Re: Fair large change to contributors

> >>> * I suggest we leave it at the old URL as well, and not remove it. For ppl
> >>> coming in from old links. It can still be community section, and it can
> >>> still have a different title, though
> >> This should be a redirect, which I can put in place with PHP. We should
> >> not duplicate content like that.
> >
> > It's not really duplicate. Just add two rows in the dispatcher.
>
> If it presents the same content under /developers as under /community it
> is duplicate and that is dumb. I would agree that having a redirect is
> appropriate.

Matter of definition. Data exists once and is presented twice. Kinda like normalisation :-P

> >>> * Let's not delete the old developer.php file and create a new identical
> >>> one. Modify the one that's there instead. Filenames don't have to match,
> >>> that's what the dispatcher takes care of.
> >> That would be inconsistent with the rest of the portal layout.
> >
> > that'sjust temporary. The system is set up so you can do this stuff without kiling version control, by using the dispatcher. Please do so.
>
> Exactly how are we killing version control? Subversion handles removal
> of files just fine.

Bad choice of words. But if your patch is applied, I can't see a diff of your changes. And it's still logged as one ad and one remove. so I can't easily diff between tomorrows version and one two steps back.
And since there really is no gain, imnsho, I see no reason to take away that ability.

> >>> * You added a new template, but didn't remove the old one. My suggestion -
> >>> keep the old one, don't add the new one.
> >> I should remove the old one. I must of missed it, sorry.
> >>
> > No, you should *use* the old one...
> >
>
> Disagreements are the heart of open source.

Indeed :)

/Magnus

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

On Wed, 5 Dec 2007 00:04:52 +0100
"Magnus Hagander" <magnus(at)hagander(dot)net> wrote:
>
> Bad choice of words. But if your patch is applied, I can't see a diff
> of your changes. And it's still logged as one ad and one remove. so I
> can't easily diff between tomorrows version and one two steps back.
> And since there really is no gain, imnsho, I see no reason to take
> away that ability.
>

O.k. I just checked and if I use svn mv instead of svn rm and svn add,
then the changelog will follow. That should remove your concern yes?

jd(at)scratch:~/test/test$ svn add test
A test
jd(at)scratch:~/test/test$ svn commit -m "this is a test"
Adding test
Transmitting file data .
Committed revision 1.
jd(at)scratch:~/test/test$ svn mv test test2
A test2
D test
jd(at)scratch:~/test/test$ svn commit -m "moving test2"
Deleting test
Adding test2

Committed revision 2.
jd(at)scratch:~/test/test$ svn log
- ------------------------------------------------------------------------
jd(at)scratch:~/test/test$ svn log test2
- ------------------------------------------------------------------------
r2 | jd | 2007-12-04 15:42:52 -0800 (Tue, 04 Dec 2007) | 1 line

moving test2
- ------------------------------------------------------------------------
r1 | jd | 2007-12-04 15:42:40 -0800 (Tue, 04 Dec 2007) | 1 line

this is a test
- ------------------------------------------------------------------------
jd(at)scratch:~/test/test$ svn log test
svn: 'test' is not under version control
jd(at)scratch:~/test/test$

Sincerely,

Joshua D. Drake
- --

=== The PostgreSQL Company: Command Prompt, Inc. ===
Sales/Support: +1.503.667.4564 24x7/Emergency: +1.800.492.2240
PostgreSQL solutions since 1997 http://www.commandprompt.com/
UNIQUE NOT NULL
Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate
PostgreSQL Replication: http://www.commandprompt.com/products/

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHVeZuATb/zqfZUUQRAhgrAJ4jiGd9VMUa1WML0Z8YgdTMNSWL3ACdFZWZ
PJttm+n48pLjkfncjGAiJ08=
=5zcv
-----END PGP SIGNATURE-----

Joshua D. Drake wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Wed, 5 Dec 2007 00:04:52 +0100
> "Magnus Hagander" <magnus(at)hagander(dot)net> wrote:
> >
> > Bad choice of words. But if your patch is applied, I can't see a diff
> > of your changes. And it's still logged as one ad and one remove. so I
> > can't easily diff between tomorrows version and one two steps back.
> > And since there really is no gain, imnsho, I see no reason to take
> > away that ability.
>
> O.k. I just checked and if I use svn mv instead of svn rm and svn add,
> then the changelog will follow. That should remove your concern yes?

Magnus' concern is svn diff, which does not really work. (I've been
annoyed by this).

--
Alvaro Herrera http://www.PlanetPostgreSQL.org/
"Postgres is bloatware by design: it was built to house
PhD theses." (Joey Hellerstein, SIGMOD annual conference 2002)