Actions

Ontolog Forum

Session Planning
Duration 1 hour
Date/Time December 04 2019 17:00 GMT
9:00am PST/12:00pm EST
5:00pm GMT/6:00pm CET
Convener Ravi Sharma

Ontology Summit 2020 Summit Planning

Agenda

  • The session will continue planning the main summit sessions.

Conference Call Information

  • Date: Wednesday, 04-December-2019
  • Start Time: 9:00am PST / 12:00pm EST / 6:00pm CET / 5:00pm GMT / 1700 UTC
  • Expected Call Duration: 1 hour
  • The Video Conference URL is https://zoom.us/j/689971575
    • iPhone one-tap :
      • US: +16699006833,,689971575# or +16465588665,,689971575#
    • Telephone:
      • Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 558 8665
      • Meeting ID: 689 971 575
      • International numbers available: https://zoom.us/u/Iuuiouo
  • Chat Room

Participants

Proceedings

[11:52] RaviSharma: welcome to Dec 4 session of Ontology Summit 2020.

[12:17] David Eddy: As always... I most strenuously object to EXCLUDING / IGNORING Un-natural language.

[12:18] John Sowa: I have to leave soon.

[12:18] John Sowa: But I wanted to make two points.

[12:20] John Sowa: First, the choice of ontology (or at least which microtheory of an ontology to use) must be made on the basis of the immediate context.

[12:20] Olaf Dammann: Need to leave at 12:30. Thanks.

[12:21] John Sowa: Second, the choice of microtheory must be based on some systematic method

[12:21] David Eddy: @JohnS... agree... pretty much always seems to implicit assumption that CONTEXT is fixed... when in fact it is often morphing from moment to moment as people modify their thinking & understanding.

[12:21] John Sowa: For more detail, see the article http://jfsowa.com/pubs/dynonto.pdf

[12:22] John Sowa: I have to leave now.

[12:26] David Eddy: A KG issue I have not seen yet... how is the contents of the KG kept current? Assume "contents" comes from many different sources.

[12:30] Jack Ring: Is KG a system or artifact?

[12:30] Jack Ring: Is KG a claim or an agreement?

[12:30] Jack Ring: Your knowledge or OUR knowledge? Please clarify presumed Summit beneficiaries?

[12:30] Ram D. Sriram: Theme 1: What are the fundamental issues involved in knowledge graphs? What the domains that we will need to address (we can look at all the proposals that NSF is funding right now)? See https://www.nsf.gov/od/oia/convergence-accelerator/Award%20Listings/track-a.jsp

[12:35] RaviSharma: Ram suggests a panel of speakers answering common Qs

[12:35] RaviSharma: Standards for KGs

[12:36] RaviSharma: Who and how will these be developed

[12:36] RaviSharma: Domains and use cases were discussed

[12:37] RaviSharma: interoperability

[12:39] janet singer: Possibly ensure coverage of talks that foreground one of three foci: 1) KG - what? (Setting in context, including history) 2) KG - how? (Includes technologies, standards 3) KG- why? (What value, to whom, use cases)

[12:40] Jack Ring: seems like you are intending to communicate fellow researchers rather than how practitioners can evolve KG's.

[12:42] Jack Ring: How long is a typical KG valid? www.Starkermann.com says 'until next person joins the project.'

[12:43] Jack Ring: Not how but how often

[12:45] RaviSharma: dimension i theme okn usecase different ways than usual domains

[12:45] Jack Ring: Top Quadrant Corp addresses most of this. How will your Summit differ from their material?

[12:46] RaviSharma: how are themes different from domains

[12:47] David Eddy: here's how... https://tdan.com/semantics-in-metadata-repository-and-systems-integration-efforts/6123

[12:47] RaviSharma: janet - why would be addressed more, Ram is saying using text or ?

[12:47] RaviSharma: so the sense is to select a few major themes

[12:48] David Eddy: The Zachman Framework (John Sowa part author) does as good a job as any at addressing the how/why/where/when, etc.

[12:48] Jack Ring: Domain=healthcare. Theme = building, efficacy assessment

[12:49] David Eddy: "knowledge" modifies as it moves down & up organizations

[12:49] David Eddy: offer an alternative...

