Chessguy pointed out that it's currently hard to play along with the monad wars code.

It would be nice for the posts to be “literate haskell”, where sections preceded by “>“
characters are valid Haskell. The idea is great – that you can mix
sections of introduction and description with sections of actual code,
ending up with an article that is also executable code! Which is very
much the style of these posts, but right now I'm being too lazy to go
that extra step:

• sometimes I show multiple version of the same function, (some of them might not work)
• I tend to introduce imports as needed but (I believe) they need to be at the top of the file
• I don't always repeat functions from earlier posts but just refer to them

So I've posted the current source to my subversion repo.
As you can see they're currently related to the number of the
associated post, and contain different areas of functionality: this is
actually how I'm working on it for the moment – I'm hoping to put
together some of the pieces in part 5 or so (Blog Driven Development is
a rather odd way to structure your work but there you go…)

Update: Vincenz on #haskell convinced me that I really should try literate haskell – watch this space…

I'm now probably going to fall off the internet for a week while I move country and job.  I'll be at the London Perl Workshop this Saturday and will talk (for a whole 5 minutes!) about Monad Wars. Maybe see you there :D