*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
That's a tricky question whose answer depends on a lot of factors.
After thinking long and hard, we have decided to use a system that is
based on projects and history. For example, we would have
/home/projects/xyz
and underneath:
01-Sequences
02-DataProcessing
03-ReflectionFiles
04-xyz-023-SeMet
05-Ligand-Soaks-1
06-Ligand-Soaks-2
...
underneath, say '04-xyz-023-SeMet', we have:
01-CCP4
02-CNS
03-MR
04-Phasing-Another-Try
05-Close-To-Giving-Up
06-Wait-A-Sec-I-Got-It
...
and so on.
The numbering scheme allows us to keep track of the history of a
project, and it makes navigation very easy, because one just has to
type in '05' and hit the tab key. Overall, directory names can get
very long still, so it would be nice not to have any software
limitations there.
I think the challenge is to find a good balance between hierarchical
trees and flat file structures.
So, in a nutshell, we work in a project-oriented way (as opposed to
method-oriented), combined with a history-oriented approach (somewhat
like CVS). The project-centric scheme, of course, keeps together
nicely all files associated with a project, which is good for making
backups/archives. This scheme also allows several different people to
work on the same project, which is what we frequently do, without
messing up each others contributions and without having to look for
them all over the place. A methods-oriented lab may be better served
by a methods-centered approach.
There are probably as many solutions as there are people, plus one.
Best - MM
On Apr 21, 2006, at 4:44 AM, Eleanor Dodson wrote:
No solution from me but:
This raises another Q - is there an optimal "style" for directory
structure?
Eg I would have suggested to a grad student that they should a)
either put each project into a projectname directory and work
within that only - they can find all refinement runs by just listing
ls -lt *ref*log , or at best creating one sub directory for MR,
exptlphase, refinement, etc
Eleanor
Donnie Berkholz wrote:
Hi all,
This problem may only be affecting me, but it's quite annoying. I
tend to store my work in ever-increasing depth of subdirectories,
so I may have a PDB somewhere like research/projectname/ref3/ref4/
ref5/ref6/ref...18/filename.pdb.
The problem is that Coot insists that its main window be wide
enough to display the full path to the open file. In some cases,
that's wider than the resolution of my monitor.
Can this be fixed?
Thanks,
Donnie
<ccp4.vcf>
------------------------------------------------------------------------
--------
Mischa Machius, PhD
Associate Professor
UT Southwestern Medical Center at Dallas
5323 Harry Hines Blvd.; ND10.214A
Dallas, TX 75390-8816; U.S.A.
Tel: +1 214 645 6381
Fax: +1 214 645 6353
