[darcs-devel] GHC's darcs evaluation wiki page

[Mark Stosberg]

>
> "bugs: we run into darcs bugs other than the conflict/merging bug on
> a regular basis."
>
> This is true, but I think things are improving.

Yes they are. I would say that the bug tracker has about every
significant bug in it. Darcs 2 addresses many major bugs, and our open
bug count is lower than comparable projects (SVN, for certain). Git
doesn't even have a bug tracker. The statement should really be
qualified. If we don't know what specific bugs they are running into,
there's no way to address them, short of solving every bug.

> "Uncertain future: no critical mass of hackers/maintainers. The
> technical basis is not well enough understood by enough people."

I know that as potential hacker I'm turned off when I see lines like this:

  wspfr _ _ NilRL _ = return Nothing
  wspfr jn matches (p:<:pps) skipped

( from SelectChange.lhs, line 183)

In the programming culture I come from, if you name a variable "wspfr",
"nilrl", "jn", "p" or "pps", there ought to be a comment close by
explaining what the heck you meant.

This compact, cryptic notation definitely shows roots in cultural math
equations, and in my view makes the code much less accessible.

     Mark

-- 
http://mark.stosberg.com/