RE: lime:entry

Hi all,

Indeed it is missing, as Christian observed. In this case it's just a fix on the diagram as the specs say there is (so my ok on the poll in case)

Just a caveat in case anybody is in the mood for a change (not necessarily to be discussed now, it can be on the table in case of a future revision of the model): I have some perplexity about it (I swear I didn't make it disappear from the diagram :P ) .

The rationale is: the CBD description of the Lexicon would explode.
Support is: in most vocabularies, membership properties (at least when potentially large numbers are involved) usually go from the contained to the container (so, having a property lime:inLexicon or any other name, going from the entry to the lexicon), e.g. as in SKOS, where there's a property skos:hasTopConcept for top concepts (usually a limited number) while there's no property inverse of skos:inScheme, as this would make the CBD of the scheme explode. Indeed, consumers could be interested in retrieving metadata about a scheme without being flooded by the list of all contained concepts.

Just to complete the argument: as I understand that, for lexicons, one might want to have an object representing the list of entries (e.g. an index), one possibility could be to reify such index (not supposed to contain any metadata) and make it part of the Lexicon object.

Obviously, in dedicated tools for ontolex (e.g. VocBench, or Lexvo) the lime:entry property can be accounted for; I'm more concerned about things such as standard http resolution.

Anyway, I don't have a strong opinion on that, I just thought it was worth mentioning.

Cheers,

Armando


From: Philipp Cimiano <cimiano@cit-ec.uni-bielefeld.de>
Sent: Monday, February 8, 2021 11:06 AM
To: public-ontolex@w3.org
Subject: Re: lime:entry


Yes we can update it. I can run an informal poll over the list to make sure everyone agrees.

Philipp.
Am 08.02.21 um 10:51 schrieb John McCrae:

Hi Manuel,

If I remember correctly, you created this diagram, although I could be wrong.

At any rate, we note corrections to the specification here:

https://github.com/ontolex/ontolex<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fontolex%2Fontolex&data=04%7C01%7C%7C6e55a58943214baad0d808d8cc195b5a%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637483756560184276%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=o007hCzswWsgpaEV50jWYsLpo3s2Eh%2BgO8PSZi6J9zU%3D&reserved=0>

At some point (up to Philipp?), we should update the published version.

Regards,

John
On 07/02/2021 17:13, Manuel Fiorelli wrote:
Hi Chiarcos, All

I've looked at the diagram and confirm your finding. Additionally, I noticed that the box representing the class lime:Lexicon uses

  *   ontolex:language which should be lime:language
  *   lime:linguisticModel which should be lime:linguisticCatalog
In fact, I'm unsure whether we have a process to make this kind of corrections and publish a revision of the specifications.

Regards,
Manuel

Il giorno dom 7 feb 2021 alle ore 15:25 Christian Chiarcos <christian.chiarcos@web.de<mailto:christian.chiarcos@web.de>> ha scritto:
Dear all,

looking at https://www.w3.org/2016/05/ontolex/#metadata-lime<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.w3.org%2F2016%2F05%2Fontolex%2F%23metadata-lime&data=04%7C01%7C%7C6e55a58943214baad0d808d8cc195b5a%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637483756560184276%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=Z5rfU4Iq88ypbqdl1j6jTu0EJJ9W%2BfuTsr9FQQmTsNc%3D&reserved=0>, I realized that the property lime:entry is missing in the diagram.

Best,
Christian


--
Manuel Fiorelli

--

Prof. Dr. Philipp Cimiano

Head of Semantic Computing Group

Cognitive Interaction Technology Center (CITEC)

Faculty of Technology

Bielefeld University

Phone: +49 (0) 521 106 12249