Need to be able to specify conditions for TransferValue

[philbarker]

From Credit Transfer Agreement Research Briefing, slide 26 from meeting 2:

Currently there is no way to specify these conditions on a TransferValueProfile (they can be specified elsewhere, though)

Adding TVP to the domain of requires and commonConditions would do the trick:

Notes

• this only shows the "dumbest" approach of expressing the conditions in text: ConditionProfile supports direct machine-readable references to requirements like minimumAge, place of residence, experience etc.
• see also issue Negative requirements and exceptions on Conditions #1003 on Negative requirements for exceptions.

[siuc-nate]

Slight nitpick with the diagram and the usage of alternativeCondition: Given its domain/range/definition, the structure would be more like this:

This is mostly because of the "and from which any one path fulfills the parent condition" part of the definition for that property, which (at the time we wrote that) was intended to avoid the need for something more complex like we see in pathways with requiredNumber.

Overall though, I think this makes sense - but I would also suggest considering what the pros and cons might be of treating Transfer Value Profile more like a component in a pathway, given both the role of transfer in pathways generally, and TVPs already being a sort of "additional layer of assertions by a third party about what's needed to complete the underlying resources" which is, at a high level, a fairly pathway component-y behavior. I think exploring that line of thinking further would be interesting, if nothing else. And, it might end up making more sense to use constraints (whether via condition profiles or directly on TVPs or via something akin to ComponentCondition) to cover the complex requirements and/or the "negative" requirements, too.

For what it's worth, I had done some musing on this topic a little while back in #947, but it didn't really go anywhere. I don't have a strong opinion of it right now, but I bring it up in case something might spark an idea.

[philbarker]

@siuc-nate hmm, yes, I see what you mean by your "slight nitpick", good catch. Thank you.

I will look at treating TVPs like a component in a pathway, in fact we were discussing something kind of similar yesterday for a different use case, but I think we want to avoid the more complex pathway expressions (unless, that is, we wind up with equally complex condition expressions doing the same thing).