Re: The Problems of Perl: The Future of Bugzilla

Aaron Trevena wrote on 5/12/07 6:08 AM:

> After so many years I'm pretty surprised that none of
the bugzilla
> developers have worked on something to make bundling
CPAN modules
> easier, or even used anything like PAR - every wheel
seems to be
> reinvented.

This is mostly in response to complaints from people trying
to install
Bugzilla.  Our end-users seem to detest external
dependencies, so we
always have to evaluate how much bang we're getting out of a
perl module
before adding it, and if it's a wheel that's ridiculously
easy to
reinvent, we probably will, to avoid the dependency.  If
it's a mature
module with lots of bang for the effort (it will actually
save us a lot
of time writing code) then we'll probably use it.

> I'm also surprised that there is no pagination for
results, and that
> every result for a query - even if there are thousands
are always
> fetched and loaded into tons of hashes - no wonder it's
memory hungry
> if you write code that eats memory like candy.

Yeah, there's times when you actually want all of them at
once, but it
shouldn't be the default.

> I spoke to Max about solving this latter problem, and
I'm still
> interested in doing so, but to be honest the
combination of the
> Mozilla Foundation/Corporation leaving the project to
rot with no
> funding or support,  so much crufty out of date code
and a couple of
> vocal people clamouring to stop useful development and
instead switch
> to more fashionable languages makes me think
"stuff it, it's easier to
> use rt or trac than put up with that".

Given the discussion that's happened here so far I think
it's extremely
unlikely that Bugzilla will switch to a new language, at
least for the
core code.  I think the discussion is worth having because
its a
critical analysis of our code and it helps us understand the
code better
and points out some things we need to pay attention to, but
I don't see
the project moving to a new language anytime soon.

> If anybody is actually interested in some constructive
feedback on
> what needs doing to improve the bugzilla project,
that's fine, but I'm
> not going to waste my time, when people who make money
from bugzilla
> are wasting there's rewriting it poorly in a language
that makes them
> feel warm and fuzzy.

I tend to agree with you here.

-- 
Dave Miller                                   http://www.justdave.net/

System Administrator, Mozilla Corporation      http://www.mozilla.com/
Project Leader, Bugzilla Bug Tracking System  http://www.bugzilla.org/

-
To view or change your list settings, click here:
<http:
//bugzilla.org/cgi-bin/mj_wwwusr?user=bondyahoo.com>

On Sat, May 12, 2007 at 09:27:00AM -0400, David Miller
wrote:
> Our end-users seem to detest external dependencies, so
we
> always have to evaluate how much bang we're getting out
of a perl module
> before adding it, and if it's a wheel that's
ridiculously easy to
> reinvent, we probably will, to avoid the dependency. 
If it's a mature
> module with lots of bang for the effort (it will
actually save us a lot
> of time writing code) then we'll probably use it.

Why make the end user pay the tax?  Package all the modules
we need with the bugzilla distro.  This is how Krang and
Twiki operate to varying degrees.  Whining about the
failures of Perl's module system when we're not doing what
works well for other groups shipping Perl web applications
seems a bit strange to me.

I'm not sure what scripts and tools krang and twiki use to
maintain their own lib trees, but it can't be harder to
write than dealing with all of the end user complaints.

-- 
</chris>

"The problem with troubleshooting is that trouble
shoots back!"
-
To view or change your list settings, click here:
<http:
//bugzilla.org/cgi-bin/mj_wwwusr?user=bondyahoo.com>

On 5/12/07, David Miller <justdavebugzilla.org> wrote:

> Bugzilla.  Our end-users seem to detest external
dependencies, so we
> always have to evaluate how much bang we're getting out
of a perl module
> before adding it,

If so, then I really wonder about the rationale to replace
the old
framework for sending mails (as used in 2.22) with the new
framework.
I fail to see the advantage, but I clearly do see the
disadvantage
with the approximately 20 or so different modules for the
simple
purpose of sending mail.

Jochen


-- 
My cats know that I am a loser who goes out for hunting
every day
without ever returning as much as a single mouse.
Fortunately, I've
got a wife who's a real champ: She leaves the house and
returns within
half an hour, carrying whole bags full of meal.
-
To view or change your list settings, click here:
<http:
//bugzilla.org/cgi-bin/mj_wwwusr?user=bondyahoo.com>