Blog post on git vs. bzr movement history

[Alexander Belchenko]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

James Westby пишет:
> Hi,
> 
> I don't know if anyone has seen this
> 
> http://renesd.blogspot.com/2007/07/revision-tracking-of-functions-is-more.html
> 
> I tend to agree with him, and I was going to write something similar,
> but he beat me to it.
> 
> The after the fact processing to detect moves and copies seems to
> support this much better, and upfront specification would become unruly
> for some changes I would guess.
> 
> Does anyone have any insights on the issue?

I think it's thoughts of blind man about colors of rainbow.
His point is valid only within small range of corner cases,
but breaks for zillion small misfeatures in more wide sense.

At first, at least it require from tool deep understanding of source
code itself, but world is not limited to C/C++/Python.v And even
in this case it will not play well with macros or automatic
code generation.
At second, it requires that all moved chunk of codes will commited
at once. Otherwise, if user does selective commit (is not rare thing,
is it?) it breaks such heuristic very easily. And users will start
complaints about "bugs" that actually are not bugs but misfeature
instead.

My 2 cents, (or as we russian people says: "my 5 kopek").

[µ]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGpvU6zYr338mxwCURAp7bAJ4hj1CQ+lOv1bO0H6R1cOFW8IfV7gCeIxsD
dJl51q+aR3sZ7jKxMI/MPN8=
=jLRo
-----END PGP SIGNATURE-----