[essdiffraction, DREAM, BEER] Comments on Sciline workflow

[celinedurniak]

Comments shared with @nvaytet and @jokasimr
Posted here as records

• it would be useful to have a more compact display of the workflow, which could be interactive and display additional info by clicking on parts of the workflow (like details about d-spacing, docstrings about the method, coordinates of the output, link to the documentation...)
• it would also be useful to know which settings are required or optional.
• are you planning to have "forks" in the workflows? For example by giving the option to convert to d-spacing or Q, saving as xye or CIF, stripping Vanadium peaks or not...

[nvaytet]

1. I agree it would be nice, but right now I see this kind of work as low-priority. We have an awful lot of reduction workflows to get working and this nice-to-have would probably come later down the line
2. Not sure where you would like to see required and optional settings? The .visualize() method already shows you this with red boxes (all black boxes mean they already have a default value). But I guess you meant somewhere else than in the output of visualize?
3. Different kinds of forks would be implemented in different ways. For example in the SANS workflow you can choose to compute a final result with either Q or QxQy as coordinates depending on which target you request (wf.compute(IofQ) or wf.compute(IofQxQy)). This would be the same probably for "d-spacing or Q", and "saving as xye or CIF". For stripping peaks or not we have different options which would be easiest to discuss in person. No best solution fits all has been found here.