we don't have a bugzilla

Raphaël Jacquot wrote:

would seem like a good idea, no ?

http://www.murrayc.com/blog/permalink/2008/04/25/postgresql-has-no-bugzilla/

Before you come trolling on this (or any other) subject, please read the
voluminous debates that have taken place about it. Apparently you think
it's something we have never considered, which in light of the product
we maintain would be more than remarkable.

Having done that, please endeavour to make an actual contribution to the
discussion.

cheers

andrew

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

On Sat, Apr 26, 2008 at 4:13 PM, Andrew Dunstan wrote:

Before you come trolling on this (or any other) subject, please read the
voluminous debates that have taken place about it. Apparently you think it's
something we have never considered, which in light of the product we
maintain would be more than remarkable.

Having done that, please endeavour to make an actual contribution to the
discussion.

Hi Andrew,

Let's be fair. It would be an almost impossible task to make any
sense of the archives on this topic without dedicating tens of hours
to the task, and having access to a better way of reading the archives
than is offered at archives.postgresql.org.

Raphaël, there have indeed been vast debates about this. My version
of the "executive summary" would be that some people want a tracker,
some people think email is good enough, and the people who do want a
tracker have different opinions about which tracker would be best for
the project. There's a wiki page about the (daunting) set of features
which would be required in a tracker to make everybody happy[1]http://wiki.postgresql.org/wiki/TrackerDiscussion.

We are currently experimenting with using a wiki page to organise our
queue of patches[2]http://wiki.postgresql.org/wiki/CommitFest:May, and there has been some effort to set up a test
bugzilla installation[3]http://wiki.postgresql.org/wiki/Tracker:BugzillaTest but as yet there has been no viable consensus
as to adopting a particular tracker.

Cheers,
BJ

[1]: http://wiki.postgresql.org/wiki/TrackerDiscussion
[2]: http://wiki.postgresql.org/wiki/CommitFest:May
[3]: http://wiki.postgresql.org/wiki/Tracker:BugzillaTest

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)
Comment: http://getfiregpg.org

iD8DBQFIEsya5YBsbHkuyV0RAnOyAKDI7Ygcb8m689IJtMcGQtmMq+5CPwCeMIsk
/+eXsTtGNVVWIfsvNsEyW7A=
=6DTB
-----END PGP SIGNATURE-----

Brendan Jurd wrote:

Having done that, please endeavour to make an actual contribution to the
discussion.

Hi Andrew,

Let's be fair. It would be an almost impossible task to make any
sense of the archives on this topic without dedicating tens of hours
to the task, and having access to a better way of reading the archives
than is offered at archives.postgresql.org.

+1

The idea that anyone would waste their time (and yes it is a complete
waste of time) reviewing the "archives" which are almost impossible to
search unless you know "exactly" what you are looking for is ridiculous.
Andrew, I am frankly surprised that you would have that response for the
individual.

How do you know he was trolling? Perhaps he is just surprised (which is
the impression that I got) that considering we are such a mature project
that we don't have a bugzilla or other such beast.

Sincerely,

Joshua D. Drake

Joshua D. Drake wrote:

Brendan Jurd wrote:

Having done that, please endeavour to make an actual contribution
to the
discussion.

Hi Andrew,

Let's be fair. It would be an almost impossible task to make any
sense of the archives on this topic without dedicating tens of hours
to the task, and having access to a better way of reading the archives
than is offered at archives.postgresql.org.

+1

The idea that anyone would waste their time (and yes it is a complete
waste of time) reviewing the "archives" which are almost impossible to
search unless you know "exactly" what you are looking for is ridiculous.
Andrew, I am frankly surprised that you would have that response for
the individual.

I entered "bugzilla" on the archives search page and got this link,
right out of the recent discussion, at the top of the list:

http://archives.postgresql.org/pgsql-hackers/2008-04/msg00764.php

That and a few similar results might have given the OP some hints about
what people are thinking and why.

So your suggestion that the search engine is useless is just plain wrong
in this case.

How do you know he was trolling? Perhaps he is just surprised (which
is the impression that I got) that considering we are such a mature
project that we don't have a bugzilla or other such beast.

Well, his post added precisely nothing to the debate. Personally, when I
find something surprising my first reaction is to go looking for a
reason, and only to ask about it if I can't find answers. As I have
demonstrated, finding clues would have been absurdly simple, despite
your assertion to the contrary. I just don't see any point at all in
posting a link to some random blog where mostly anonymous posters
comment on our lack of a bug tracker. A reasoned contribution to the
debate on the other hand will be welcome.

