabino Mullane wrote:
> Simon Riggs wrote:
> > The main reason is that Slony is Copyrighted PGDG, so we
> > own the code and it is of course BSD licenced.
>
> As an aside, how can copyright be assigned to a non-defined
> group (a concept really, as near as I can tell). Is the PGDG
> actually defined anywhere yet? If not, anyone want to take
> a stab at it?
>
> IMHO, we need to get this resolved at some point - either have
> the code owned by their respective contributors (e.g. Linux)
> or by a legal entity (e.g. Apache Foundation). The former may
> be what we actually have anyway.
>
> Copying to advocacy as someone there may have the answer.

AFAICT we have the former (code is owned by respective owners). AIUI, in most
European countries copyright is considered naturally given rights that you
have and that you cannot give away. In the U.S., you can give copy rights
away, however you can only do so to a defined legal entity, of which the PGDG
is not one. This can change somewhat depending on country and depending upon
employer agreements, but since no one is employed by the PGDG, it's mostly
moot from what I can tell.

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

On Thu, Nov 08, 2007 at 03:38:29PM -0000, Greg Sabino Mullane wrote:
> As an aside, how can copyright be assigned to a non-defined
> group (a concept really, as near as I can tell).

This is an interesting question, and one that the IETF faced a number of
years ago, winding up only recently. There are some possibly troubling bits
of news in the various IETF archives on this topic (in particular, pay
attention to the creation of the IETF trust).

That said, I have some reason to believe that the actual problem was not
that the copyright wasn't owned by a particular legal entity, but that there
were some individuals who were more or less threatening to prevent any new
work happening in order to satisfy their own agenda. The IETF decided to
compromise under the circumstances. (This is about all I know of the
topic.)

All of that said, the creation of the IETF Trust has resulted in some nasty,
corrosive discussions; significant legal costs; and a great deal of
distraction from the actual work of producing standards.

AFAICT, no harm was actually done over the years by the funny copyright
notices on IETF documents. So I suggest to leave well enough alone for the
time being. But I am not, to my chagrin, a lawyer; so if we think we need
legal advice on this topic, I suggest we make a request to FG, asking for
legal advice on the topic, "Do we need copyright assignment?" rather than
the topic, "To whom should copyright be assigned?"

> Is the PGDG actually defined anywhere yet? If not, anyone want to take a
> stab at it?

Unless the "stabber" in question is actually a lawyer with specialisation in
corporate holdings, I'd like to ask that prospective stabbers not do this.
Bad formulations that might be used in any future legal discussion are in
practice considerably worse than no formulation at all.

A
--
Andrew Sullivan
Old sigs will return after re-constitution of blue smoke

"Greg Sabino Mullane" <greg(at)turnstep(dot)com> writes:

> Simon Riggs wrote:
>> The main reason is that Slony is Copyrighted PGDG, so we
>> own the code and it is of course BSD licenced.
>
> As an aside, how can copyright be assigned to a non-defined
> group (a concept really, as near as I can tell). Is the PGDG
> actually defined anywhere yet? If not, anyone want to take
> a stab at it?

As a non-lawyer I don't think this is an issue. The group exists, it's just
that it's (merely) a group of individual people. Ie, the copyright is owned
jointly by every contributor and that's all the notice is saying. The PGDG
doesn't exist separately from the members as a legal person like a corporation
might.

It may be that the notice is insufficient for some legal purposes because it
doesn't explicitly name a legal person. However under the Berne convention
notices are actually irrelevant to claiming ownership anyways. The only impact
an insufficient notice might have is to make it harder to sue for damages (we
would still be able to sue to stop further infringement but not damages until
after the notice).

I suspect it's not actually true that it's insufficient notice but that's
something a lawyer should be able to answer easily. I also am entirely
skeptical that we care about being able to sue for damages.

If this were a GPL project it might matter more. But given that it's BSD we're
only interested in the copyright notices to protect ourselves from someone
else claiming they wrote it and we're infringing. Not to be able to pursue
someone else for infringing on our copyright.

--
Gregory Stark
EnterpriseDB http://www.enterprisedb.com
Ask me about EnterpriseDB's Slony Replication support!

On Nov 8, 2007, at 9:28 AM, Bruce Momjian wrote:
> Tom Lane wrote:
>> Simon Riggs <simon(at)2ndquadrant(dot)com> writes:
>>> My understanding was that we were trying to show equal favour to
>>> all of
>>> the various solutions. This was a reason not to do that.
>>
>> The reason for taking a "balanced approach" is that no one solution
>> fits everyone's needs. I don't think the core docs should be pushing
>> Slony more than other solutions.
>
> We do mention Slony for in-place upgrades because if its
> capabilities to
> work across Postgres versions.

I'm pretty sure Skytools/Londiste works across versions too.
Presumably, any replication that's not based on binary format should
work.
--
Decibel!, aka Jim C. Nasby, Database Architect decibel(at)decibel(dot)org
Give your computer some brain candy! www.distributed.net Team #1828