Re: satellite meeting IUCr 2005

[Favre-Nicolin Vincent <vincefn@users.sourceforge.net>]

	Hi,

   The program looks extremely interesting.

> 4)Lenght? Current proposals:
> A.Short workshop (1-2 days): Powell, Vincent, Gross-Kunstleve, Brown?
> B.School (5 days): Spek
> C.Something in between (3 days): Gualtieri, Watkin?

  I think 3 days is a minimum for a 'school'. Indeed, most international 
crystallographic congresses already begin by a one day dual workshop (ccp4 + 
ccp14), so a single or even 2 day would certainly be too short. Of course it 
will be more expensive, but otherwise it would not really be a school.

> 5)Topic? Proposals/Ideas
> A.Macromolecular computing: Powell, Vincent
> B.Macro- and small molecules computing with special attention to the powder
> method: Lachlan, Gualtieri, Viterbo
> C.Introducing CIF (and XML systems): Brown
> D.3 streams: macromolecules, small molecules/powders, general interest (the
> first two in parallel sessions, the last in plenaries at the beginning of
> each day): Spek
> E.Crystallographic computing (general): Software to software interfaces,
> software libraries, program validation: Watkin

   5D (or B) look good to me too.

   However I think it would also be interesting to have a half day at the 
beginning to talk about the foundations of modern crystallographic 
algorithms, before focusing on the applications. I believe Maximum Likelihood 
(ML) would be worth spending some time on it, as it is now more and more used 
(phasing, but also refinement on intensities, and now powder diffraction), 
and most people are only familiar with traditionnal least squares.
   This could fit in 5D, by just pushing back the separation in parallel 
sessions by (half) a day.

  	Regards,
		Vincent
-- 
Vincent Favre-Nicolin
Université Joseph Fourier
http://v.favrenicolin.free.fr
ObjCryst & Fox : http://objcryst.sourceforge.net