chat-transcript_20100315d.txt anonymous morphed into PeterYim PeterYim: . Welcome to the OntologySummit2010: March 15~16, 2010 Symposium (Face-to-Face Workshop) (22Q7) Theme: Ontology Summit 2010: Creating the Ontologists of the Future (2861) 5th in the series of a 3-month annual event by and for the Ontology Community. This Summit is co-organized by Ontolog, NIST, NCOR, NCBO & IAOA (2992) * Co-chairs: Dr. SteveRay & Professor BarrySmith Please refer to agenda, dial-in and other details on the session page at: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2010_Symposium . PavithraKenjige: Hi Peter anonymous morphed into JulitaBermejoAlonso Tom Brunner: IS the audio line open. It appears to be quiet. PeterYim: yes, Tom ... it's startingnow Simon Spero: twitter #tag so for is #ontologysummit2010. Suggest #ontsum2010 PeterYim: @Arturo - slide#16 ... the OOR team would want to offer our platform to host your recommended ontology education program repository/registry anonymous morphed into BruceBray anonymous morphed into FabianNeuhaus RaviSharma: Great Observation from Dagobert Soergel in terms of at least a course in ontology like bio informatics in medical education, thus requiring projections of Teachers required by types of expertise required to cater to the students numbers projeted. RaviSharma: Todd: if we looked at the answers by expertise type 3 categories described by Peter, we can try and answer part of your questions. However direct survey of ontology, related to ontology type of categories would better answer your Q. RaviSharma: Amanada in yr slide 7 one of the numbers should be 37% rathr than 3.7%? RaviSharma: Amanda: extremely important to continue extend these surveys Simon Spero: Hello posterity anonymous1 morphed into DeborahMacPherson RaviSharma: Mike Grunninger: your articulation of disciplines, is it typical such as Computers-Applied Logic-Mechacical Engg and now interacting with bio-informatics? Would it be possible to extract respondent types with 1-2-3 disciplines from Amanda's Data? Simon Spero: Disciplines are artificial constructions evolved to keep people in their own ivory silos and stop them attacking your own RaviSharma: Mike: What skills does an ontologist need to distinguish whether a given problem posed to them is or is-not amenable to ontology approach? anonymous1 morphed into Cory Casanave RaviSharma: Simon: I agree aand the diffusion boundaries are most interesting areas where people like me enjoy working, it is different for every one, some are SMEs in one or two deeper study areas. Cory Casanave: The US Dial in number does not seem to be correct? RaviSharma: Perwill answer shortly cory RaviSharma: Peter will answer shortly PeterYim: @Cory - are you using 218-844-8060 - see: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2010/WorkshopRegistration#nid297W PeterYim: @Cory - if the number is correct, please try again (sometimes it takes a few tries, if the circuits are busy) RaviSharma: Mike: you answered part of my Q in 3 but what does it really take to realize that either our tools are not applicable, or not yet there to apply or deny ontology solutions to a given problem. RaviSharma: Mike: This Core skill or optional skill ought to also address developmens needed in ontologies for them to be useful for practical applications and identify Gaps where progress in ontology logic stat tools where gaps could increase productivity? Simon Spero: Skills and Knowledge lists could do with some Semantic Factoring. Cory Casanave: @PeterYim - did eventualy work, but the volume on the phone is very very low PeterYim: let me have the speaker get closer to the microphone, let me know if that works Gary Berg-Cross: Just for the record, according to some sources in 1875 William James set up the first laboratory of experimental psychology. This was four years before Wundts laboratory in 1879. James was not recognized as having the first running laboratory in psychology because his laboratory was used mainly for teaching demonstrations. RaviSharma: Simon: we would practice what we are talking about by doing that, and would be nice to identify skills, lists that do not matter for future ontologist for now at least? anonymous1 morphed into JulitaBermejoAlonso Simon Spero: Baader, Franz (2003). The Description Logic Handbook: Theory, Implementation, and Applications. Cambridge, UK: Cambridge University Press. JulitaBermejoAlonso: sorry, I had to go. Who is talking now? RaviSharma: Julia: Barry Smith RaviSharma: Julita: I meant your name PeterYim: see: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2010_Symposium#nid28QR ... on slide#20 now RaviSharma: Barry: your point proves that even for formal educators and good professionals, tenure track is absolutely ridden with failures, should we not evolve better and less