Eric Kow writes:

 > So this discussion seems to mix tactics (choice of words) with strategy
 > (choice of things to say).

Good point!

 > Maybe we really should stop trying to say that darcs is smart and
 > focus on the fact that it is very easy to use, that it has a unique
 > user interface that lets pick and choose patches without any fuss
 > (maybe we can talk more about a notion of selective undo).

I think that's right.  You know, Linus makes a big deal of how git is
*stoopid*.  That doesn't seem to bother git fans one bit. ;-)

OTOH, we should not forget that implementing patch theory is a large
part of what makes Darcs easy to use.  That's what I was trying to get
at with the talk about Darcs having "named patches" that you can
select (at least conceptually, in simple cases) independently of each
other.

Maybe "patch-focused" is a good term?  But I really want to link that
to a higher conceptual level where the user selects features to merge
in rather than individual lines of code (which is what a patch
"really" is, right?)  So (sorry Daniel!) I'm not happy with that as a
buzzword, either. :-)  Somehow I really want to say "Darcs is based on
a theory of merging features into a line of development" but without
telling any lies. :-)

BTW, I'm not very happy about removing the words "patch theory" from
the home page myself.  But that may be the easiest way to discipline
ourselves to remember what we were looking for before we got hooked on
the patch theory Kool-Aid.
_______________________________________________
darcs-users mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-users

Reply via email to