When @pkienzle converted to the new jitter-view orientation representation he dropped python support, judging that it would be too slow to be useful, and so not worth the extra hassle to maintain it in both python and C (release 0.97). Similarly, he never implemented magnetism for pure python models.

That Users can no longer produce pure python oriented/magnetism plugins needs to be clarified in the documentation (plugin.rst).

The Structure factors along with a number of other more complex models do not list the math used to compute them. Every model should have that which should of course match what the code is doing.

These omissions can at times be irritating when trying to answer a question about the model, often with the user sitting next to me and explaining how they just need to go to the documentation. Usually the documentation does refer to the underlying articles with equations but that can be a huge barrier.

One caveat may be that some models can be quite complex and the math itself may look daunting if not placed judiciously in the docs?

The corfunc_help.rst committed to fix #1225 needs copying to ​https://github.com/SasView/docs for the new online documentation (see Ricardo's email to developers of 11/02/19).