GroupI canSAS-XII

From canSAS
Revision as of 13:26, 22 September 2022 by Butler (talk | contribs) (Created page with "Outcomes from in person group discussion: * We should write a paper about resolution. This should cover what it is and its origin, in what cases ignoring it can be problematic...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Outcomes from in person group discussion:

  • We should write a paper about resolution. This should cover what it is and its origin, in what cases ignoring it can be problematic, and best practices in reporting and treating it in analysis. The target audience should be practitioners who have not yet encountered resolution issues as well as reviewers who can be pointed to this paper when they have data quality concerns that are actually due to resolution effects and are in fact being treated correctly.
    • ACTION: Tim Snow (Diamond) to start a paper on overleaf and add all interested parties.
  • Facilities should be responsible for removing ALL artifacts. This probably cannot include resolution, however the facilities should still be responsible for providing the correct math to model the resolution in a format that standard analysis packages can use.
  • There needs to be an extension to the NXcanSAS standard to allow analysis software how to interpret what a series of data entries represents: a series (in time, temp, etc); several panels of a single point in time, 4 magnetic cross sections, other?.
    • ACTION: Jeff Krzywon (NIST), Peter Beaucage (NIST), Tim Snow (Diamond) and other interested parties should draft a proposed extension, which should utilize existing structures as much as possible and then submit to the data formats group for discussion and eventual release.
  • reduced 2D data from multiple detectors contains all the information needed for fitting and is adequate for visualization. However it lacks information regarding the solid angle subtended by each pixel, which would be helpful in visualization and required for performing properly normalized 1D averages. What is the most reasonable way to provide this information? Provide a pixel shape and size for each pixel? Provide a "calibration" file with that information? Some other information? ... or is it hopeless?