Slashdot Mirror
Bug Reporting Etiquette
Jamie Zawinski writes "Mozilla.org has a new article on Bugzilla Etiquette. Relevant to more than just Bugzilla, this should be required reading for anyone who wants to file a bug about any product, no matter what bug tracking system is in use. I especially like the mention that "'Open Source' is not the same as 'the developers must do my bidding.'"" Update: 03/19 21:26 GMT by T : If that link doesn't work for you without cutting and pasting, reader Stephen Ostermiller suggests "you might want to use this link which appears to be the same document mirrored elsewhere."
Doing system testing, one of the most common mis-reported bugs is due to "pilot error", whether that be a misconfigured server, improper use of the application, a missing configuration parameter, etc.
Why do these result? Because people often do not read the documentation (setup instructions, release letters, etc).... OR because things like a README were not included!
--------
Free your mind.
Realize that the developerr has NO idea what is going through your mind, or what you did to make the bug happen.
CLEARLY report the problem like you are explaining it to a 3 year old.
Report your entire platform. OS, vers, ram, vid card, prod vers, etc.
If you take the time to clearly report the bug then you will not waste your time and the developer's time by having to respond to another email asking for more detail.
You SAVE time by being extra clear when you report the issue.
When I was at Macromedia, I determined that a mis reported bug uses 4-6x the manpower than one clearly reported one. This also means that poorly written bugs limit the number of bugs that can be fixed before shipping because of the extra manpower and time used to pass them back and forth.
Good bug:
QA reporter -> Qa manager -> Bug meeting -> Developer -> Bug addressed -> Original reporter -> Bug closed
Bad bug:
QA reporter -> QA manager -> Bug meeting -> Developer -> Bug addressed -> Bug returned to QA -> More info dded, more time spent getting more details -> bug returned to developer -> Bug addressed -> Original reporter -> Bug closed
I know this doesn't look like 6 x the time but when measured, it got up to that high. Especially if a bug has to return more than once.
- Zav - Imagine a Beowulf cluster of insensitive clods...
I have to salute any of these developers that don't rip any of the whiners a new one. I recently went to Magic Workstation which is a site for beta game software. The message boards are filled with crap like:
You mean we have to wait another half month to download it? What are the updates, anyways?! and
i AM WAITING TO LONG!!!!!!!!!!!!!!!!.
Personally, I'd be giving these guys the verbal-bitchslap, but they and other developers I've seen are pretty calm and take whining as a part of the process.
It doesn't matter how politely I submit my bug report, it still gets ignored with the classic phrase, "we are concentrating our resources in other directions"
:-) But it's still a fair response - "thanks for the bug report, but we aren't going to fix it. However, if it's important to you, you have the source."
I've never seen anyone use such a management-speak phrase in bugzilla.mozilla.org.
Gerv
(document author)
The Bugzilla Helper is actually quite good. (You'll probably have to cut'n'paste this link)
http://bugzilla.mozilla.org/enter_bug.cgi?format=g uided
It forces you to search for a dupe before submitting. It forces you to enter build IDs, steps to reproduce, and other critical information.
It should also search for dupes before letting you submit the bug based on keywords in your bug report.
Reporting bugs through the Bugzilla Helper should also be made mandatory. Sure that may annoy power users. However, even power users are liable to miss critical info in their bug reports.
Dude. Chill. Remember that while reading TFM is an excellent map from {Linux commands} to {what they do}, there is not a simple inverse map from {what you want to do} to {Linux command}. apropos and Google are decent steps, but they don't always give you the answer even if you know how to search.
It is much easier just to ask someone who knows what they're doing and can answer in 5 minutes than to spend hours googling and crawling the man pages. I have several friends who have been using Linux for a year or so, and when we can't figure something out, we IM / write each other and ask, and it's better that way. If you have a fish that you're not going to eat, why not give it to a man?
I hereby place the above post in the public domain.
One of the biggest problems with bug management is the ability of the bug viewing audience to find bugs that may match the problem they are having.
I've worked within a major corporation (RAID queries) and also worked closely with Mozilla and in both instances I found it extremely frustrating to find bugs, often when I knew the exact keywords that are associated with a given bug.
If there was a search tool akin to Google for searching accurately through large bug databases where users could easily find bugs then the issue of duplicates would be solved almost entirely.
"Unlike most of you, I am not a nut." - Homer J. Simpson
Dude. Chill. Remember that while reading TFM is an excellent map from {Linux commands} to {what they do}, there is not a simple inverse map from {what you want to do} to {Linux command}. apropos and Google are decent steps, but they don't always give you the answer even if you know how to search.
Ok, asking for a hosting providor that costs less than $100 when I post a comment that says, "There are plenty of hosting sites that offer dedicated servers for around $100" is the line that I will not respond.
It is much easier just to ask someone who knows what they're doing and can answer in 5 minutes than to spend hours googling and crawling the man pages. I have several friends who have been using Linux for a year or so, and when we can't figure something out, we IM / write each other and ask, and it's better that way.
There is a difference between IM'ing a friend and emailing someone you don't know. Finding out a simple solution in google takes them less time.
If you have a fish that you're not going to eat, why not give it to a man?
Because it promotes laziness, however, I will gladly trade it for services. I do not give things away, I make people work for them. Very few things in life were given to me, and I had to work for everything I have. I'm not saying it's the way I'll raise my kids, but I'm not going to help someone out just because they need a fish. They're going to work for it, then they get a fish.
Dacels Jewelers can't be trusted.
Respect is a two way street. It would be nice if the Mozilla Developers would show the same respect they desire from bug reporters to bug reporters. The fact is you can get rude and dismissive comments from the Mozilla developers even if you do your absolute best to be as respectful to them as possible. In fact you can get the attitude from the Mozilla developers even before you make a comment see here.
That said, I think the Mozilla developers are doing a GREAT job and I have a ton of respect for them. I just woouldn't ever want to work with them.
hear me out. sorry for the long mail.
i use linux 24x7 on various machines and also use windows and mac's and solaris. i aint a newbie. i remember when i was playing around with apt-get
i removed the
once by mistake [instead of just the archives/*.deb] and tried to use apt after
that. the errors it spewed out were _BS_
and totally non-related, especially it spewed
it out multiple times. me not being a debian developer [yet] and thinking that this is something that can be improved - i looked at the bug reports. none of it at that time had this, so i thought i would report. before that i decided to check out the #debian channel on irc.debian.org whether this was ok.....my mistake! it was a horrible experience. basically the people said i was a) a troll b) just want to point out problems in debian c) a deadrat user
d) maybe i should stick with windows e) some affirmative responses. no, it was not SOME
folks just trolling as usual on irc, they are easy to spot. its more the uniform mindset of the "power users" and "developers". i got so pissed off and i have never been in irc after that. well i _know_ these things happen and can shrug them off. i also can assure you that i am not hyping up a single incident.
coming to the pt, wtf are you guys thinking newbies will do ? you are telling bug reports to be "proper" ? yeah right. i cant begin to tell you the hypocrisy that exists in the minds of most of the developers/power users.
frankly it feels like most of the folks write
open source software for _their_ name and use
it to show _their_ coolness rather than anything else. i know everybody isnt like that, probably pareto's rule applies in that 20% of the folks do 80% of the work.
thanks for reading.
vv