Re: journaling in contrib ... - Mailing list pgsql-hackers

From Marc G. Fournier
Subject Re: journaling in contrib ...
Date
Msg-id 20020814141053.D35100-100000@mail1.hub.org
Whole thread Raw
In response to Re: journaling in contrib ...  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: journaling in contrib ...  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-hackers
On Wed, 14 Aug 2002, Tom Lane wrote:

> Gavin Sherry <swm@linuxworld.com.au> writes:
> > On Wed, 14 Aug 2002, Bruce Momjian wrote:
> >> I think this belongs on gborg.  Would you create a project there?
>
> > A number of people at OSCON did consider this to be a nice contrib
> > feature. Out of curiousity, what makes it more suitable for gborg?
>
> I think Bruce is unhappy with the size of the tarball (220K!), and
> I was too when I first saw it.  But it turns out that nearly all of
> it is a copy of Hans' slides from his OSCON talk, which I don't think
> is appropriate to include in contrib anyway.  (For one thing, not
> everyone can read .sxi format.  I can't at the moment.)
>
> I'd suggest dropping the talk slides (and you might as well flatten the
> thing into one directory).  Perhaps instead the README could include a
> pointer to where to find the talk slides on-line.  That'd bring it down
> to half a dozen K which is a more appropriate size for a contrib item
> (and hopefully will not trigger Marc's wrath ;-)).

S'alright, I figured I'd read all the posts on the subject before I
responded in agreement with Bruce ...

*If* we start referring ppl and projects to GBorg, more ppl will go see
what is available ... how many ppl *look* in contrib to see what is
avalable in there?  I know that unless I'm searching for something
specific, I never check out what is in contrib, but more oft search places
like freshmeat to see if someone has done it already ...

The original reason for contrib was *if* someone submit'd a patch to
extend the server, but we weren't quite sure whether it should go in, we'd
give it one release in contrib to see if it would be useful before either
trashing it, or incorporating it ... the reality of the situation: we've
never trashed anything, and rarely incorporated ...

Anything in contrib that can be built seperately from the server code,
that just requires libpq and headers, should be pulled and distributed as
seperate modules, which has the added benefit that, if listed on GBorg,
search engines will pick up the modules ...

And the whole arg that someone threw out about 'nobody maintaining them if
they aren't part of the distribution' ... so?  if nobody is maintaining,
then who is using??



pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: Open 7.3 items
Next
From: Bruce Momjian
Date:
Subject: Re: Open 7.3 items