Giordon Stark

> * Breaking the schema to use a list of strings for the POI instead of a space-separated string of POI names seems like a good idea. The latter seems...

You can look here (`fixed_poi_fit` is only a few-lines wrapper around `pyhf.infer.mle.fit` which handles all situations): https://github.com/scikit-hep/pyhf/blob/1884c6c08bc5b629724dd9ef3c60ec4386353b94/src/pyhf/infer/mle.py#L196-L202 setting `fixed_params[index] = True` and `init_pars[index] = fixed_value` is all you need, for...

`pyhf xml2json` functionality works normally in the JSON format. It's just not something that's implemented in terms of the functionality. The serialization should be ok.

@amartyarej how quick of a timescale are you asking for? We have a ton of changes in the pipeline.

> Should we maybe have this test in addition to the old test instead of replacing it? We probably need the old test, but change the bounds to allow the...

> Is this actually necessary? There wasn't anything in particular related to the validation that required the POI to be negative. I think the concern that we don't cover qmu...

> @kratsg @lukasheinrich it has been a long time so can you help remind me of the plan? Was the idea to make a new `expected_result_1bin_lumi` fixture and model to...

> without renaming everything and creating a new spec? no, hence why the need for refactor later. #1241.

Oh, I misunderstood. The names don't need to change. Only need to copy out the spec for one of the existing ones, add "qmu" to the name instead -- and...

JSON doesn't know about tuples vs lists fyi. But in your code example, what happens with ```python spec["measurements"][0]["config"]["parameters"].append({"name": "mu", "bounds": [(0, 10)]}) ``` instead? You are appending bounds as a...