| layout | title | author |
|---|---|---|
page |
ORSO - file formats - ToDo list |
Jochen Stahn |
Some of the items mentioned below are discussed in more detail at
www.reflectometry.org/file_format/specs_discussion
The orsopy version 1.0 was released by the general meeting and is thus officialy justified. But how do we handle minor changes (new key words, correction of terms) or additions from there on? Who is deciding?
Present practice for small changes is that the orsopy authors discuss via github and finally release the result. This pragmatic approach is working fine and we should keep it. But larger changes (which might interfere with other interests inside of orso, e.g. the simple model language) and deviations from formaly agreed principles (opening of the header structure to non reflectometry data) should be made by a wider body. The file formats group is not suitable for this since it is even smaller than the orsopy developers group.
The orso chairs might be the right instance in between the general meetings - but this means some preparation and discussion of (all) the chairs berfore the decission.
- Who allows minor changes to
orsopy? - Who decides on major additions or changes of the principles?
all orso chairs
2022-12-19
The .ort header is in many cases also suitable for other types of data.
Jochen thinks it would be nice if e.g. an omega scan has almost the same
header as an omega-2theta scan....
The differences would be:
- free choice of all columns (order and content)
- the first line can not be the one reserved for R(q)
- the ending can not be
.ort
Consequences:
- adaption of
orsopy
- Do we want this?
- How much standardisation outside the
.ortwill we provide?
Jochen
2022-12-19
There is a number of optional columns which will be used quite frequently - especially when used for x-ray data. Brian suggested to give them standardised names so that the analysis software can be simpler. Suggestions are:
- incident angle
- counting time per point
- attenuation factor
- wavelength
- absolute counts
- incoming beam divergence
- prepare a list of physical properties
- suggestions for names or key words
- Do we want this?
- If so: which physical properties with which names and probably symbols (short notation)?
tba.
?
Brian asked for a clarification / definition in case information is given as a column:
- Does it overwrite the header content?
- Should there be a key in the header pointing to the column?
Artur's suggestion: the analysis software should check the column description and decide what to use. Recommendation: 'column beats header', but no further rules. This approach requires standardised column descriptions. A link in the header to the column (name) helps to avoid ambiguities. There should be a check in orsopy if the column exists when link is present.
- Which approach: key, internal link or just a rule how to prioritise column and header?
- If so: definition of key / link / rule.
tba.
For non-experts it is quite difficult to figure out how to use orsopy. There should be some guide on how to start and how to add header info.
Andrew MCC
Ability to have an error defined for a quantity in the header, either implemented similar to how quantities are allowed to have a range, or similar to how columns are allowed to be an error of another column.
?
The output of lab x-ray reflectometers is be both, a raw and (partially) reduced data file. Thus there must be more information provided. An example is the cathod material, the chosen line(s) with name, energy and relative intensity, probably the monochromator....
tba.
tba.
In the meantime several corrections and additions have been made to orsopy.
- When will the next version be released?
- Who is deciding this?