stressful job-value systems? RexBrooks: Can barely hear Barry. RexBrooks: Can you get a microphone to Barry? PeterYim: giving him a different microphone now ... any better? RexBrooks: Is it not the case that the formalisms of logic in philosophy allow the computability of ontologies? RexBrooks: Yes, a little better. Thanks. RaviSharma: Michael Uschold: Philosophy is in a way more core than Logic and math and also history of original capture of thoughts, thus by definition philosophy is essential pre-requisite, but tailored to ultimate objectives of course and skills? RexBrooks: @Ravi: It would also help bring our philosophers back into the practical world a bit more. They have much more to contribute, but I think they find their academic environment more comfortable than solving real world problems that they are uniquely qualified to help solve. anonymous1 morphed into AmandaVizedom Brand: Comment and Question: The academic trend seems to be adding informatics (mostly technology) to disciplines (e.g. bioinformatics) and having experts write books to combine both technology and ontolog (e.g. Professor Ken Baclawski "Ontologies for Bioinformatics"). Is this an easier way rather than trying to have separate departments of ontology? RaviSharma: Fabian:Enterprise Architecture is an area I have been practising 10 years, I would be happy to contribute some text for Fabian and Denise. AmandaVizedom: I agree with the concern that a new department is too likely to be a new silo. Ref my previous comments about departments as inherently stagnating, and add the stew of academic politics and rivalries. I also don't think that philosophy or any other existing department is a good home for ontology. I'd argue that the interdisciplinary nature of applied formal ontology is inherent and valuable. It can also be skewed and even broken, and that will make the ontology less valuable. For these and other reasons, I think that an interdisciplinary center with faculty co-tenured in many different departments is better, both practically and for the well-rounded maturation of the field itself. RaviSharma: Rex: Are not Barry and Amanda both philosophers great examples of ontologists from Philosophy who are producing solutions and also growing and supporting human capital in these areas? I am sure there are others as well as Barry mentioned Pat Hayes started in Philosophy. Simon Spero: @Amanda: iSchools tend to have a lot of double-appointments Simon Spero: @Amanda: And a lot of people who have single appointments but work closely with other depts. Simon Spero: I graduated from Imperial: still has no Computer "Science" department. Department of Computing, yes. AmandaVizedom: Centers have billets and budgets as well. In the example I've referenced, the Minnesota Center for Phil/Hist/Soc of Science has cotenured faculty across depts, is better funded than the phil dept, gets strong support from the university administration because of the broad participation it attracts, and has existed for decades. Simon Spero: Amanda: Right RexBrooks: @Amanda: I agree, I didn't mean to suggest that ontology should live inside a philosophy department, only that philosophy is indispensable and engaging that faculty on practical matters is something that would be valuable, after an initial period of acclimatizing them RexBrooks: I thing that computer science departments are too bound to software development and already give too little emphasis to information theory and information science. Simon Spero: Rex: what must come from philosophy? FOPL, some. Ontology (classic)- a bit. Epistemology - lots (degrees of belief, update/revision). Temporal/Spatial -lots Simon Spero: Rex: What must come from LIS/Cog Science: Human Information Behavior; concept theory ; Knowledge Organization Systems; working with inter-disciplines AmandaVizedom: Extending the example: I do think the MN center and programs provide a model that we would do well to study and to some degree emulate: The MN Phil Sci center: http://www.mcps.umn.edu/; The cross-disciplinary Studies of Science and Technology graduate minor: http://www.mcps.umn.edu/grad/program.html; An example of the spreading phenomenon inspired by the MN Center's smashing success: http://www.pitt.edu/~pittcntr/About/history/history_1.htm Simon Spero: Amanda: Center is a bad word in some states right now Simon Spero: What is it like to be an ontologist? AmandaVizedom: Picking up on current speaker's comments: Agree that we need to acknowledge the multiple parentage, in more than words. We need to be structurally open to that continued life-blood. Simon Spero: What needs to come from b-schools? Cory Casanave: Note that we dont have modeling departments or architecture departments . Likewise formal modeling as a skill set is required across many disciplines and is