Hi,
On Fri, Apr 20, 2012 at 05:07:51PM -0400, Jason Swails wrote:
> On Fri, Apr 20, 2012 at 3:36 PM, Scott Brozell <sbrozell.rci.rutgers.edu>wrote:
> > Yes this is one reason that i do not like git - i want to know what
> > other people have done Before i pull their commits.
> > git diff --stat origin/master !* |& less
> > does not do it; what i want is the analog of
> > cvs -nq update !* |& less -p '^[^?]'
>
> But git is great this way! Just do a "git fetch". This synchronizes
> *your* repository's knowledge of the remotes with what the remotes actually
> are at that time, leaving your local branches completely untouched.
>
> :) (Linus has far fewer nice things to say about CVS -- it's documented on
> Youtube :))
It is great that we are making progress on git - after just a few years
we can now add git fetch to the wiki ;-)
As far as cvs vs git, this is the same old story:
in a vacuum, git is better than cvs just as f90 is better than f77
and C++ is better than C. But git forces me to learn new syntax
and offers me no useful functionality beyond cvs; in comparison,
almost every C program is a C++ program (and likewise for f77 and f90)
so C++ does not force a C programmer to learn new syntax.
`
Of course when someone else uses a new f90 or C++ feature then the
learning curve is forced upon the rest. So good practice is to choose
new features prudently; the recent issue with f90 kind may be a lesson
(although i have seen non-amber f90 code that has used kind well).
scott
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Mon Apr 23 2012 - 12:30:02 PDT
