Can fogbugz track case dependencies?

Question

Can fogbugz track case dependencies?

Posterity

Answer 1

Yes and no. Cases can be linked to each other, but if you're looking for a tree of cases (prerequisites and such), you need FogBugz 7 or later.

If you're using FogBugz 7.3 or greater, you can now use the Case Dependency Plugin, which was released in April 2011.

Bases

Answer 2

FogBugz has long supported case "relation", which creates an ad hoc link between cases simply through adding "case 1234" to any note. Downside: these are not removable, and this persists into FogBugz 7. (We tried to figure out how to do it right, but just ran out of time, so we left current behavior.)

FogBugz 7, newly released, has added parent-child hierarchy, to allow you to split up a master case into its constituent parts, or to aggregate similar requests under one umbrella case.

FogBugz 7 also offers milestone dependencies, where one milestone cannot be started before another is complete. This only applies to the scheduling features of the software. We don't actually prevent anyone from working on cases in the dependent milestone.

We feel these features represent the real world of dependencies as they exist among different parts of a project.

We intentionally did not implement any sort of Bugzilla-style blocking, for several reasons. First, it can be horrendously inefficient, allowing people to ignore work that they could easily do if it was in front of them. Second, it can cause a morass of interdependencies. Third, it also allows the use of the software as a social bludgeon, ("I can't start stubbing out functions until Jeff has finished his mock-ups.") which is something we try to avoid. We make social software... in that we prefer to let social problems be solved socially and software problems be solved with software. The intentional omission of blocking or hard dependency between cases is part of this philosophy.

That said, FogBugz 7 is highly extensible, with plug-ins, tags, custom fields, and lots of other goodies. If blocking is what you want, I'm sure someone will be able to cobble something together.

Allahabad

I agree completely that full blocking can lead to bad things. What I think would be more useful is a "do next" feature, where I can be prompted what case to work on next when this one is resolved. And this can be implemented easily as a plugin! Now just to find the time to do it... ;) – Superstratum

@Superstratum Actually you can do this now with the backlog plugin by just assigning an order to your cases. next in the order is the one you work on next. – Semirigid

Rich, I agree with your reasoning regarding blocking on start, but this reasoning does not hold for blocking on resolution. I do want to be able to build a chain where case X cannot be resolved until case Y is resolved. The parent-child relationship does not address this today. – Terret

Answer 3

You didn't define what you mean by dependencies exactly, but if you mean that the resolution of one case requires the resolution of others - formally the answer is no. However, you can refer to other cases from a base case and FogBugz will track the cross references. For example, if you say "see case 2031" in the text of one case, the 2031 portion will turn into a hyperlink and both cases will now report that they refer to each other (both forwards and backwards). It's a pretty cool feature actually.

Aforethought

Answer 4

FogBugz 7 now supports sub-cases. This may or may not solve your problem, depending on how you want to handle it.

Orleans