Brief article on benchmarks of Python repository with leading DVCSen

[Matthieu Moy]

"Stephen J. Turnbull" <stephen at xemacs.org> writes:

>  > I'm repeating myself, but you are _not_ explicit about the change it
>  > content with bzr. The changes in content are _detected_ after the fact
>  > by bzr, with a diff algorithm. As any decent VCS would do.
>
> No, decent VCSes don't use diff algorithms,

THEY DO, FOR THE CONTENT OF FILES.

(in case you didn't notice, the sentence contains the word "content",
and neither "file name" nore "file identity")

Try it: open a file in your text editor, add a line in the middle,
save, run "$vcs diff". It will show

--- file
+++ file
 context
+line you've added
 context

And now, tell me how $vcs noticed that the line has just been added,
if not with a diff algorithm.

(And no, diff algorithms are not 100% reliable, it does happen that
the diff doesn't actually show your intention. Not a coincidence if
there are actually several diff algorithms.)

> they flag files with changed content. That way almost all changes
> that humans make will lead to a perception of O(1) performance for
> $VCS diff (and related operations).

That's a totally different issue. Fast performance for diff can be
achieved with a purely snapshot oriented system (like Git), without
marking who changed from a revision to another (hint: checksums).

-- 
Matthieu