not, in general, taught as something to apply to these related skill sets. AmandaVizedom: @Simon: Institute? AmandaVizedom: @Simon: Very much like what it is like to be a bat. Simon Spero: (in re current speaker): This is top down approach Simon Spero: ETL Simon Spero: Extract, Torture, Lose AmandaVizedom: In fact, the data silo problem is in some cases an extension of an Enterprise Architecture problem -- including the problem of making Enterprise Architecture implementable. Simon Spero: @Amanda: do we need an archive of bad ontologies RexBrooks: @Simon: Game Theory; Set theory;Catastrophe Theory are also needed, and relate strongly to Cory's note on modeling. Simon Spero: Rex: Set theory, mandatory (but need for logic) Simon Spero: Rex: Game theory - needed if opponent is involved (IC applications) AmandaVizedom: I don't think that we will evolve to fit the right environment if we start with *any* existing department. I think more and better study of fielded ontology (and fielded ontologists) need to serve as our cross-disciplinary anchor. I also thinking that there is funding to be had on cross-disciplinary projects that meed current user needs. Simon Spero: Amanda: I think that describes iSchools, to the extent that they exist AmandaVizedom: Most EA artifacts don't go down/up to the information level. Stop at exchange of information bearing objects. Can use semantic technologies to bring it further and see how IT activities relate to EA models etc. RaviSharma: Barry: agreen that is immediate value to X-Agency EA but there are many related areas such as Complex Event Processing.BPM and Organization goals incl Citizen services for fed and Sstae levels. RexBrooks: Game theory is used for who needs what questions. Opponents not necessary, just different sets of interests. Simon Spero: Rex: Not sure if I'd label that as game theory, but accept conclusion Simon Spero: Decisin theory RaviSharma: Leo: well articulated, thanks. AmandaVizedom: Philosophers tend to be good at conceptual modeling, but depends on temperament and particular program whether they can model concepts in context/use or think about problem-solving like an engineer needs to or think about the cognitive factors shaping the way the information comes in and may need to go out. AmandaVizedom: Every other discipline mentioned has similar areas of special strength and weakness. Cory Casanave: When the concdeptual model is used as the basis for the executable model is is not put on the shelf. We want to augment the logical model so the executable artifacts emerge. RexBrooks: @Simon: good enough. I also combine it with Catastrophe Theory for adding in factors for cognition, performance, unanticipated external variables, making decisions highly problematic, especially in emergencies--the domain I live in most. AmandaVizedom: We need each other from the beginning; the maturation of the discipline requires these forces and insights from the beginning. It should not wait until there is a stable, codified base that is skewed toward one parent or another. RexBrooks: Can't hear Barry at all, now. AmandaVizedom: What about cognitive science / cognitive studies? AmandaVizedom: That is, as a recent case with a history that may be instructive. RexBrooks: Can't hear this speaker either. Simon Spero: Garca-Solaco, Manuel, F`elix Saltor, and Mal u Castellanos (1996). Semantic Heterogeneity in Multidatabase Systems. In: Object-oriented multidatabase systems. Ed. by Omran Bukhres and Ahmed K. Elmagarmid. Prentice Hall. Chap. 5, pp. 129195. ISBN: 0131038133, 9780131038134. anonymous1 morphed into MarleneRockM RaviSharma: Amanda: that effort would be really worthwhile as we would get clearrer x-disciplinary linkages underlying ontology training and edcation requirements PavithraKenjige: AI,Knowledge Management, Semantic Web.. all of these part of Computer Science discipline, but specialization.. PavithraKenjige: They all use Ontology PavithraKenjige: So I agree with the concept of Ontology being a Science rather than worry about Systems or SOftware Engineering etc.. RaviSharma: Fabian: request - Barry' Comments to be captured for Communique I am requesting Fabian to kindly note comparison bet math and engg and also physics, etc. AmandaVizedom: Different comment: Meant to say earlier, don't want to forget: I've mentioned working with "ontologists" who are unqualified and without training. I've not given equal time to something else I've experienced: working with "ontologists" who have been through some of the existing training programs, academic or commercial, and are just as unqualified. It's only the specifics that vary. This is a strong driver for my suggestion that the survey mechanisms be refined and continued, and that the results be a continuing ingredient in curriculum design. Simon Spero: Hey - why don't we do ontology on this? Simon Spero: What departments/disciplines exist Simon Spero: What skills are relevant to ontology AmandaVizedom: Agree with Steve Ray: We can't do this at the same time. Simon Spero: What knowledge is obtainable once we have those skills AmandaVizedom: @Peter Brown: Absolutely agree that message needs attention, but also think it can be separated. Here's why... GeraldRadack: Ada Lovelace (1815-1852) is generally credited with being the first programmer. Her field was mathematics. AmandaVizedom: We perceived going in that there is already demand, and the survey results strengthen that perception. We can say: We are not here making the case for the value of ontology. We are saying: it's clear that some people value ontology and want/need ontologists already. To meet that demand, we need training programs. Let's look at what they need to contain, starting from those cases and our varied background knowledge and experience. Simon Spero: what Amanda said RaviSharma: Mike Grunninger: Please add - Uncertainties and Strengths and Types of Relationships (predicates - by relevence by domain for exapmle) are imporatent to include in Communique as knowledge gaps or research areas for future ontologists related training. Simon Spero: "What, if anything, is an ontologist?" AmandaVizedom: On Ontology Evaluation: Critical and under-developed. Existing stuff is mostly the "easiest" formal analysis. Much harder, and urgently needed: sophisticated ways to assess ontology quality in situ. Integration of such evaluation at various points in the lifecycle. Understanding of how factors such as the application context, the user context, and ??? bring to the fore particular dimensions of quality, metrics and ways of evaluating. Training is needed on the methods that exist, but ad-hoc methods being used in projects, or dreamed up in projects but never implemented, need to be researched. anonymous1 morphed into Joanne Luciano Joanne Luciano: Regarding Methods and Metrics for Ontology Evaluation - of Utility in an application and for reuse (in another application) see: http://ontology.buffalo.edu/08/IDO/Luciano.pdf (powerpoint presentation) - Simon Spero: Where is the gallery of ontological horrors? AmandaVizedom: Noting JoanneLuciano's comments just now about metrics and measures: This kind of thinking about eval is just the kind of research we need in current projects. To my knowledge, Joanne is one of the very few people who have done such research. Simon Spero: Good examples, horrible warnings RaviSharma: Jim Disbrow: Appreciate your example of multiple language emaning and reasoning -linguistic eqvivalences and meaning of transliteration semantically meanings Simon Spero: Danger: Translations may be indeterminate. approach all rabbits with caution RaviSharma: Jim: inductive deductive and semantic language translations? Simon Spero: I tried to call you earlier. Is your phone on? AmandaVizedom: Like Barry, I get many emails about openings, and the ones I get probably have little overlap with the ones he gets, as those I see are not in biological ontology, for the most part. Moreover, I'm aware of many searches going on *for which I never see a posting or listing*. Sometimes this may be because there is a mismatch between where the ontologists are and where the jobs are posted, and/or between the language used. In other cases, positions are likely filled by personal reference. Simon Spero: "Where the ontologists are." So we *are* going for early education RexBrooks: If you are going to use the term "Semantic Technolog of technologies, you should also explain the difference between those technologies and the Semantic Web or there will be confusion? RaviSharma: If we agree also suggested by Terry Longstreth present here Let us just add Research and graduate level Learning areas in communique. RexBrooks: Let me correct what I wrote a moment ago: If you are going to use the term "Semantic Technology" or "Technologies", you should also explain the difference between those technologies and the Semantic Web or there will be confusion! RaviSharma: Fabian: Please see above, this would perhaps address your observation? Gary Berg-Cross: Modify the last sentence in (2BZ to read "This is a large obstacle for communication between ontologists and between ontologists and the users of ontologies. AmandaVizedom: Very Important Reason for Certifications: I am often asked to provide suggestions for such un-posted openings. This is in a work context, where my team provides technical support to theirs. I am a contractor and the folks asking are contractors; we work for different companies, on different contracts, for different sub-organizations... of the government. I can't point them to specific people or companies, ethically or legally. If there were any