Markus,
I share your dissatisfaction with "part of" because that language construct hides many different conceptual relationships that should be cleared out, I think we'll have some community discussion work to do in that regard. One of the uses is: what is the relationship between a human and his behavior? I would say that the "human" <has been defined as having> "human behavior" (or the reverse). But if you have a better suggestion to express this concept I would be really glad to hear it.
Now that you mention it, yes, I agree that only a property called "corresponds with item" makes sense in this context, but not the inverse.
I would like to make a further distinction regarding constraints. The nature of constraints is not to set arbitrary limits but to reflect patterns that naturally appear in concepts. On that regard, I hate the word "constraint", because it means that we are placing a "straitjacket" on reality, when it is the other way round, recurring patterns in the real world make us "expect" that a value will fall within the bonds of our expectations. I think that we should seriously consider using the term "expectation" from now on because we don't "constrain" the values per se, we "expect" them to have a value, and when the value departs from the expected value, then it sets an alarm that might reflect an error or not.
Once made that distinction, yes, you are right, considering that we are separating properties and items, our expectations do not belong to the data itself, they belong to the property.
However, I would like to go to bring the conversation to a deeper level. What is that what makes the concept of "addition (Q32043)" to be that? What is in "physical object (Q223557)" that we, sentient beings, can perceive and agree to treat as a concept? I mention those two because one is purely abstract, and the other one is purely physical. And I would say that "addition (Q32043)" <has been defined as having> "associativity (Q177251)" and "physical object (Q223557)" <has been repeatedly observed to have> "density (Q29539)". We can argue whether the second is an expectation or not, but the first is definitely not, someone defined an "addition" like that and this information can be sourced. Even more, we could also say that also "physical object (Q223557)" <has been defined as having> "density (Q29539)", and I guess we could find sources for that statement too.
With all this I want to make the point that there are two sources of expectations: - from our experience seeing repetitions and patterns in the values (male/female/etc "between 10 and 50"), which belong to the property - from the agreed definition of the concept itself, which belong to the data
Cheers, Micru
PS: this is a re-post because my previous message was bounced back "for being too long" :)