On Sun, 21 Apr 2013, Yves Dorfsman wrote:
On 2013-04-21 09:57, Brad Beyenhof wrote:
Although I'm surprised it doesn't have any comment capability.
I'm not sure exactly what you mean by this, but I definitely value the
commit-style nature of git as a version-tracking mechanism, where
rdiff-backup
(or duplicity) just copies filesystem snapshots. Version control != backup,
no
matter how incremental/differential the system.
git and friends:
capture changes when a human thinks changes were made and should be recorded.
Fewer revisions, easy to search, every revision is meaningful.
You will miss every non-planned changes.
rdiff-backup and friends:
capture ALL changes (if run frequently enough).
A lot of revisions, might be difficult to find what you are looking for.
Revision are per time slice, you have to correlate them to events yourself.
I actually like to do both.
I use git and similar (where I have gotten around to setting them up :-) and I
try to have all changes manually checked in. But in addition, I have a cron job
that runs regularly (at least nightly) to catch changes that happened that
weren't manually checked in. I give these automated checkins a different comment
so it's obvious which is which. It avoids a user checking something in that
includes a bunch of stuff that other people changed.
David Lang
_______________________________________________
Tech mailing list
Tech@lists.lopsa.org
https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
http://lopsa.org/