easily recognizable cue that a candidate is qualified, I could and would eagerly point them to that. There isn't -- no degree, no title, no field. A certification would be a good start. PavithraKenjige: I can not hear anything? Simon Spero: Amanda: The only onto courses I've seen graded were projet based RaviSharma: Mike Grunninger for Communique: It is also important to distinguish between two streams, ontology-science as suggested by Prof Barry, and Also the other important track of Practical applications of ontologies -ontology-engineering-Computational ontologies and related learning of available tools, standards and related domains. Simon Spero: Amanda: Are there any multiple-choice questions for specific sets Simon Spero: Denise++ Simon Spero: People with CS and DB backgrounds find some things easy Simon Spero: But they have to unlearn others Simon Spero: People with cataloguing background find completely different things trivial/difficult AmandaVizedom: Of the people I have worked with people who were hired as ontologists, there is a considerable range of both formal training and experience. No area of either formal training or experience has been a reliable predictor of possession or lack of requisite skills or knowledge, or ability to learn them. AmandaVizedom: @Simon: Not understanding the question (about m-c ques)... lost context? AmandaVizedom: It sure would be nice if people doing ontology for the semantic web learned about ontology more broadly! RaviSharma: Mike Grunninger: I see a sense to iclude semantic web as technology of interest, RDF OWL reasoning, DL, etc. RaviSharma: Fabian: Please see above for Mike. RexBrooks: My comment on the Semantic Web had nothing to do with interest-drivers. I'm concerned that if we make all uses of technology or technologies, semantic technology or semantic technologies, leaving a reference to Semantic Web out will cause confusion in public perception where they hear "Semantic Web" frequently enough that it could be problematic if they construe our use of semantic technologies to refer to semantic web when there is only some overlap. PavithraKenjige: First of all, Semantic Web has to use Ontologies.. so Ontologies can be used help interoperability.. anonymous1 morphed into PhilKalina PavithraKenjige: Semantic web is registered trade mark and not a concept? Simon Spero: The phrase "Semnatic Web" can be used to lure journalists in, like "Puppies" for children. PavithraKenjige: I would agree with using the word "Semantic web technologies" AmandaVizedom: Perhaps use opportunity to mention Semantic Web in such a way that makes clear what the relationship is to our focus. It is an example of the need -- the rise of sem web comes with a need for well-trained ontologists. Without well-rounded ontologists, sem web projects face danger of making errors that ontology learned about 20 years ago, causing project failure. The same is true of other projects that have grown out of the Sem Web, such as Linked Open Data. RexBrooks: +1 Amanda AmandaVizedom: ... But many semweb projects are currently working without knowledge of ontology principles and history. Even if they want to do otherwise, they do not know how to find people who do have the knowledge. PavithraKenjige: The concept of Semantic Web is used interchangeably with Web 3.0 PavithraKenjige: So I am not sure whey this causes so much of confusion? Simon Spero: http://www.dsoergel.com/cv/B58.html AmandaVizedom: Same is true of Semantic Enterprise, Semantic Interoperability, and other currently recognized areas of activity and buzz. Simon Spero: "SemWeb Proposal for an open, multifunctional, multilingual, system for integrated access to knowledge about concepts and terminology" Gary Berg-Cross: Here is the paragraph from the 2008 communique. Simon Spero: "Expanded version of a paper published in Advances in Knowledge Organization v.5 (1996):" Gary Berg-Cross: "We believe in the promise of semantic technologies based on logic, databases and the Semantic Web, a Web of exposed data and of interpretations of that data (i.e., of semantics), using common standards. Such technologies enable distinguishable, computable, reusable, and sharable meaning of Web and other artifacts, including data, documents, and services. We also believe that making that vision a reality requires additional supporting resources and these resources should be open, extensible, and provide common services over the ontologies. (1GUK)" Simon Spero: Predicative use of copula GeraldRadack: I suggest adding "linked data" as a line under Core Knowledge, item 8. AmandaVizedom: Unrelated comment, triggered by Jim Disbrow's remark: Major mistake ontologists need to avoid: ontology is not made up of language. AmandaVizedom: To be more clear: need to understand the difference between ontological