Should I keep todo lists in source control?

Question

I have some trouble figuring out the best way to store my programming todo lists.

I consider the following:

• one todo list in source control for each project
• a master todo list (with general tasks) in a personal folder in source control

How do you find that?

What would you suggest?

Edit:

Thanks for the suggestions. I use a bug tracking system (BugTracker.NET) for bugs and for tasks that involve requests from other people such that they can see the status. And I use //TODO in code.

I have a lot of additional notes about what to do. Would you recommend also putting that in the bug tracker (especially if it is not possible to put them as //TODO in code)?

Answer 1

why don't you use a bug tracking system?

examples : Bugzilla Mantis Trac

and many others ...

Answer 2

You should keep your todos in your code. Preferrably linked with a bugtracking program. And then you should use a documentation generation program that catches all todos and writes them to a list with links to the relevant parts of the code.

A good example is Doxygen. Given the todo in code:

// TODO: fix potential non-assigment of var
int my_var;

Doxygen will be able to pull this information (you can even set up filters for arbitrary annotations, like FIXME BUG LOOK_HERE and so on) and a) leave a todo entry for the specific class/interface and b) compile a list of todos for the complete project.

Also, your todos and todo-lists will be version controlled and the lists (ie. the documentation) are easily generated from scratch.

So, to sum: A combination of Doxygen, a scm system (any will do) and bugzilla will get you up and running in no time.

Update: Check out this git hook that creates Github issues from TODOs in your checkins

And a general note to the question of whether or not using TODOs in your code is a Good Thing(TM): A fool with a tool is still a fool.