1. I am not saying our global optimisation methods are
better than direct methods. In fact, read
Shankland et al, J Mat Chem 1997,7(3), 569-572
for a discussion of optimal data collection wrt
direct methods and organic powder data. The
potential is enormous.
2. When our software is released is our business. It may
be released when it is bug free and user friendly.
Are we supposed not to publish until that point ?
It's the ideas that are important, not the software.
If someone else cares to implement them, then good luck
to them.
3. Do you really think I've never indexed a powder / determined
a space group ? I know full well that these are often the
rate limiting steps, and that they often fail or are
ambiguous. That does not mean that when I run a global
optimisation structure solution, that I'm prepared to hang
around for hours or days when I can have the answer in seconds.
3. I'm not interested in comparisons with direct methods
at the moment. The global optimisation method is of
interest in its own right. We've got enough on our
plate without entering into competitions ! Neither
am I handing out hard earned data sets until I am
finished with them. But I will say that we are routinely
solving flexible *molecular* structures with more than 25
non-hydrogen atoms with 2A resolution data so you should not
consider that my unwillingness to participate in a challenge
is in any way indicative of a lack of confidence in our approach.
4. As for the molecule with > 1000/s, you can read all
about it, and others, when we publish and not before.
Apart from the Z. Krist. paper, we have currently two
papers in press and a further one submitted. I will e-mail
you the references when they come out.
For the last time*, the point is that if you are going to
use a *global optimisation method* to solve a powder structure,
you are better off using integrated intensities + a covariance
matrix. That's our experience, in theory & in practice.
All we're trying to do is advance the field : if people choose
to ignore the work, or challenge it, that's fine by me.
These opinions are mine, they may be Bill's too : I hope so.
I just can't find him at the moment !
And on that note....
Best wishes,
Kenneth
-- Dr. Kenneth Shankland,Building R3,ISIS Facility,Rutherford Appleton LaboratoryChilton,Didcot,Oxon OX11 0QX,United KingdomTel : +44 1235 44 6381 Fax : +44 1235 44 5720Email : k.shankland@rl.ac.uk