structures and linguistic structures. AmandaVizedom: +1 Deborah re: cross-domain ontology interoperability and reasoning. Add: Ontology modularity and how to use it. Simon Spero: Amanda++ Simon Spero: That's what Special Librarians have to do GeraldRadack: "ontology is not made up of language": But practical ontologies use inguistic information, e.g., names and definitions of classes and properties. AmandaVizedom: Another thing I'd like ontologists to know about: the history (to whatever date) of attempts to 1) turn ontologists into SMEs, 2) turn SMEs into ontologists, 3) find technological ways to minimize the bottleneck and lossiness of SME>Ontologist>Formalism>machine flows,... Simon Spero: ""There is in my opinion no important theoretical difference between natural languages and the artificial languages of logicians; indeed, I consider it possible to comprehend the syntax and semantics of both kinds of language within a single natural and mathematically precise theory." - Montague GeraldRadack: Amanda: Why do newly created ontologies need to know (1) and (2)? That seems more like something project managers should know, so that they bring on properly qualified ontologies. Therefore, I think it should be in the key findings, but not in the body of knowledge for ontologists. AmandaVizedom: Tools, work processes, feedback processes, effective/ineffective elicitation techniques, effective/ineffective validation techniques, importance of in situ SME validation. Simon Spero: Also: http://t2.gstatic.com/images?q=tbn:HYWmFmD2PdPntM:http://www.hoosierwoodindiana.com/Smee1.JPG RaviSharma: Earth, atmosphere, ocean and Natural resources, Energy, Environment, smart grid. for peter Simon Spero: Intolerance++ DeborahMacPherson: A couple changes for Any domain could be an application domain for ontologists. For example, ontologies are used in the following areas: (2AV5) Natural sciences (biology and and biomedical informatics, physics, astronomy, geology) (2AV6) Business (enterprise modeling, enterprise integration, manufacturing systems) and E-Commerce (2AV7) Government (public administration, health system, military, intelligence community, security) (2AV Education (2AV9) Cultural resources (museums, digital libraries) (2AV10) DeborahMacPherson: OOPS the changes are:A couple changes for Any domain could be an application domain for ontologists. For example, ontologies are used in the following areas: (2AV5) Natural sciences (biology and and biomedical informatics, physics, astronomy, geology) (2AV6) Business (enterprise modeling, enterprise integration, manufacturing systems) and E-Commerce (2AV7) Government (public administration, health system, military, intelligence community, security) (2AV Education (2AV9) Cultural resources (museums, digital libraries) (2AV10) (New) Built Environment (Emergency Management, Sustainable Design, Smart Grid) also change are used "in the following areas" to "in, across and between the following areas" Simon Spero: [How long is the communique supposed to end up? How is the target audience? Is it it industry, academia (as industry), academia (as researchers?) RexBrooks: Deborah+1 RaviSharma: Deborah: Great additions and edits please continue to send to Fabian. DeborahMacPherson: Thanks - sent to list - sorry will have to miss F2F tomorrow but will dial in tomorrow RaviSharma: Fabian: Please Include a sentence such as following.....While ontology related learning set is mutifaceted it is also recognized that there are atleast two types of tracks for example formal or ontology science and another related to Ontology engineering, ontology applications, Computational Ontology, related to certifications, degrees, industry or standards oriented areas. AmandaVizedom: @Gerald: Maybe, in theory. But IME the project managers have no idea about any of this. I have come into projects in which the entire strategy was based on the belief that some people (not necessarily ontologically trained)could specialize in eliciting all the relevant knowledge from the SMEs, this could be captured in spreadsheets and handed off to a small group of ontologists who could then formalize it; the result could then accurately describe the content of data services, with sufficient accuracy to support discovery, cross-community use, management based on authoritativeness..... No one who knows the history I mentioned would make such assumptions. When I've brought up examples of the relevant research (e.g., DARPA/RKF), it was brand-new news to all. The managers, in many cases, rely on the tech leads for such knowledge. Simon Spero: This is the site for a course taught by Kristina Spurgin last year: http://www.infomuse.net/520/ Simon Spero: The mind map expands deeper PeterYim: thanks everyone ... see you all tomorrow PeterYim: -- session ended: 5:05pm EDT --