Re: [Gnu-arch-users] Re: darcs vs tla

[Stephen J. Turnbull]

>>>>> "John" == John A Meinel <John> writes:

    John> In tla, though, you don't need the binary distinction
    John> because it always preserves the blobs you give it. If diff
    John> is able to create a textual patch, that is what you get,
    John> otherwise it reverts to the whole blob approach.

I don't really understand why darcs requires this, but one thing I
noticed is that darcs represents a binary file as hex inside the
patch.  The patch is serialized in a single file, none of this bizarre
"changeset is a whole tree with lots of stuff you've never seen before
in it".

Another thing about darcs is that darcs really does think of patches
as an operator algebra.  darcs is not a revision control system in the
sense that CVS or arch is; darcs is an editor with changeset (rather
than text) semantics.  It's more clumsy to use than vi or emacs at the
moment, but that will change, I suspect.

David Roundy seems to be very concerned with these concepts.

    John> mv and darcs mv are about the same. But you can't use a GUI
    John> navigator and have it run darcs mv.

Speak for yourself.  :-)  *I* can, because my navigator is dired.
Even if it's not directly customizable (but I bet it is already), I
can fix that, simply by rebinding R to dired-darcs-mv in dired-map:

(defun dired-darcs-mv (target)
  (interactive "FMove to: ")
  (shell-command (format "darcs mv %s %s" (dired-get-filename) target)))

:-)

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.