ppy/chat-transcript_unedited_20131031a.txt ------ Chat transcript from room: ontolog_20131031 2013-10-31 GMT-08:00 [PDT] ------ [9:12] PeterYim: Welcome to the = RulesReasoningLP: mini-series session-02 - Thu 2013-10-31 = Session Co-chairs: Dr. LeoObrst (Ontolog; MITRE) & Dr. HaroldBoley (RuleML; U of New Brunswick) Topic: Concepts and Foundations of Rules and Ontologies: Logic Programs, Classical Logic, and Semantic Web - I Panelists / Briefings: * Dr. HaroldBoley (University of New Brunswick; RuleML) - "From Data to Knowledge through Grailog Visualization" * Dr. BenjaminGrosof (Benjamin Grosof & Associates) - "Hilog, Defeasibility, and the Foundations of Practical Meta Knowledge: A Brief Introduction" * Professor GeorgGottlob (Oxford) - "Datalog+/- a Unifying Framework for Ontological Reasoning and Query-Answering" Logistics: * Refer to details on session page at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2013_10_31 * (if you haven't already done so) please click on "settings" (top center) and morph from "anonymous" to your RealName * Mute control (phone keypad): *7 to un-mute ... *6 to mute * Attn: Skype users ... see: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2013_10_31#nid3ZTO ** you may connect to (the skypeID) "joinconference" whether or not it indicates that it is online (i.e. even if it says it is "offline," you should still be able to connect to it.) ** if you are using skype and the connection to "joinconference" is not holding up, try using (your favorite POTS or VoIP line, etc.) either your phone, skype-out or google-voice and call the US dial-in number: +1 (206) 402-0100 ... when prompted enter Conference ID: 141184# ** Can't find Skype Dial pad? *** for Windows Skype users: Can't find Skype Dial pad? ... it's under the "Call" dropdown menu as "Show Dial pad" *** for Linux Skype users: please stay with (or downgrade to) Skype version 2.x for now (as a Dial pad seems to be missing on Linux-based Skype v4.x for skype-calls.) Attendees: AlanRector, AlexShkotin, AliHashemi, AmandaVizedom, BenjaminGrosof, BobSmith, BobbinTeegarden, BrandonWhitehead, BrianHaugh, ConradBock, DennisPierson, ElieAbiLahoud, FranLightsom, FrancescaQuattri, GenZou, GeorgGottlob, HaroldBoley, HensonGraves, JuanSequeda, KenBaclawski, LamarHenderson, LeoObrst, MichaelGruninger, MikeBennett, MikeDean, PeterYim, SimonSpero, TaraAthan, ToddSchneider, == Proceedings == [8:38] anonymous morphed into Conrad [9:19] anonymous morphed into BrandonWhitehead [9:27] PeterYim: Hi everyone! [9:27] PeterYim: Hi AlanRector ... does the vnc access (with the local viewer) work for you now? [9:29] MikeDean morphed into MikeDean [9:30] anonymous morphed into TaraAthan [9:31] Conrad morphed into ConradBock [9:31] anonymous1 morphed into GeorgGottlob [9:32] anonymous1 morphed into BobbinTeegarden [9:32] anonymous morphed into FrancescaQuattri [9:36] anonymous morphed into BrianHaugh [9:36] PeterYim: == LeoObrst and HaroldBoley starts the session ... see slides at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2013_10_31#nid3ZT9 [9:38] anonymous morphed into FrancescaQuattri [9:40] PeterYim: == HaroldBoley presenting ... [9:41] DennisPierson morphed into DennisPierson [9:52] anonymous morphed into JuanSequeda [10:05] SimonSpero: @HaroldBoley: has there been any user studies/ cognitive modelling of how easily these graphical notations are understood (before training, after n-hours training, n-days/weeks after training?) [10:08] MikeBennett: @Simon I could not help thinking of the guy from Australia who presented at the OMG in March about taking a more scientific approach to the graphics of modeling languages. Can't remember his name (anyone?) [10:12] ElieAbiLahoud: @MikeBennett: Dr Daniel Moody, http://www.omg.org/news/meetings/tc/dc-13/special-events/Physics_Notations_Tutorial.htm [10:12] MikeBennett: That's the guy! Ta. [10:12] HaroldBoley: @SimonSpero, I didn't use parallel ('control') groups but sequential groups when teaching Semantic Web Techniques (http://www.cs.unb.ca/~boley/cs6795swt/notes.html), noticing that recent student groups understood Semantic Web Logics better and faster than earlier ('pre-Grailog') groups attending this course (some with interdisciplinary background). Quantitative studies should follow next, including colleagues from the Social Sciences. If you are interested, please do contact me. [10:21] HaroldBoley: @MikeBennett, Yngve Lamo does 'typing homomorphism' (linked from http://www.cs.unb.ca/~boley/talks/RuleMLGrailog.pdf, slide 115). [10:06] LeoObrst: Harold, can you describe how a modal expression would be depicted in Grailog? I.e.,with necessity/possibility operators. [10:17] HaroldBoley: @LeoObrst, modal expressions use complex nodes as shown in slides 104ff of the long version of this talk (http://www.cs.unb.ca/~boley/talks/RuleMLGrailog.pdf). [10:28] MikeBennett: Thanks @Harold that looks very interesting. [10:03] PeterYim: == BenjaminGrosof presenting ... [10:17] PeterYim: [Grosof: note - typo on slide#4 ... Benjamin: please update, and I will swap it in.] [11:29] PeterYim: re: [10:17] the BenjaminGrosof slide deck has now been updated ... Thank you, Benjamin [10:19] HaroldBoley: @Benjamin, 'backquote' is visualized with the (snipangle) 'instantiation' boxes introduced of Grailog. [10:21] SimonSpero: The use of backtick threw me for a bit [10:21] SimonSpero: [Also, KR macros are awesome] [10:24] LeoObrst: Benjamin, do you have "contexts", i.e., for examples like "believes( john, ${ likes(mary,bob) } )" on slide 3? I.e., the truth value of the embedded argument is local? Or not? [10:30] AmandaVizedom: +1 for Leo's question about contexts, or something (perhaps uses the "holds" trick) that leverages all of this great support and enables modularizing reasoning, or defining permitted paths of reasoning across contexts of various kinds, vs. non-permitted? [10:48] BenjaminGrosof: Answer to Leo's question on contexts: the truth value of the embedded argument is relative to the believer John, essentially. [10:26] AmandaVizedom: re: slide 6: Another (suggested but not stated) very important kind of reasoning enabled by these KR tools: for federated data (including linked data) applications, not only *having* provenance but *using* provenance to tailor how information from different sources is treated... and being able to capture & expose rules about *that* in the ontology/knowledge base. [10:32] AmandaVizedom: ... or defines 'contexts' in which rules, prioritizations, etc., apply? [10:52] BenjaminGrosof: +1 to Amanda's comment about using provenance. [10:34] ConradBock: Does well-founded LP fit in somewhere? [10:50] BenjaminGrosof: Answer to Conrad's question: LP and its extension Rulelog discussed here is under well founded semantics, cf. my survey presentation on 10/24/2013 Ontolog Forum session - ref. http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2013_10_24#nid3ZSA [10:51] BenjaminGrosof: That's the semantics most frequently used due to good computational complexity and algorithmic characteristics. [10:39] ElieAbiLahoud: Thank you Benjamin and Harold. I'am afraid I have to drop-off now. [10:38] PeterYim: == GeorgGottlob presenting ... [11:05] BenjaminGrosof: wrt Georg's slide 30: note that stratified naf ("negation as failure") is a special case of well founded semantics, and also a special case of stable semantics / answer set semantics. [11:06] LeoObrst: @Georg: on slide 9, is linear datalog related to linear logic? [11:18] GeorgGottlob: @Leo: "linear" here is not related to linear logic; it just means only one atom in rule bodies [11:09] PeterYim: === GiorgioOrsi (one of GeorgGottlob's postdocs) saying a few words about his implementation of Datalog+/- ... re. slide#37 [11:11] PeterYim: == Q&A and Open Discussion ... [11:09] AmandaVizedom: @BenjaminGrosof: What tooling exists that implements / supports working with Hilog? [11:11] BenjaminGrosof: Answer to Amanda question about Hilog tooling: [11:11] BenjaminGrosof: A number of systems support restricted Hilog. E.g., Jena permits variables in predicate position. [11:12] BenjaminGrosof: The most sophisticated implementation is in Flora-2, used in Coherent Knowledge Systems software commercially now. (By most sophisticated: I mean both expressive and efficient. Coherent has a proprietary extension of Flora-2.) [11:37] BenjaminGrosof: More answer to AmandaVizedom question about tools for Hilog: There are two aspects: engine and UI. Engine techniques for handling functions come from Prolog (LP) and FOL. Unification is important. LP uses tries for indexing and "tabling" for efficiency/termination. State of the art on tabling is in XSB and Flora-2, then extended proprietarily wrt Hilog in Coherent Knowledge Systems software. Coherent also supports Hilog, esp. for textual logic (see Ontolog Forum 6/20/2013 session - ref. http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2013_06_20 ) with particular features in its proprietary (but available) UI and so to an extent did the UI in the earlier Vulcan's SILK (code not available publicly, team disbanded). [11:13] AlexShkotin: Are there any use cases of usage of that languages in any domain areas? [11:19] BenjaminGrosof: Answer to Alex question: yes, there have been many usages of Hilog. [11:20] AlexShkotin: @Benjamin, do you have a storage of Hilog ontologies. Just not to do twice:-) [11:28] BenjaminGrosof: Answer to Alex question on storage of Hilog ontologies: I don't understand the question. [11:29] AlexShkotin: @Benjamin, like repository to download Hilog ontologies. [11:39] AlexShkotin: I just have asked about Hilog ontologies repository as we have for OWL 2. [11:31] SimonSpero: Alex: http://flora.sourceforge.net/ [11:31] SimonSpero: @Benjamin: why SVN in 2013, instead of a Github repo? [11:24] AlexShkotin: If we have some knowledge representation as text (for ex. on CNL) we can compare different formal languages as a tools to use. For ex. we convert DB to CNL and then to OWL2:-) [11:27] AlexShkotin: So CNL representation of a knowledge may be most general against formal languages. [11:35] PeterYim: Join us again, same time next week (Thu 2013-11-07) for the OntologyBasedStandards mini-series session-06 on "Ontology-based Financial Standards" co-championed by MikeBennett, BillMcCarthy and ElieAbiLahoud [11:36] PeterYim / LeoObrst: next in this series - 2013_11_21 - Thursday: RulesReasoningLP mini-series session-03: Concepts and Foundations of Rules and Ontologies: Logic Programs, Classical Logic, and Semantic Web - II - Co-chairs: LeoObrst & PascalHitzler [11:36] LeoObrst: Thanks, All! [11:37] PeterYim: great session! [11:37] AlexShkotin: Thanks [11:42] AlexShkotin: C u [11:37] PeterYim: -- session ended: 11:36am PDT -- ------