Actions

ConferenceCall 2023 03 15: Difference between revisions

Ontolog Forum

Line 78: Line 78:


== Discussion ==
== Discussion ==
== Announcements ==
[12:05] RaviSharma: welcome everyone
[12:10] Gary Berg-Cross: Note next week we will have a special session from the OBO Academy on WikiData
Tuesday, March 21, 2023 from 11:00am to 12:00pm with Speakers Andra Waagmeester and Tiago Lubiana Link https://lbnl.zoom.us/j/98450763854?pwd=N2lhajNMNER0ekZVaFlyWFM2TXUwZz09
== Main Discussion ==
[12:13] RaviSharma: Karl, generally the technology also gets tied to a window of opportunity also where it is allowed to mature, scale and respond to market needs for that technology, generally for solution.
[12:14] RaviSharma: Karl I was mentioning Gartner graph and eventually after mature stay technology oes retire!
[12:22] RaviSharma: Karl, It is always a challenge to have unity of purpose between the user and the developer. But what else would you use other than triples, for ontologies?
[12:27] RaviSharma: Karl, what alternatives beyond RDF are better for onto engg?
[12:28] Michael DeBellis: Are there any plans for RDF*
[12:28] RaviSharma: Karl in tooling list where are the use cases requirements etc.
[12:33] RaviSharma: Karl I am also supporter of visualization and hope you are suggesting that OWL etc need not be as explicit to users as KGs!
[12:35] Gary Berg-Cross: Digrams...not OWL...makes the point that it has to be human understandable.
[12:35] RaviSharma: Karl, what you are saying is that Patters and graphs should be the direction of reuse?
[12:38] Chris_Kindermann: But aren't diagrams also ambiguous? Just looking at a diagram doesn't necessarily provide me with enough information to know what exactly is meant by the diagram (coming back to the first example: given an edge between two entities: What does the edge mean?)
[12:40] RaviSharma: Karl, Excellent Digital twin Language, on right is high level or global or pattern and on left I interpret is the instantiation?
[12:45] RaviSharma: Karl, with integration of ChatGPT how do you see this progress, namely integration between Digital twin language and the URL access based ChatGPT?
[12:49] Chris Kindermann: Thanks, Karl. I agree that's a good way of looking at things.
[12:54] anonymous1: I think diagrams should not be the end-all be-all of describing the relationship. I think they're useful indicating that a relationship /exists/ rather than the exact meaning.
[12:54] anonymous1: You're right that a single edge could be more than one meaning -- for example, in the MOMo methodology, we use 17 different axioms represent a single node-edge-node construction.
[12:54] MARIA POVEDA VILLALON: We need a balance between details in the diagrams and the OWL code, but some converters exist. But the close you get to OWL the more difficult to read the diagrams are. Id love to have an standardised UML_OWL
[12:56] anonymous1: @mariapoveda -- take a look at ontouml. is this what you're looking for?
[12:56] anonymous1: you might also look at owl to sysml translations (I recall that sysml is a superset of uml)
[12:58] MariaPoveda: I know ontouml but not sure if it is what we need
[12:59] MariaPoveda: I'm working on something closer to OWL https://chowlk.linkeddata.es/
[12:59] MariaPoveda: do you have the link to sysml?
[12:59] James Overton: In hindsight, I feel like the OBO tools that we showed in the first five weeks are not the tools and techniques that connect most closely to what Cogan and Karl have been speaking about. We presented mainly on tools that help us build and integrate open community ontologies at scale, which is does not seem to be something that Cogan and Karl need.
[13:00] Cogan: I think research into fuzzyowl/dl has been quite slow. You might look into more recent stream reasoning techniques for changing knowledge/data.
== Knowledge Graphs ==
[12:29] Ken Baclawski: Karl, you might look at my work on KGSQL that goes well beyond RDF Star.  In KGSQL every triple is reified.  See https://ontologforum.com/index.php/KGSQL
[12:55] James Overton: I think that the most interesting thing to discuss with Ken would be LinkML <https://linkml.io/>, which OBO people such as Chris Mungall are developing. Unfortunately we do not have enough time to get to the interesting bits today.


== Resources ==
== Resources ==

Revision as of 16:22, 18 April 2023

Session Design Patterns
Duration 1 hour
Date/Time 15 Mar 2023 16:00 GMT
9:00am PDT/12:00pm EDT
4:00pm GMT/5:00pm CST
Convener Gary Berg-Cross

Ontology Summit 2023 Design Patterns

Helping scientific researchers make better use of ontologies

Agenda

  • Karl Hammar Applied Ontology Engineering: The Missing Pieces Slides
  • Karl Hammar is a senior technical program manager at Microsoft where he works on building industry-specific cloud solutions. Previously he headed the Department of Computing at Jönköping University, Sweden. Dr Hammar's research has focused on methods and tools supporting ontology engineering in scenarios ranging from building automation to veterinary syndromic surveillance. He obtained his PhD from Linköping University. His most recent and notable research and development work targeted the Smart Buildings domain where he among other things co-founded the RealEstateCore ontology and consortium -- a leading edge technology in the commercial real estate vertical, and the recommended solution for use with the Azure Digital Twins platform.
  • Video Recording

