Matthew Newville,
Consortium for Advanced Radiation Sources, University of Chicago,
Chicago, IL
Using FEFF for XAFS analysis is a marvelous model of
using theoretical physics to solve everday problems in a wide variety of
scientific fields. The XAFS community is
well-served by FEFF because
- FEFF is not too difficult for a typical scientist
doing XAFS to use for simple XAFS problems.
- It's easy enough for the experts to write analysis codes aimed at a
more general audience that can do fairly sophisticated XAFS modeling.
There are dozens of such programs based on FEFF.
- It's not too difficult for a knowledgeable person to modify
for difficult XAFS problems or other spectroscopies (such as
DAFS or XPS).
These aspects of FEFF's usability (easy for an expert,
but definitely not a black-box) allow the theory to grow and be applied in
new ways. This might be the biggest reason for FEFF's
success.
A Photo-Electron Theory Center should view itself as a User
Facility, with a focus of providing theoretical codes to the
knowledgeable analyst of all sorts of spectroscopies (XAFS, DAFS, XPS,
XMCD, resonant-Raman scattering, inelastic scattering, ...).
I'd really like a Photo-Electron Analysis ToolKit that:
- provides ``under-the-hood'' access to the codes --
overlapped atomic orbitals, scattering path geometry, scattering matrix
elements, sum-over-scattering-paths, Debye-Waller factors, ....
- uses a consistent and common input/output mechanism, so that
different parts of the codes can be pieced together in new ways.
- is aimed at the serious analyst, or writer of analysis programs.
If
you do the physics, we'll write the GUIs.
Download a PostScript version of APS-slides.ps,
the 6 overhead slides presented at the workshop.