ppy/OntoIOp-confcall-n44_chat-transcript_20131009b.txt ------ Chat transcript from room: ontoiop_20131009 2013-10-09 GMT-08:00 [PDT] ------ [7:55] PeterYim: . = OntoIOp team-confcall (n.44) - Wed 2013.10.09 = * Date: Wed 9-Oct-2013 * start-time: 8:00am PDT / 11:00 am EDT / 5:00pm CEST / 4:00pm BST / 12:00am[+1] KST / 1500 UTC ** ref. world clock - http://www.timeanddate.com/worldclock/fixedtime.html?month=10&day=9&year=2013&hour=8&min=30&sec=0&p1=224 * Duration: 1.0~1.5 Hrs. * in-session chat-workspace: http://webconf.soaphub.org/conf/room/ontoiop_20131009 * shared-file workspace for this session: http://ontolog.cim3.net/file/work/OntoIOp/Team_confcall/2013-10-09_team-confcall_n.44/ Remarks: * the session may be recorded for archival purposes. Unless otherwise documented, participants agrees to this by virtue of their participation at the session. * In case we have to mute everyone (due to extraneous noise or echo) - Mute control: *7 to un-mute ... *6 to mute * for Windows Skype users: Can't find Skype Dial pad? ... it may be under the "Call" dropdown menu as "Show Dial pad" * for Linux Skype users: please note that the dial-pad is only available on v4.1 (or later or the earlier Skype versions 2.x,) if the dialpad button is not shown in the call window you need to press the "d" hotkey to enable it. (--CLange) AGENDA: (a) brief report from the OMG New Jersey meeting [Gruninger] and discussion (b) planning for the OMG proposal submission Attendees: TillMossakowski (chair), ChristophLange, MariaKeet, MichaelGruninger, PeterYim (scribe), TerryLongstreth == Proceedings: == [8:02] PeterYim: == TillMossakowski starts the meeting ... [8:04] TillMossakowski: I have started to redesgin the OntoIOp homepage at: http://ontoiop.org [8:04] TillMossakowski: have confirmed with Elisa that we can openly publish the RFP [8:04] PeterYim: Till: agenda for today (a) brief report from the OMG New Jersey meeting [Gruninger] and discussion; (b) planning for the OMG proposal submission [8:05] PeterYim: === MichaelGruninger reports on the OMG New Jersey meeting in September [8:06] PeterYim: @Maria @Till ... I muted both of you (as you are introducing noise to the line) ... please un-mute with *7 before speaking [8:06] TillMossakowski: for the record: here are Michael's notes: [8:06] TillMossakowski: == review of issues brought up at the New Jersey meeting ... 1. This is too ambitious -- this project is trying to solve all interoperability problems. - the scope of the metalanguage is too broad and ambitious (Note: I think we need to emphasize heterogeneous interoperability among ontologies; some people thought we were specifying mappings between all possible languages and ontologies) 2. This sounds like a research project, not a standards project. Response: Provide formal definitions with citations to the literature where these definitions have been stated. e.g. what is the definition of a logically heterogeneous ontology? 3. The role of translations and mappings - will translations between specific languages (e.g. OWL-2 and CL) be part of the standard? - if so, will only one translation be allowed? - what is the relationship between different translations between the same two languages? - will mappings between specific ontologies be part of the standard? 4. What is the need for the metalanguage? This justification needs to be aligned with the mandatory requirements. Show how the requirements and related to the use cases. 5. How detailed are the metalevel relationships? For example, consider the Date-Time Vocabulary (dtv). Do we simply say that dtv_owl is an approximation of dtv_cl or does the standard specify in what way it is an approximation, saying what is lost and what is preserved. 6. Are the conformance criteria for logical languages specified with enough detail to be testable? [8:12] PeterYim: Michael: lots of comment from Nicolas Rouquette - "Rouquette, Nicolas F (316A)" [8:12] TillMossakowski: ad 1: we should emphasize that only a small subset of languages and translations needs to be covered (comparable in number to that of ODM). Moreover, the emphasis is on the meta language, not on the concrete languages and translations. Thanks to Elisa for mentioning fUML; this is a good point for keeping the inclusion of UML manageable. We can also refer to DOL, which is a possible proposal answering the RFP, with a number of logics and translations and even tool support available. [8:14] TillMossakowski: the reference to DOL is problematic, because DOL will be an answer to the RFP. So we should leave that out. [8:17] TillMossakowski: we should have two separate requirements: 1) small core set of languages to be covered, 2) extensibility for any logical language with a model theory [8:18] TerryLongstreth: You should be able to describe the history of DOL (and it's predecessor projects) without specifics of the DOL proposal [8:18] TillMossakowski: ad 2: we have enough references to cite [8:19] TillMossakowski: Heterogeneous Logical Theory should appear in the glossary [8:20] TillMossakowski: Some useful reference is Oliver Kutz, Till Mossakowski, Dominik Luecke (2010). Carnap, Goguen, and the Hyperontologies - Logical Pluralism and Heterogeneous Structuring in Ontology Design. In Logica Univeralis, 4 (2), pp. 255333. http://www.informatik.uni-bremen.de/~till/papers/Hyperontology.pdf [8:26] TillMossakowski: ad 3: [8:26] TillMossakowski: - specific languages and translations should be part of an informative annex [8:27] TillMossakowski: - there should be more than one translation allowed, with default translations [8:28] TillMossakowski: - the relationship between different translations often is that there are different trade-offs, and they have different properties [8:29] TillMossakowski: Michael: we need to be more precise about what are different, and different but equivalent translations [8:29] TerryLongstreth: How can we establish conformance for an extensibility requirement? [8:32] TillMossakowski: I would say: we require a mathematical definition of a notion of logic (e.g. institutions), and the meta language should work for all logics falling into this definition [8:33] TillMossakowski: ad 3: mappings between specific ontologies will not be part of the standard [8:33] TillMossakowski: Maybe we even need more use cases for those requirements that have not been "grounded" by a use case yet. [8:34] TillMossakowski: ad 4: Maybe we even need more use cases for those requirements that have not been "grounded" by a use case yet. [8:34] TillMossakowski: we need to look at the table of features in the old working draft [8:35] TillMossakowski: ad 5: Do we simply say that dtv_owl is an approximation of dtv_cl ? Yes! [8:36] TerryLongstreth: "approximation" is equivalent to a lossy transformation; i.e. unidirectional and non-reversible [8:40] TillMossakowski: there is research about approximating OWL in EL, and Prop in HornProp, and also some initial work about approximating FOL in OWL. [8:43] TillMossakowski: usually, in the research literature, approximations are required to be maximal. If they are not, use an interpretation of theories. [8:45] TillMossakowski: action: Michael asks Mark about DTV [8:46] TillMossakowski: ad 6: I just repeat: we require a mathematical definition of a notion of logic (e.g. institutions), and the meta language should work for all logics falling into this definition [8:52] PeterYim: Till: next meeting in 2 weeks, as usual - OntoIOp confcall (n.45) Wed 2013.10.23 start-time: 8:00am PDT / 11:00 am EDT / 5:00pm CEST / 4:00pm BST / 12:00am[+1] KST / 1500 UTC [8:54] PeterYim: [action] Till will lead the edit of the RFP and put a version on google-doc for collaborative contribution on editing suggestions and proof-reading ... Till will later convert this back to a word-doc for submission to OMG [8:55] PeterYim: -- session ended: 5:51pm CEST -- ------