Re: Re: On the _need_ to vacuum...

* Paul M Foster <paulf(at)quillandmouse(dot)com> [010429 10:35] wrote:
> On Sat, Apr 28, 2001 at 10:22:53PM -0700, Alfred Perlstein wrote:
>
> > * geustace(at)godzone(dot)net(dot)nz <geustace(at)godzone(dot)net(dot)nz> [010428 21:44] wrote:
> > > I am rather staggered by a developer considering it necessary to
> > > attempt to cooerce the core development team into including a patch.
> >
> > I'm assuming you refer to the updated page at:
> > http://people.freebsd.org/~alfred/vacfix/
> >
>
> I have to agree with this. Alfred's free to do what he likes. I don't
> recall that he mentions whether this patch is Open Source.

It's amazing that someone with your writing skills is so lacking
in reading comprehension.

The patch is free for inclusion in Postgresql source code, the
patch is NOT free to end users unless laudered through the Postgresql
source tree.

> If it isn't,
> then this is all moot. It can't be included in PostgreSQL because of
> licensing issues. If it _is_ Open Source, then Alfred is free to charge
> for it. _However_, he makes the threat of potential legal action if you
> should broadly disseminate a previously downloaded copy of the patch.

That has been removed, I will not bother doing so and I'm probably
not within my rights to do so, there now you have it, an admittion
that I believe I'm not within my rights to own this code as it has
been previously released. I do with you'd read on before deciding
the mirror the damn thing every which place.

> That's not only not Open Source, it's ANTI- Open Source. On that basis
> alone, I would be averse to including it in PostgreSQL. The coercion
> issue is secondary and childish.

You're entitled to your opinions no matter how misformed and foolish
they are.

There is no coercion involved.

I also don't think it helps the project for me to be giving Joe
End User patches that can speed up vacuum that basically discourage
him from using the latest and greatest version, there's a few
problems in the 7.0.3 tree that are fixed in 7.1. 7.1 needs
mainstream use. And hey, who knows I may actually recoup my
investment that has me seemingly forever stuck with 7.0.3. If someone
really need the patches they will just have to pay for them.

By not having these patches available for 7.1, large sites such as
my previous employer can not even consider testing it. With gracious
help from the developers and a lot of work on our part we were able
to track down and fix several obscure bugs related to corruption
and crashes in 7.0.3, unfortunetly we can't deploy 7.1 so we don't
know what's lurking there.

Anyhow, I really shouldn't have bothered explaining this to someone
that thinks that open source == free beer. I guess it was really for
the rest of the people that think I'm being a jerk.

--
-Alfred Perlstein - [alfred(at)freebsd(dot)org]
Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 years of accumulated wisdom.