Perhaps it wasn't a troll. If it wasn't I apologise. My other points
remain, however.

(BTW, I have long been on the record as being in favor of using a
tracker for both bugs and features, and I did work some years ago on
making mainline Bugzilla fully support Postgres.)

cheers

andrew

Andrew Dunstan wrote:

I entered "bugzilla" on the archives search page and got this link,
right out of the recent discussion, at the top of the list:

http://archives.postgresql.org/pgsql-hackers/2008-04/msg00764.php

That and a few similar results might have given the OP some hints about
what people are thinking and why.

How would he know to search at the archives?

* There is no archives signature at the bottom of -hackers lists

* The mail-pref link when followed doesn't provide a link to the archives

* The website? Hmm I guess I click support and then a link called
mailing lists, oh and then there is a link somewhere down the middle of
the page that says, "archives".

Lastly, why would I search the archives?

There is this vast assumption that anyone (I do this too) will just
"know" to do something in this community. I think that assumption is far
off base.

IMO, a more appropriate response would have been something like:

Thanks for the discussion, this is something our community is currently
considering. You can view these two links if you are interested.

<link>
<link>

All we did with our response is say...

Hey, you are a troll, go away. When in fact he likely wasn't.

Well, his post added precisely nothing to the debate. Personally, when I
find something surprising my first reaction is to go looking for a
reason, and only to ask about it if I can't find answers. As I have

As my wife often tells me, "you are not like everyone else".

The first natural response from most people that I encounter, is to ask
not to research.

demonstrated, finding clues would have been absurdly simple, despite
your assertion to the contrary. I just don't see any point at all in
posting a link to some random blog where mostly anonymous posters
comment on our lack of a bug tracker.

Well I won't disagree that the post was offtopic for the list.

(BTW, I have long been on the record as being in favor of using a
tracker for both bugs and features, and I did work some years ago on
making mainline Bugzilla fully support Postgres.)

Right, which is also why I was surprised by your reply.

Sincerely,

Joshua D. Drake

"Joshua D. Drake" <jd@commandprompt.com> writes:

How would he know to search at the archives?

If he knew enough about the community to post in -hackers (as opposed
to, say, -general or -novice) he should certainly have heard of the
mailing list archives. *You* might find 'em useless but I don't.

regards, tom lane

Tom Lane wrote:

"Joshua D. Drake" <jd@commandprompt.com> writes:

How would he know to search at the archives?

If he knew enough about the community to post in -hackers (as opposed
to, say, -general or -novice) he should certainly have heard of the

You think so? (not being sarcastic).

mailing list archives. *You* might find 'em useless but I don't.

I didn't say I found them useless. I said they are useless unless you
know what you are looking for. I am pretty sure that when it comes to
PostgreSQL, you always know what you are looking for.

Anyway, now I am off-topic, so I am done with this thread.

Sincerely,

Joshua D. Drake

On Sat, Apr 26, 2008 at 08:54:46AM -0700, Joshua D. Drake wrote:

How would he know to search at the archives?

* There is no archives signature at the bottom of -hackers lists

Maybe because there's a perfectly functional archive link in the mail
headers? And because there's an RFC that tells us how such headers
are supposed to work?

A

--
Andrew Sullivan
ajs@commandprompt.com
+1 503 667 4564 x104
http://www.commandprompt.com/

Andrew Sullivan wrote:

On Sat, Apr 26, 2008 at 08:54:46AM -0700, Joshua D. Drake wrote:

How would he know to search at the archives?

* There is no archives signature at the bottom of -hackers lists

Maybe because there's a perfectly functional archive link in the mail
headers? And because there's an RFC that tells us how such headers
are supposed to work?

A

Oh so there is.

As a lot of people use gui apps, (I do seem to recall that mail cli
shows the full headers) most would never see the detailed headers of an
email. I know I very rarely look at the full source of an email.

--

Shane Ambler
pgSQL (at) Sheeky (dot) Biz

Get Sheeky @ http://Sheeky.Biz

Shane Ambler wrote:

Andrew Sullivan wrote:

Maybe because there's a perfectly functional archive link in the mail
headers? And because there's an RFC that tells us how such headers
are supposed to work?

