Closed cmungall closed 5 years ago
Note that punning OP and class is a use case for CRO: https://github.com/data2health/contributor-role-ontology/issues/14
Uhm, unfortunately it's also an OWL 2 DL violation :-(
I thought punning on class and object property was permitted? https://www.w3.org/TR/owl2-new-features/#F12:_Punning
I have plenty of comments that I could address to the people who wrote the specs. Most of them are rude and therefore I shall keep quiet.
What's my best bet for actually seeing what is permitted? Is the one I linked to the best or is there others?
I'm not aware of any other source.
What's my best bet for actually seeing what is permitted? Is the one I linked to the best or is there others?
That's it @cmungall. From my understanding, it is permitted. Property punning between properties is illegal, but punning of a property name and class name or individual name is allowed.... as far as I can tell.
p.s. for years (!) I thought it was illegal
p.s. for years (!) I thought it was illegal
Me too.
Aside from punning, there's a problem in the example because Equivalentclasses expects at least two arguments. I assume that's an issue in the example, not in CRO?
Let me know if you need a release of this fix urgently.
as for me, I'd like to have have 5.1.11 released soon
Cool, I'll release once we've reached a decision on undeclared obo tags.
Aside from punning, there's a problem in the example because Equivalentclasses expects at least two arguments. I assume that's an issue in the example, not in CRO?
The example is manchester / frame syntax though...?
Anyway thanks for the fix! Don't need a release urgently. But I don't want to be the holdup on the release, I'll answer in the other ticket about obo tages
The example is manchester / frame syntax though...?
I wasn't referring to the syntax, it's the OWL axiom that needs two or more. Syntax might allow writing it (functional syntax complains, because it's closer to the spec in its grammar) but it's still a bad axiom. Not important here though.
Ah, I see, I used EquivalentClasses
which is not a construct in Manchester.
Class: Test
EquivalentClasses: creator_of value x
I meant EquivalentTo
which is Manchester
Note that there was a similar issue fixed in #548 but there wasn't a test case, so this seems to be a new case.
The following ontology puns
creator_of
, but it is impossible to use the punned OP in a HasValue expression:yields