From OntologPSMW

Jump to: navigation, search
[ ]


Open Ontology Repository initiative - Planning Meeting - Thu 2008-01-03     (1)

We are using this conference call to explore and possibly plan on kicking-off an Open Ontology Repository initiative, which a few of the folks in the community feels that a collaborative effort on such would be a meaningful thing to move forward with --Conveners: Peter P. Yim, Leo Obrst & MikeDean     (1A)

  • Shared-screen support (VNC session), if applicable, will be started 5 minutes before the call at:     (1C6)
    • view-only password: "ontolog"     (1C6A)
    • if you plan to be logging into this shared-screen option (which the speaker may be navigating), and you are not familiar with the process, please try to call in 5 minutes before the start of the session so that we can work out the connection logistics. Help on this will generally not be available once the presentation starts.     (1C6B)
    • people behind corporate firewalls may have difficulty accessing this. If that is the case, please download the slides below and running them locally. The speaker will prompt you to advance the slides during the talk.     (1C6C)
  • Please note that this session will be recorded, and the audio archive is expected to be made available as open content to our community membership and the public at-large under our prevailing open IPR policy.     (1C10)

Attendees     (1D)

Agenda Ideas     (1F)

(please insert your ideas below)     (1F1)

Agenda & Proceedings     (1G)

1) Welcome & calling the meeting to order     (1G1)

  • Leo & Mike co-chaired the call and welcomed everyone.     (1G2)
  • Peter volunteered to take notes on the proceedings     (1G3)

2) Participants' self-introduction (affiliation, pertinent work, etc.) and expectations from this call or the initiative (2 minute each)     (1G6)

3) Key Issue(s) discussion - "Open Ontology Repository initiative - Planning":     (1G8)

Discussion:     (1G10)

  • Needs & Requirements:     (1G11)
    • scope - keeping it lean (Duane) or rich in services (ref. details on Leo's slides)?     (1G11A)
    • centralized (and mirrored) vs. federated repositories     (1G11B)
    • how open? - Peter: totally "open" (malicious intents excluded, of course)     (1G11C)
  • Goals & a Mission Statement:     (1G12)
    • Mike Dean - first draft: "To promote the global use of ontologies by developing and maintaining an open, collaborative ontology repository."     (1G12A)
    • keeping it open on whether we will be hosting a repository or not.     (1G12B)
    • Farrukh's update: "To promote the global use of ontologies by enabling and facilitating open, federated, collaborative ontology repositories."     (1G12C)
    • John & Leo both feels that we should state that we ARE going to operate a repository (Duane's "1." should do)     (1G12D)
    • Duane's addition: "mission should be to: 1. establish a hosted registry-repository; 2. promote the global use of ontologies by enabling and facilitating open, federated collaborative ontology repositories,; 3. establish best practices for expressing ontology and taxonomy work in registry-repositories."     (1G12E)
    • Final edit: "promote the global use and sharing of ontologies by: 1. establishing a hosted registry-repository; 2. enabling and facilitating open, federated, collaborative ontology repositories; 3. establishing best practices for expressing interoperable ontology and taxonomy work in registry-repositories."     (1G12F)
    • Peter moves to adopt this (final edit version) working mission statement (subject to final confirmation by the working group); Rex seconded the motion; all agreed and adopted by this meeting.     (1G12G)

Duane Nickull: Registry-Repository facitilies should generally not contain any reasoning components.     (1G22)

R&R is simply a storage facility. Reasoning engines use R&R API's as a source of knowledge.     (1G23)

Duane Nickull: IN ISO/IEC 11179 2004 part 3, the purpose of the registry is to be loaded with the     (1G24)

first order logic that constraints further binary or n-ary relationships to a set number of types     (1G25)

(Symmetrical, Assymetrical etc). These can be inherited and labelled to build custom representations     (1G26)

of taxonomy concept maps.     (1G27)

Duane Nickull: Also - mapping does not work unless one considers the some context aspects. One gotcha     (1G28)

is the heirarchy in XML. A data element "FirstNameOfPerson" is not the same if it occurs in a path     (1G29)

Mark Feblowitz: I've had to ring off for another call. I'm quite interested in this topic. I'll be     (1G31)

listening to the recording later, and will be interested in any follow-on.     (1G32)

anonymous2: Looking at the R&R as analogous to an archive, I'm in agreement with Duane that the     (1G33)

registry should contain just first order logic.     (1G34)

Lee Feigenbaum: Open Anzo semantic middleware platform:     (1G35)

John Graybeal: Speed of initial roll-out will be important for some participants -- having a few     (1G36)

capabilities early will be extremely valuable.     (1G37)

John Graybeal: So I love the use cases/requirements -- thanks very much for providing those!     (1G38)

(Prioritization of them will be key.) And am eager to participate in a collaboration.     (1G39)

John Graybeal: Must sign off for now. Looking forward to the results.     (1G40)

Farrukh Najmi: Starting point for reading on ebXML RegRep and     (1G41)

Lee Feigenbaum: I got disconnected. WIll try to rejoin.     (1G43)

Li Ding: centralized semantic web ontology search engine -     (1G44)

It offers submit, index, search, version-tracking and REST web service.     (1G45)

Duane Nickull: Here is a registry&ontology project I did for the Canadian Government:     (1G46)

Anders Tell: Ref to: Architectural principles and point of views, IEEE1471, ISO 42010:2007     (1G48)

Mike Dean: To promote the global use of ontologies by developing and maintaining an open,     (1G49)

collaborative ontology repository.     (1G50)

anonymous2 morphed into Gerry Creager     (1G51)

Gerry Creager: I'm inclined to agree that the following are good points:     (1G52)

1. Federated repository structure although a distributed system should be employed.     (1G53)

Thus, we need to work toward interoperable and scalable repositories as individual ones evolve.     (1G54)

