Poll: What bug tracking software do you use?
As requested by tmpk: http://news.ycombinator.com/item?id=901314
"Which of the following do you use in your job/start-up: Also, if you could provide the rough number of users that use your particular bug tracking system installation that would be helpful."
(I added a few extra of my own too)
(
72 comments
| 44 points
| on HN
)
This weekend I just finished getting the git and mercurial integration working. My eyes hurt.
More info about BugTracker.NET at http://ifdefined.com/bugtrackernet.html.
It's always interesting to see the writers of these tools "in person"! Thanks for posting.
It also comes with a helpdesk that allows features and suggestions be fed in.
In my opinion, whatever is handling your development process would ideally handle your bugs too. If the bugs aren't up there and visible as something you need to do before anything else then you don't have control of the quality of your products.
http://www.targetprocess.com/
Requires .Net, ISS, etc.
We have: Users: 7 Feature requestors: 8 Product owner: 1 Bug reportees: Thousands (anyone from any of the clients where our software is implemented).
Used in the past: Trac, Request Tracker ( http://bestpractical.com/rt/ ), FogBugz.
I like Trac better. It has a more intuitive interface, and with the various plugins you can argue it does way more.
I've previously used Unfuddle and found it easy and intuitive. I liked it a lot. I've also met the two guys who started it a couple times. They are based in Honolulu, it is in Ruby/Rails, and they host at Rackspace. They seem to be doing very well.
I've looked, but never used a few of the others like Github, Bitbucket, google code, and Trac.
I'm not crazy about it, but I've never met an issue tracking system I am crazy about. DevTrack can be configured pretty well to suit our workflow, but I still run up against inscrutable problems sometimes. This may be due in part to the configuration, but it says something that we're still having problems with that.
See: http://leetcode.net/blog/2009/01/integrating-git-svn-with-ma...
http://pivotallabs.com/users/dan/blog/articles/965-story-tas...
Previously I used FogBugz, which was a joy to use, how I miss it now!
EDIT: We are a team of 8 people using it, 2-4 per project.
I was eyeing youtrack the other day (http://www.jetbrains.com/youtrack/index.html), looks pretty nice, and at 150$ for 20 users it's pretty cheap.
Currently trying Lighthouse for a small 2 person project.
I'm actually thinking that Google Wave could be interesting here. We've all got accounts and I'm thinking of giving it a go just to see what happens.
The problem I have is how do you your tracker data out at the end of the project, for archival or moving to another tracker?
Do any trackers provide a simple export to XML?
I use BitBucket. It's not the best per-se but as I almost exclusively use Mercurial and host on Bit Bucket it makes sense.
I am not sure I would recommend it for serious projects; the hosting is stellar (and hg rocks as SCM :D) but if you have the time to integrate with Lighthouse or other such apps it's probably worth it.
Some sort of "roadmap" feature. The ability to manually mark a revision as being relevant to the ticket (like the auto feature you have). When a revision (or revisions) are related to a ticket display a prominent, separate, link(s) to them in the ticket. Due date and importance for tickets. Highlight the importance in the list view (e.g. urgent = red etc.). When a revision is "tagged" in the mercurial repository create a link to a "changelog" that lists fixed issues for that tag name (if it exists: version or milestone - or even a separate tagging system).
That's just a few off the top of my head. (sorry; I always seem to be dissing you guys :( not intentional)
I find it a bit frustrating (missing features; Unix compatibility issues), but the user interface was easier for our non-technical staff to learn than was the interface for RT, which was our initial choice.
https://www.labkey.org/issues/home/Developer/issues/list.vie...
If I had known then what I know now, I would not have gotten involved.
users: 20
Internally - 10, bespoke project stakeholders - up to 100 at a time, community developers - 100s (we've open sourced our core framework).