search

OBOFoundry / OBOFoundry.github.io

Metadata and website for the Open Bio Ontologies Foundry Ontology Registry
http://obofoundry.org
Other
166 stars 204 forks source link

Add recommendation for textual definitions of relations #846

Closed jamesaoverton closed 2 years ago

jamesaoverton commented 5 years ago

Principle 6 "Textual Definitions" says "Textual definitions SHOULD follow Aristotelian form (e.g. “a B that Cs” where B is the parent and C is the differentia), where this is practical." This applies best to class definitions. I would like to expand this recommendation to apply to relations: "a B between D and R, in which C", where D is the domain, R is the range, and C is the differentia.

I've used this pattern in RO Core, and I think it works well. For example RO:0000053 'has bearer' is defined as "A relation between an independent continuant (the bearer) and a specifically dependent continuant (the dependent), in which the dependent specifically depends on the bearer for its existence". When fixing some things on today's OBI call we came across some problems for which this would be a good general solution.

So I propose expanding that sentence to "Textual definitions SHOULD follow Aristotelian form where this is practical: e.g. “a B that Cs” where B is the parent and C is the differentia; or in the case of relations “a B between D and R, in which C” where D is the domain and R is the range."

cmungall commented 5 years ago

I think we should give this a go, but I recommend first testing on a broad section of RO before instituting as a principle. I don't see any reason why it wouldn't improve things. The definitions would be wordier as we'd use terms like "the anatomical structure" rather than variables such as "S", but I think most users would be happy with this.

On Mon, Mar 18, 2019 at 9:48 AM James A. Overton notifications@github.com wrote:

Principle 6 "Textual Definitions" says "Textual definitions SHOULD follow Aristotelian form (e.g. “a B that Cs” where B is the parent and C is the differentia), where this is practical." This applies best to class definitions. I would like to expand this recommendation to apply to relations: "a B between D and R, in which C", where D is the domain, R is the range, and C is the differentia.

I've used this pattern in RO Core, and I think it works well. For example RO:0000053 'has bearer' is defined as "A relation between an independent continuant (the bearer) and a specifically dependent continuant (the dependent), in which the dependent specifically depends on the bearer for its existence". When fixing some things on today's OBI call we came across some problems for which this would be a good general solution.

So I propose expanding that sentence to "Textual definitions SHOULD follow Aristotelian form where this is practical: e.g. “a B that Cs” where B is the parent and C is the differentia; or in the case of relations “a B bewteen D and R, in which C” where D is the domain and R is the range."

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/OBOFoundry/OBOFoundry.github.io/issues/846, or mute the thread https://github.com/notifications/unsubscribe-auth/AADGOUfigQ6bjkJJlk0583ulw7H0cWQxks5vX8M5gaJpZM4b6R5I .

cmungall commented 5 years ago

What should the style be when D=R?

Perhaps "A relation between two Xs where the first X [optional qualifier describing role] <...> the second X [optional qualifier describing role]"

On Mon, Mar 18, 2019 at 10:39 AM Chris Mungall cjmungall@lbl.gov wrote:

I think we should give this a go, but I recommend first testing on a broad section of RO before instituting as a principle. I don't see any reason why it wouldn't improve things. The definitions would be wordier as we'd use terms like "the anatomical structure" rather than variables such as "S", but I think most users would be happy with this.

On Mon, Mar 18, 2019 at 9:48 AM James A. Overton notifications@github.com wrote:

Principle 6 "Textual Definitions" says "Textual definitions SHOULD follow Aristotelian form (e.g. “a B that Cs” where B is the parent and C is the differentia), where this is practical." This applies best to class definitions. I would like to expand this recommendation to apply to relations: "a B between D and R, in which C", where D is the domain, R is the range, and C is the differentia.

I've used this pattern in RO Core, and I think it works well. For example RO:0000053 'has bearer' is defined as "A relation between an independent continuant (the bearer) and a specifically dependent continuant (the dependent), in which the dependent specifically depends on the bearer for its existence". When fixing some things on today's OBI call we came across some problems for which this would be a good general solution.

So I propose expanding that sentence to "Textual definitions SHOULD follow Aristotelian form where this is practical: e.g. “a B that Cs” where B is the parent and C is the differentia; or in the case of relations “a B bewteen D and R, in which C” where D is the domain and R is the range."

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/OBOFoundry/OBOFoundry.github.io/issues/846, or mute the thread https://github.com/notifications/unsubscribe-auth/AADGOUfigQ6bjkJJlk0583ulw7H0cWQxks5vX8M5gaJpZM4b6R5I .

jamesaoverton commented 5 years ago

While textual definitions are required by Principle 6, the wording makes it clear that Aristotelian definitions are just recommended. I'm proposing this as a recommendation with the same (weak) force, more as a starting point than anything else. I have no desire to revise existing definitions en masse. If you can think of a way to make that intention more clear, please let me know.

I think your D=R case would also be a good addition, although if we have more than three examples we might want to restructure the text with bullet points or something.

nlharris commented 2 years ago

What's the status of this?

cmungall commented 2 years ago

I suggest we consolidate here: https://github.com/oborel/obo-relations/issues/523

And from an OBO side we just say "for relations follow the patterns decided in RO" and link to the documentation page that we will add to close the RO ticket

nlharris commented 2 years ago

So I can close this issue?

wdduncan commented 2 years ago

IIRC, there was push back on using Aristotelian definitions on an RO call a while back. The person (sorry I can't recall her name) said that the curators couldn't understand the Aristotelian definitions. FWIW, I support adding the recommendation.

cmungall commented 2 years ago

This issue can be closed by a PR to principle 6 that clarifies the status of the principle for non-classes.

I propose a slight variation to what James wrote:

"Textual for classes, definitions SHOULD follow Aristotelian form where this is practical: e.g. “a B that Cs” where B is the parent and C is the differentia. In the case of relations (object properties), we recommend following the patterns established in the Relation Ontology. In the case of annotation properties, we recommend following the patterns established in OMO."

cmungall commented 2 years ago

Or we could consider merging this issue into #956, but it has been open a while

jamesaoverton commented 2 years ago

I agree that #956 addresses this discussion, so I'll close this issue.