Conference Call Information

  • Date: Wednesday, 15 Mar 2023
  • Start Time: 9:00am PDT / 12:00pm EDT / 5:00pm CET / 4:00pm GMT / 1600 UTC
    • ref: World Clock
    • Note: The US and Canada are on Daylight Saving Time while Europe has not yet changed.
  • Expected Call Duration: 1 hour
  • Video Conference URL
    • Conference ID: 837 8041 8377
    • Passcode: 323309
  • Chat Room

The unabbreviated URLs are:

Attendees

Discussion

Announcements

[12:05] RaviSharma: welcome everyone

[12:10] Gary Berg-Cross: Note next week we will have a special session from the OBO Academy on WikiData

Tuesday, March 21, 2023 from 11:00am to 12:00pm with Speakers Andra Waagmeester and Tiago Lubiana Link https://lbnl.zoom.us/j/98450763854?pwd=N2lhajNMNER0ekZVaFlyWFM2TXUwZz09

Main Discussion

[12:13] RaviSharma: Karl, generally the technology also gets tied to a window of opportunity also where it is allowed to mature, scale and respond to market needs for that technology, generally for solution.

[12:14] RaviSharma: Karl I was mentioning Gartner graph and eventually after mature stay technology oes retire!

[12:22] RaviSharma: Karl, It is always a challenge to have unity of purpose between the user and the developer. But what else would you use other than triples, for ontologies?

[12:27] RaviSharma: Karl, what alternatives beyond RDF are better for onto engg?

[12:28] Michael DeBellis: Are there any plans for RDF*

[12:28] RaviSharma: Karl in tooling list where are the use cases requirements etc.

[12:33] RaviSharma: Karl I am also supporter of visualization and hope you are suggesting that OWL etc need not be as explicit to users as KGs!

[12:35] Gary Berg-Cross: Digrams...not OWL...makes the point that it has to be human understandable.

[12:35] RaviSharma: Karl, what you are saying is that Patters and graphs should be the direction of reuse?

[12:38] Chris_Kindermann: But aren't diagrams also ambiguous? Just looking at a diagram doesn't necessarily provide me with enough information to know what exactly is meant by the diagram (coming back to the first example: given an edge between two entities: What does the edge mean?)

[12:40] RaviSharma: Karl, Excellent Digital twin Language, on right is high level or global or pattern and on left I interpret is the instantiation?

[12:45] RaviSharma: Karl, with integration of ChatGPT how do you see this progress, namely integration between Digital twin language and the URL access based ChatGPT?

[12:49] Chris Kindermann: Thanks, Karl. I agree that's a good way of looking at things.

[12:54] anonymous1: I think diagrams should not be the end-all be-all of describing the relationship. I think they're useful indicating that a relationship /exists/ rather than the exact meaning.

[12:54] anonymous1: You're right that a single edge could be more than one meaning -- for example, in the MOMo methodology, we use 17 different axioms represent a single node-edge-node construction.

[12:54] MARIA POVEDA VILLALON: We need a balance between details in the diagrams and the OWL code, but some converters exist. But the close you get to OWL the more difficult to read the diagrams are. Id love to have an standardised UML_OWL

[12:56] anonymous1: @mariapoveda -- take a look at ontouml. is this what you're looking for?

[12:56] anonymous1: you might also look at owl to sysml translations (I recall that sysml is a superset of uml)

[12:58] MariaPoveda: I know ontouml but not sure if it is what we need

[12:59] MariaPoveda: I'm working on something closer to OWL https://chowlk.linkeddata.es/

[12:59] MariaPoveda: do you have the link to sysml?

[12:59] James Overton: In hindsight, I feel like the OBO tools that we showed in the first five weeks are not the tools and techniques that connect most closely to what Cogan and Karl have been speaking about. We presented mainly on tools that help us build and integrate open community ontologies at scale, which is does not seem to be something that Cogan and Karl need.

[13:00] Cogan: I think research into fuzzyowl/dl has been quite slow. You might look into more recent stream reasoning techniques for changing knowledge/data.


Knowledge Graphs

[12:29] Ken Baclawski: Karl, you might look at my work on KGSQL that goes well beyond RDF Star. In KGSQL every triple is reified. See https://ontologforum.com/index.php/KGSQL

[12:55] James Overton: I think that the most interesting thing to discuss with Ken would be LinkML <https://linkml.io/>, which OBO people such as Chris Mungall are developing. Unfortunately we do not have enough time to get to the interesting bits today.

Resources

Previous Meetings

 Session
ConferenceCall 2023 03 08Modules and Patterns
ConferenceCall 2023 03 01Panel
ConferenceCall 2023 02 22Ubergraph
... further results

Next Meetings

 Session
ConferenceCall 2023 03 21Wikidata
ConferenceCall 2023 03 22Wikidata
ConferenceCall 2023 03 29Synthesis
... further results