[12:50] Jack Ring: the ZF is the larger impediment to the assessment of efficacy and value of a or alternative KG's.

[12:51] David Eddy: It's a start...

[12:51] David Eddy: "sense making"... good phrase

[12:52] Jack Ring: Zachman personally agreed circa 2005.

[12:52] David Eddy: But John Z never got close to actual implementation.

[12:53] David Eddy: Understand... Row 6 of ZF is 100% guaranteed machine readable.

[12:53] Jack Ring: Yes, it is a start --- down the wrong path.

[12:53] David Eddy: offer a better path.

[12:54] janet singer: Jack you can set your identity using the settings button toward the top of the screen

[12:54] David Eddy: Read the paper... DOWN will not work. WORK happens bottom-up... far, far away from management that has moved on.

[12:55] David Eddy: Unfortunately the business intent has long since been striped out of the operational systems.

[12:55] janet singer: I think its a good idea to use the Zachman-Sowa framework as a starting point for setting themes

[12:56] BobbinTeegarden: John Zachman was concentrating more on process, not structure, although he (with Sowa) were brilliant at pointing at context for meaning. Ontologies and knowledge graphs are the structure side of the process/structure dynamic.

[12:57] David Eddy: Depends what you mean by "structure"

[12:58] David Eddy: I wrote that TDAN paper as a business school discussion case... to make people THINK. Force people to THINK about what happens in organizations.

[12:58] David Eddy: I'm simply not familiar with TOGAF.

[12:59] David Eddy: Such modeling frameworks tend to be totally uninterested in the AS IS operational systems.

[12:59] Jack Ring: Jack Ring: A better path is to make 'a better path' the objective of this Summit. Janet seems to be trying. Consider alternatives such as Top Quadrant. There are others that are actually helping practitioners leverage sharing of knowledge claims.

[13:00] David Eddy: @Jack... does Top Quadrant deal with operational systems or just models

[13:01] David Eddy: I'll hail TQ... yet again...

[13:02] janet singer: Any framework that can help us ensure we span from the data-level issues to delivery of business value is helpful for setting themes

[13:02] David Eddy: If one bothers to read the TDAN document... management quickly looses interest in models (high level Zachman BSP work)... its the work that helps the 2am 3rd shift operator that has sticking power.

[13:03] BobbinTeegarden: Most of the current semantic tools vendors (TopQuadrant and Franz included) enable both modeling environments (as and to be inclusive), and are now hooking ontologies into realtime/sensornet environments.

[13:03] Jack Ring: David, Both. They make W3C almost useful. I am not involved but have known them since 1993.

[13:04] David Eddy: @Jack... I've been around CASE tools since 1982 (Excellerator)

[13:05] David Eddy: ZF, as you say... JUST a starting point. Objections, weaknesses can be noted.

[13:05] George Hurlburt: KG implies dynamics, suggesting rapid state changes. Many of the various frameworks (e.g. DoDAF) tend to bypass state changes, making the resulting output static without regard to the time domain.

[13:06] David Eddy: @Janet... thx

[13:06] David Eddy: next week

[13:06] Jack Ring: David, then you are very aware of their failure to achieve responsive, Zero-Defects Computer code.

[13:06] BobbinTeegarden: @George: look at the new SySML

[13:07] David Eddy: Not interested in weapons systems, zero defects. Plenty of bigger issues to challenges

[13:07] David Eddy: Think Westpac fined AUS$700M

[13:07] RaviSharma: Thanks for participation - Ken will be back next Wednesday.

[13:09] janet singer: For next week we can consider using ZF as sensemaking device for setting themes that go from data to value delivery, and start identifying substantive issues (like judging efficacy of KGs, keeping KGs fresh)

[13:09] RaviSharma: m and others kindly capture your good ideas and either email or send to mail list or to ken

[13:12] RaviSharma: John ZF in second and third rows goes to conceptual and physical data models and from there to incorporating subsequent outputs

Resources

Video Recording

Previous Meetings

... further results

Next Meetings

... further results