News for hackers. Stuff that matters.
Re:Ask HN: Best issue tracker in 2015? by WorldMaker on Friday August 07, 2015 @05:12PM (#10025046)
Honestly, as something of a minimalist, I've been happiest with Github's issue tracker lately. GitLab's looks similar in screenshots, but I've not used GitLab. (Github also has an Enterprise version if you need on-premises install these days.)

It has just enough features to get the work done but also not feature/configuration overload so that it's also not eating a lot of "management time from you". Github in particular you can tell that Github themselves dogfood their issues and PRs and want them to be great products for themselves.

There are actually more than one "Kanban overlays" for Github issues I've seen. The first example to my mind is https://waffle.io/

I would presume if GitLab has similar APIs for their issues there may be a Kanban overlay for it as well.

If you are already using GitLab then maybe you should just use GitLab's issue tracker? I think the convenience of using your source control host as your issue tracker is a big, useful deal. (#synergy)

Re:Ask HN: Best issue tracker in 2015? by trcollinson on Friday August 07, 2015 @05:26PM (#10025123)
It took me a bit of "Selling" to management to begin to use Github issues as a solution at a number of organizations but every time I have it has been a rousing success! Most of the concerns around using issues have been that there "aren't enough fields for all the things we want to track on an issue." After a bit of discussion it is usually decided that they should try to track without so many fields and see what happens.

Ultimately, people love the simplicity and the integration with the repository. I know for a fact that GitLab's has a very similar mechanism and works wonderfully. Good suggestion and a very successful one!

Re:Ask HN: Best issue tracker in 2015? by WorldMaker on Friday August 07, 2015 @05:52PM (#10025235)
Issue tracking is one of those areas in computing where there's a compulsion to over-organize things and store a lot of peripheral information that may or may not be germane to the actual problem of issue tracking. It's fascinating how much applying the YAGNI [You Aren't Going to Need It] principle to Issue Tracking reveals. A lot of fields are either to duplicate stuff that should already be searchable (in the main body, for instance) in a good issue tracker [1]. A bunch of other fields are business process related and it's not surprising how many of those business processes themselves could use a good YAGNI refactor, and the few that remain after tree-shaking are often easy enough to store in a label, a "tag" in the title of an issue, or a status on a kanban board.

[1] I honestly think that a lot of blame goes back to the very slow search of early issues trackers (including and especially, but not alone, Bugzilla) and needing to lump things into specialized fields to get any sort of search improvements.