2. Individual software elements should be purpose-driven rather than large code-sets     (1G55)

(unix model of services).     (1G56)

3. Services will be what draws in users. Therefore, apps will need to evolve that are     (1G57)

user-requirement driven rather than simply providing a sterile API.     (1G58)

anonymous1 morphed into Leo Obrst     (1G59)

Rex Brooks: I got disconnected, but I will also make sure that the group I work with is informed about this.     (1G60)

It is the Integrated Response Services Consortium, focused on Emergency Management and Health Informatics,     (1G61)

putting together a very domain-specific Service-Oriented Architecture. I hesitated to muddy the waters by     (1G62)

suggesting that this effort, by its very nature, will tend toward developing a de facto SOA.     (1G63)

Duane Nickull: Think mission should be to establish a hosted registry-repository     (1G64)

Rex Brooks: However, I think Leo's point about the need for developing services to make use of the     (1G65)

registry-repository--provide benefits--is important.     (1G66)

Li Ding: a question, how the statement will be evaluated?     (1G67)

To promote the global use of ontologies by enabling and facilitating open, federated collaborative     (1G69)

ontology repositories.     (1G70)

Mike Dean: i like farrukh's proposal     (1G71)

Gerry Creager: That'd work for me.     (1G72)

Anders Tell: Farrukh new statements is a step forward.     (1G73)

Li Ding: The goal is to promote usage of ontology, but the approach is not limited to ontology repository     (1G74)

because there are many other factors, such as tools supporting ontology, mapping and integration services,     (1G75)

visualization services.     (1G76)

Rex Brooks: I agree with Farrukh's suggestion.     (1G78)

Duane Nickull: + 1 to Farrukh's statement.     (1G79)

Duane Nickull: mission should be to:     (1G80)

1. establish a hosted registry-repository;     (1G81)

2. promote the global use of ontologies by enabling and facilitating open, federated collaborative     (1G82)

ontology repositories,;     (1G83)

3. establish best practices for expressing ontology and taxonomy work in registry-repoositories     (1G84)

John Graybeal: Farrukh's candidate does not say we will develop anything, only that we will     (1G85)

facilitate development of things. Don't know if thats an issue or not.     (1G86)

Duane Nickull: yes - that is why I added the other parts     (1G87)

Farrukh Najmi: +1 on Duanes 3 part mission     (1G88)

Farrukh Najmi: ebXML RegRep 4.0 is being defined now. Interested parties should feed requirements     (1G89)

into that work.     (1G90)

Farrukh Najmi: I can be contact point     (1G91)

Mike Dean: I'd like to see the goal addressed up front, as promote the global use of ontologies     (1G92)

by 1. ... 2. ... 3. ...     (1G93)

Ken Baclawski: I would like to see interoperability explicitly mentioned as a goal for the     (1G94)

Ken Baclawski: best practices.     (1G95)

Duane Nickull: interoperability and other requirements couldbe collected and documented in a     (1G96)

Requirements Document which would help us all understand what we might want to concentrate on.     (1G97)

Duane Nickull: interoperability is a requirement that reg-rep can help solve. interoperability is not     (1G98)

what a reegistry is all about ipso facto     (1G99)

Rex Brooks: I think a requirements document with use cases is a good idea. Registries contribute toward     (1G100)

that, enable it, but don't enforce or produce it.     (1G101)

Gerry Creager: Interoperability, correctly stated, isn't the sole function of an registry. However, if     (1G102)

it doesn't help promote it from the start, we lose the opportunity to see a real federation evolve easily.     (1G103)

Ken Baclawski: I agree that the registry is not about ontology interoperability per se. That is why I     (1G105)

only want it mentioned in the best practices part.     (1G106)

Li Ding: the best practice part focus on ontology creators but not ontology consumers     (1G107)

Farrukh Najmi: I am in listen + chat mode now and cannot speak into conf due to some logistical issues.     (1G108)

If any one needs my input just look for response in chat window.     (1G109)

Lee Feigenbaum: I like OOPS but I don't know what the P and S stand for     (1G111)

Farrukh Najmi: Can someone speak my suggestion of ONTOREP as I think I cant be heard     (1G112)

Li Ding: Collaborative Open Ontology Library     (1G113)

John Graybeal: COOL is used elsewhere, will be confusing     (1G116)

Lee Feigenbaum: OREP?     (1G117)

Li Ding: Oori is MIT's student pungmul group     (1G119)

4) Next Steps & Action Items:     (1G122)

  • Our next meeting/call on this initiative will be two weeks from now (exact date/time to be confirmed on the mailing)     (1G124)
  • this call adjourned 12:35pm PST     (1G125)

minutes captured in real time on this wiki by Peter P. Yim / 2008.01.03-12:50 PST     (1G127)

Attendees are requested to review the notes above, help clean it up, and make amendments and edits to better     (1G128)

reflect the proceedings of the session as they see fit (particularly where they are personally involved).     (1G129)

Questions, Answers & Discourse     (1H)

  • Please mute your phone, by pressing "*2" on your phone keypad, when the talk is in progress. To un-mute, press "*3"     (1H1)
  • If you want to speak or have questions or remarks to make, please "raise your hand (virtually)" by pressing "11" on your phone keypad. You may speak when acknowledged by the speaker or the session moderator.     (1H2)
  • You can also type in your questions or comments through the browser based queue management chat tool by:     (1H3)
  • For those who have further questions or remarks on the topic, please post them to the [oor-forum] list so that everyone in the community can benefit from the discourse. (One needs to be subscribed to this archived mailing list first before posting. See subscription details [ here].)     (1H4)

Audio Recording of this Session     (1I)

This page has been migrated from the OntologWiki - Click here for original page     (1I6)