Clarifying FP7: RO re-use

matentzn commented 3 years ago

We want to clarify what is meant by "there SHOULD still be coordination with RO, for example in the form of an issue submitted to the RO tracker." in http://obofoundry.org/principles/fp-007-relations.html

From the Sep 7 OFOC call: Add something like "Each OBO ontology MUST reuse existing relations (aka object properties) that have already been declared in the Relations Ontology (RO), rather than declaring duplicative relations. In some cases it may make sense for an ontology to declare a new relation in its own ID space. In these cases, the new relation MUST be classified under an existing RO relation and there SHOULD still be coordination with RO, for example in the form of an issue submitted to the RO tracker.

cthoyt commented 3 years ago

does this apply both to relations to other named entities and literals?

It always irked me that there aren't standard properties for InChi and SMILES, and several of the other relations/properties in ChEBI as well

matentzn commented 3 years ago

Good question.. not sure. My primary concern is object properties, but there really is no rational reason behind it.

nataled commented 2 years ago

Text added. More suggestions are in issue #1680. I think this issue can be closed (a new issue should be created for @cthoyt's concern, if he so desires).

cthoyt commented 2 years ago

Thanks for the note @nataled, I just made a new issue for further discussion.

alanruttenberg commented 2 years ago

Hadn't noticed this language. The MUST is too strong. There's no guarantee that there is an appropriate superproperty unless RO includes topObjectProperty (which it can't). I think this needs to be a SHOULD, at best, with a comment addressing this.

nataled commented 2 years ago

@alanruttenberg would it help to add "if one exists" to the superproperty requirement and keep the MUST? I wasn't at the OFOC meeting where this was discussed so I'm relying on the notes captured above.

alanruttenberg commented 2 years ago

Something like that, yes. I reviewed the language and would propose the following:

Each OBO ontology MUST reuse existing relations (aka object and data properties) that have already been declared in the Relations Ontology (RO), rather than declaring relations that mean the same as an existing RO relation. Where it makes sense for an ontology to declare a new relation in its own ID space and there is a RO relation that is logically a super-property of the new relation, the new relation MUST be asserted to be a sub-property of the RO relation. In such cases, it is requested that there still be coordination with RO, for example in the form of an issue submitted to the RO tracker.

Rationale for changes:

I assumed that the scope of RO includes data properties so added that in the parenthetical. If I've got that wrong, omit that.
"duplicative" is jargon, so I rephrased to use more plain language.
The use of the term "classify" is something that is applied to classes. I rewrote that bit to make it clear exactly what should be done.
I felt the "in some cases" language was unnecessary and a bit off-putting.
I objected to the MUST for coordination with RO, based on my past experiences in trying to do so, so I reduced that to a polite request.

I wasn't sure whether the comma in "with RO, for example" should be moved to the end "with RO for example,". Please check.

alanruttenberg commented 2 years ago

BTW, the page has empty sections "Examples, and Counter-Examples". If there is nothing to say in a section then I think the section should be omitted.

cmungall commented 2 years ago

My proposed standard https://github.com/chemkg/chemrof

e.g https://w3id.org/chemrof/smiles_string

On Wed, Sep 8, 2021 at 3:29 AM Charles Tapley Hoyt @.***> wrote:

does this apply both to relations to other named entities and literals?

It always irked me that there aren't standard properties for InChi and SMILES, and several of the other relations/properties in ChEBI as well

— 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/1587#issuecomment-915116606, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAMMOJRDWEPIJVDS2MXV7TUA43G7ANCNFSM5DUMO5RA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

nataled commented 2 years ago

@cthoyt @cmungall please open a new issue regarding the lack of standard properties for InChi and SMILES to ensure that there's the proper follow-up.

OBOFoundry / OBOFoundry.github.io

Clarifying FP7: RO re-use #1587