[Nexus-developers] Workshop organization

["V. Armando Solé"]

Dear Mark,

This workshop does not exclude the one on NeXus. Each day is more and 
more evident to me that  institutes using NeXus are "cooking" their 
particular solution and a dedicated workshop for NeXus users and 
potential users is necessary. Concerning dates on the NeXus workshop, at 
the PANDATA meeting we said this year. To me, that is still valid 
because as far as I understand there may be urgency (Gerd at DESY?) unless 
you want to have the feedback of the more data analysis oriented workshop.

This post of mine to the MAHID mailing list should serve to illustrate 
in what direction the discussions go::

>> If one has followed the full NeXus convention, and wants to perform 
>> azimuthal averaging of an image obtained by powder diffraction, one 
>> would need a group where image, sample_detector_distance, 
>> pixel_size_dim0, pixel_size_dim1, direct_beam_position, detector_tilt, 
>> detector_rotation, wavelength, and perhaps something else is written. If 
>> that "image_powder_diffraction_group" is available, everybody has agreed 
>> on the names and so on, the analysis is possible irrespectively of the 
>> convention used to store the data (NeXus in this example). If we could 
>> get a consensus about the minimal set of information to perform a 
>> particular analysis, FROM THE RAW DATA, of image powder diffraction 
>> analysis and/or image SAXS (quite similar problems) and/or XRF mapping 
>> and/or XANES mapping, and so on we would have taken a huge step forward.
>>
>> The /data/data approach was intended to share simple data. I think for 
>> pure statistical analysis problems can take us quite far in the mean time.
>>

Please do not be confused, that post is for illustration. Most of the parameters (if not all) I mention ARE already in the NeXus format. What we are trying to do is to make them accessible to people AND PROGRAMS that have no clue about NeXus but that are able to read HDF5.

Armando