swarm-support
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Bug Reporting (was Re: Memory Leak-.dll problem)


From: Alex Lancaster
Subject: Bug Reporting (was Re: Memory Leak-.dll problem)
Date: 09 Mar 2000 02:47:31 -0700
User-agent: Gnus/5.070095 (Pterodactyl Gnus v0.95) Emacs/20.4

>>>>> "MD" == Doug Donalson <address@hidden> writes:

[...]

D3> I would love to give you 5 lines of code that reproduce the
D3> problem but the problem can't be recreated in 5 lines of code.  I
D3> also went to the extra effort of making it possible for you to be
D3> able to compare two versions of the sim, one with explicit removal
D3> of events from the event list and with two comment lines, one
D3> without.  Sorry if I put too much in.

MD> 4) While it is helpful to know "this didn't happen before and now
MD> it does" -- that alone does not mean there is a bug.  (Yes, I think
MD> there is one in this case.)

D3> Marcus, believe it or not, I am not a complete idiot.  You wanted
D3> specifics and I gave them to you.  It should save you quite a bit
D3> of time to know that the error occured with the updated dll's and
D3> not prior to them.

Whoa folks!  A couple of buckets of virtual cold water please! ;-)

Marcus was not intending a personal attack.  He really is actually
trying to fix the problem, not trying to make anyone feel small.  It
is the cumulative experience of dealing and integrating bug reports
from all around the Swarm community, that essentially he is attempting
to sum up.

I think the point to realise here is that writing a good bug report is
actually not an easy task, and people need to at least have a grasp of
what is required to do it well.  Behind what Marcus is saying is the
fact that good reporting is like writing good code: it can take a
considerable time and energy.  It doesn't have to be `hard' in the
sense of quantum mechanics, but it does require a certain discipline
of thinking and rigourous attention to detail that is not entirely
natural to most people (even most programmers).  I'm still learning
how to identify bugs well and describe them concisely, and it can
often require as much energy as writing the code itself.

When I started working on open source projects, I occasionally had
similar reactions to folks on various mailing lists.  I already
considered myself a decent programmer (having been at it for a 2-3
years in academia then industry), so I thought why the perceived
aggression when it came to reporting a bug?  Should be the simplest
thing in the world right?

Eventually I realised that the arguing, pruning, dissecting and
analysing that goes into the sometimes dense and heated exchanges that
is summed up by the phrase `reporting a bug' is actually pretty
helpful.  It stopped my woolly and wishful thinking, and helped me
strip the problem to the bare essence and helped tune and hone my
programming skills.  I actually learned a lot from those people and
interactions, however abrasive it seemed at first.  Free/open source
software simply wouldn't have advanced as much as it has in the last
ten years if it wasn't for exactly this kind of programming rigour.

Some companies have entire *divisions* employing nobody except testers
and debuggers who job is to do precisely these things, and do
practically nothing else; and there are textbooks on the subject (not
that I'd suggest that you go out and read too many of them).

One really good (and quick!: one page) reference is the `Bug
Reporting' sections that are included with the documentation of a lot
of GNU tools, such as Emacs or binutils.  In one page they summarize
the essentials:

 http://www.gnu.org/manual/binutils-2.9.1/html_node/binutils_23.html

Lurk on some of the kernel, glibc or general GNU/Linux tools mailing
lists, get a feel for the kind of exchanges that happens on those,
then you'll realise that Marcus is pretty damn tolerant as software
maintainers go... ;-)

My $0.02...

Alex
-- 
Alex Lancaster * address@hidden * www.santafe.edu/~alex * 505 984-8800 x242
Santa Fe Institute (www.santafe.edu) & Swarm Development Group (www.swarm.org)

                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]