Re: [sdpd] GUI Interfaces - was Re: Ask for reports on ECM19

[L.M.D.Cranswick...@dl.ac.uk]

> --- In sdpd...@egroups.com, L.M.D.Cranswick...@d... wrote:
> 
> > Given the reputation of EXPO 
> > of being very friendly and easy to use - it could be quite a 
> > rival to ESPOIR given the setup via Prespoir can be 
> > the most tedious part to get going quickly?  (at 
> > least when trying to quickly demonstrate the ESPOIR 
> > software in showing how to get a new problem working 
> > from scratch(?))

[text deleted]

Armel wrote:
> With experience, the time for building
> a datafile considerably reduces, possibly to less than
> 1 mn. And this time is not very dependent on the presence
> or not of an integrated GUI in the program. A modern text
> editor is the most marvelous GUI which I know, if compared
> to antediluvian text editors (VI and so on).

Have wanted to reply to Armel's comments before but delays 
abound.

Isn't the importance of having GUI's on "real space" 
structure solution programs that the user can be confident 
the program is running as ordered - without having to
wait days or weeks and check if a Monte Carlo
program like Espoir failed due to small typographical
error in the input file.   Spending days, weeks or months
sorting out input file problems may not allow
real experience to be generated in solving structures
from real space/Monte Carlo methods (Perhaps they
will be for-ever wondering if a structure failed 
to solve - not because it was impossible for the software - 
but some non-obvious option in the input file)

A good Graphical User Interface can greatly assist in ensuring 
the starting file is what the user intended.  (the most
error prone option most likely being the insertion of
fragments into the solving routines?).

> So, copy an old file which you remember corresponding
> to a similar case. Then change only the parameters that
> have to be changed. 

Doesn't this assume these files are correct and transferable
to a new problem - especially in the case of inserting
fragments where something new will have to be generated?

Lachlan.

PS: Another self citation alert:
Some functionality in the Crystals for Windows single
crystal suite may not be widely known but could be
useful for some of the above mentioned, model building and post 
structure solution refinement: DLS - Distance Least Squares.
Has very good graphics to help the user see what is happening.

  Setting up Crystals to perform DLS on a simple Inorganic Structure 
   http://www.ccp14.ac.uk/tutorial/crystals/dls.html
(will try and do a more complicated one as an example
in the future).

  Setting up Crystals to perform DLS on an Organic Structure
   http://www.ccp14.ac.uk/tutorial/crystals/dls_organic.html

  Setting up Planar Restraints in Crystals 
     (and doing DLS on these as well to optimise the structure)
   http://www.ccp14.ac.uk/tutorial/crystals/planar_restraint.html

  Setting up and Refining on a Rigid Body (example of a 5
   membered ring) in Crystals
   http://www.ccp14.ac.uk/tutorial/crystals/rigid_body.html


  

-- 
Lachlan M. D. Cranswick

Collaborative Computational Project No 14 (CCP14)
    for Single Crystal and Powder Diffraction
Daresbury Laboratory, Warrington, WA4 4AD U.K
Tel: +44-1925-603703  Fax: +44-1925-603124
E-mail: l.cranswick...@dl.ac.uk  Ext: 3703  Room C14
                           http://www.ccp14.ac.uk


-------------------------- eGroups Sponsor -------------------------~-~>
GET A NEXTCARD VISA, in 30 seconds!  Get rates 
as low as 0.0% Intro or 9.99% Ongoing APR and no annual fee!
Apply NOW!
http://click.egroups.com/1/9331/14/_/40740/_/970553007/
---------------------------------------------------------------------_->