[darcs-users] Re: Attaching lots of data to a patch

[Albert Reiner]

Thanks for your reply:

[Aaron Denney <wnoise at ofb.net>, Fri, 6 May 2005 16:59:38 +0000 (UTC)]:
> Sounds like it should go in some sot of documentation of the code.  The
> relations and how they're derived justify not only the change, but the
> new form.

That's precisely what I want to avoid: As the theory we're working on
keeps changing, the relations in the program necessarily also change
all the time.  There is no point in keeping the derivation of what is
likely to be just a temporarily useful result in the documentation of
the program.  OTOH, even though it is just some routine calculation I
still don't want to loose it, in case it becomes handy later on.

Anyway, if there are no better ideas than those I came up with myself,
I'll probably go with the additional repo for this kind of stuff.

Regards,

Albert