As a lot of people use gui apps, (I do seem to recall that mail cli
shows the full headers) most would never see the detailed headers of an
email. I know I very rarely look at the full source of an email.

So the proper thing to do is complain to the writer of the GUI app so
that it has an option for showing the list headers, perhaps adding a
menu entry when they are found.

CLI email clients do not typically show the full headers -- the user
must ask for them, just like a GUI app. Most of the time I don't look
at the source of emails either.

--
Alvaro Herrera http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.

On Sun, Apr 27, 2008 at 11:58:01AM -0400, Alvaro Herrera wrote:

So the proper thing to do is complain to the writer of the GUI app so
that it has an option for showing the list headers, perhaps adding a
menu entry when they are found.

At least in the case of Thunderbird, you already have an option:

http://www.juergen-ernst.de/addons/mailinglistheader.html

According to the manual, claws has a whole Message/Mailing-List
submenu with built-in support for this. (This took me approximately
10 seconds to find through Google. The claws manual looks pretty
good, I have to say.)

I was unable to find the way to to this in Lookout! I don't use the
Mac OS Mail client, but a quick look didn't yield anything obvious.
However, these both allow you to look at the raw mail message, which
has the header.

Given that List-Archive was standardized in RFC2369, which came out in
1998, and that virtually every list server software shipping since
about 2000 has it built in and turned on by default, I fail completely
to see how using something as free-form as a signature footer is
supposed to be an improvement.

A

--
Andrew Sullivan
ajs@commandprompt.com
+1 503 667 4564 x104
http://www.commandprompt.com/

andrew@dunslane.net (Andrew Dunstan) writes:

Raphaël Jacquot wrote:

would seem like a good idea, no ?

http://www.murrayc.com/blog/permalink/2008/04/25/postgresql-has-no-bugzilla/

Before you come trolling on this (or any other) subject, please read
the voluminous debates that have taken place about it. Apparently you
think it's something we have never considered, which in light of the
product we maintain would be more than remarkable.

Having done that, please endeavour to make an actual contribution to
the discussion.

This seems more than a tad unfair.

Yes, it is certainly fair to observe that there have been voluminous
debates. But it will take a whole lot of "trolling around" in the
archives to figure out the shape of the *conclusions* of those
debates.

Seeming relevant conclusions:

- Yes, there probably ought to be some sort of structured bug tracker.

- HOWEVER, there are many who prefer to use email for their work, as
opposed to being forced into some sort of "webby thing."

- That being said, experimentation is taking place for the
"commitfests" with using a wiki to track statuses of patches and
discussions.

- It is hoped that out of this experimentation, patterns surrounding
what is *truly* needful to structure will emerge. There is reason
to be hopeful about this, as the wiki-based process has been showing
some useful structures to impose already.
--
select 'cbbrowne' || '@' || 'cbbrowne.com';
http://www3.sympatico.ca/cbbrowne/spreadsheets.html
"I find it hard to believe a professional programmer could support the
concept of Open Source." -- David J. Owens <owensdj@home.net>

Chris Browne wrote:

andrew@dunslane.net (Andrew Dunstan) writes:

Raphaël Jacquot wrote:

would seem like a good idea, no ?

http://www.murrayc.com/blog/permalink/2008/04/25/postgresql-has-no-bugzilla/

Before you come trolling on this (or any other) subject, please read
the voluminous debates that have taken place about it. Apparently you
think it's something we have never considered, which in light of the
product we maintain would be more than remarkable.

Having done that, please endeavour to make an actual contribution to
the discussion.

This seems more than a tad unfair.

I'm sorry, open season to beat me up on this closed yesterday. :-)

cheers

andrew

On Mon, Apr 28, 2008 at 11:55:18AM -0400, Chris Browne wrote:

Yes, it is certainly fair to observe that there have been voluminous
debates. But it will take a whole lot of "trolling around" in the
archives to figure out the shape of the *conclusions* of those
debates.

As one of those confused, it would be really nice if someone could
summarise it all on a wiki page that we can point people to.

Have a nice day,
--
Martijn van Oosterhout <kleptog@svana.org> http://svana.org/kleptog/

Am Montag, 28. April 2008 schrieb Martijn van Oosterhout:

As one of those confused, it would be really nice if someone could
summarise it all on a wiki page that we can point people to.

http://wiki.postgresql.org/wiki/TrackerDiscussion