I'm falling into the trap, someone help!
On Wed, 2014-02-12 at 10:04 -0800, Ross Walker wrote:
> This means beyond
> simply adding it for a phd, publishing, taking a job and then never being
> heard from again. I think if people add stuff to the code base there
> should be a reasonable expectation that they will donate their time for a
> decent length into keeping it maintained and/or helping others learn /
> maintain it.
The onus there falls on the group, not the individual IMO. Furthermore,
if a method is interesting someone else will pick it up (e.g., I picked
up CpHMD after Mongan left and expanded on it) and bring it forward. If
it's not, it'll get left behind and ultimately cleared out. If it's
essential code, _someone_ will learn it. Mongan never would have added
CpHMD if it was going to tie him into Amber support into med school
[probably] and it would have been a net loss for Amber.
But enough time has been spent on these topics at this point. There's
idealism and then realism and the happy medium is often hard to find
(and harder to agree upon).
> I stripped a bunch of stuff around it in get_nb_energy.i yesterday but the
> pmemd test cases were working at that point. Unless running
> ./test_amber_serial.sh do not test everything in pmemd. :-(
I actually think it's the recent EMIL commit. Reverting the tree to
right before those commits makes the diffs go away. The nature of the
diffs make it seem like it could be an easy fix (some missing info
messages and non-trivial differences in only the last MD step).
I sent Josh an email and hopefully he can get this squared away during
his [next] working hours.
All the best,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Wed Feb 12 2014 - 11:00:02 PST
