UoM_chat-transcript_20100107g_edited.txt // PeterYim: . Welcome to the Ontolog Panel Session: UoM_Ontology_Standard - getting ready for the OASIS QUOMOS TC transition - Thu 7-Jan-2010 * Session Chair: Mr. HowardMason (BAE; ISO) * Panelists: o Dr. CharlesEhrlich (NIST) -- some historical context on the International Vocabulary of Metrology (VIM), a personal perspective o Mr. EdwardBarkmeyer (NIST) -- Scope of the OASIS QUOMOS TC work o Dr. PatHayes (IHMC) -- UoM_Ontology_Standard_CLIF_draft o Mr. DavidLeal (CAESAR) -- UoM_Ontology_Standard_UML_Model_draft o Mr. HansPeter_de_Koning (ESA/ESTEC) & Dr. JamesMasters (TopQuadrant) -- UoM_Ontology_Standard_OWL_draft . session page: http://ontolog.cim3.net/cgi-bin/wiki.pl?UoM_Ontology_Standard/ConferenceCall_2010_01_07 . anonymous morphed into HowardMason anonymous morphed into JimDisbrow anonymous morphed into MartinWeber anonymous morphed into DavidLeal anonymous1 morphed into Frank Olken anonymous morphed into Chip Masters anonymous morphed into Yavuz Eren Chip Masters: @Peter Yim: I emailed you the slides for my slot. PeterYim: thanks, Chip, glad you are alerting me to them Frank Olken: I do not seem to see slides for Charles Erlich's remarks on VIM. Am I missing something? PeterYim: Charles Erlich is speaking on a personal capacity and will not have slides today SteveRay: For the record, I'd like to thank Chuck for a great, information-packed presentation. EdBarkmeyer: The major link is to http://www.bipm.org PatHayes: Ed just made my point for me. PeterYim: @Frank - Chuck will email us some links that we will post and add the collection of resources [ update: they're in ... see: http://ontolog.cim3.net/cgi-bin/wiki.pl?UoM_Ontology_Standard/ConferenceCall_2010_01_07#nid26DX ] SteveRay: How does the VIM handle multiple human languages? Is English the normative form? SteveRay: When is the deadline to provide feedback to the VIM3 committee? PeterYim: @ChuckErlich - the ulterior motive for some of us is that we will be able to help drive "ontology-based standards" that co-exists with natural language standards, so that standards are interpretable both by humans and machine ... is this a right time (VIM3) to bring it up for the VIM WG's consideration? ... how can we best approach it? EdBarkmeyer: Per Peter's comment: the problem is that there is so much data that human interpretation is giving way to machine interpretation in many decision areas, and that means we must get the machine interpretations to match the established human ones, as distinct from what that programmer/knowledge-engineer came up with. MikeBennett: The issue of non physical quantities like cash currency amounts may be a technical issue but it also has strong implications for scope, e.g. whethe someone like me working on financial ontology needs to be aware of UoM work or not PatHayes: @mike, indeed. EdBarkmeyer: Hans-Peter's question is really also a question of scope for the QUOMOS effort: we have discusssed things like including "currency". PatHayes: re scope, I am also concerned about eg psychological data collection and processing. MartinWeber: @PatHayes: Yeah I would expect QUOMOS to support definitions of arbitrary proprietary quantities and units, too... HowardMason: Addressing these questions is a great example of the concept raised at the last summit - synergy between ontologists and standardizers MikeBennett: @ Martin & Pat I guess the question of scope is one of context - whether it's about quantification of things at the most primitive possible level, or whether (as I was starting to think) it's about creating an ontology of (physical) units of measure. Can the context be defined ontologically, obviating the English/French issue? RaviSharma: I Echo Mike Benett PatHayes: @mike on context, there isn't a single accepted technique, but semi-formally we can just be very disciplined with the use of names within the ontology. MikeBennett: @Pat that would make sense, I've just been reworking part of the EDM Council upper ontology to deal with the relationship between Value and Price (Value being a sort of platonic ideal); I can share this in questions. There are two kinds of things to which one can apply the concept of expression: expression of a platonic ideal like economic value, meaning etc., versus expression of things (incuding price again) in multiple mutable ways. PatHayes: mike, would love to see more detail on that. Yes we can be deliberately ambigouous about platonic vs concrete. [[HansPeter_de_Koning]]: @EdBarkmeyer: To be precise on your slide 2 I would replace "Metric prefixes" with "Prefixes (decimal multiples and submultiples, binary multiples)" RaviSharma: Ed: Is it not sufficient for Basic Units to first define the current measurent standard such as (made up example---emission from S to P state of hydrogen atom for frequency standard etc, and then relate freq, similarly for time which is derived unit from fequency or Cs Atom etc., etc and similarly for distance such as meter, etc.etc. these change as physics evolves but are always defined as unique at a given time, thus ontology should relate these basic standards to other standards.? RaviSharma: Ed: Similar to ODM the CLIF to OWL mapping becomes most important for UOM ontology to succeeed? PatHayes: ravi. OWL is embeddable in CLIF, so its more of an extraction than a translation. RaviSharma: Ed: What is the mechanism and which SDO will be reesponsible such that ontology would be used to synchronize and keep these standards bodies efforts (obviously related) current on a reasonable time scale, currently it is voluntary and ad-hoc at best? but going forward? PatHayes: [in reference to Ed's slides/talk ... the UoM_Ontology_Standard will be derived from its normative CLIF representation, and will also be expressed in] owl 2 dl, yes. RaviSharma: Ed: how will you relate ontology efforts such as Sweet in to this? RaviSharma: Pat: Thanks, is it true of all CLIF that there are corresponding OWL statements etc? PatHayes: ravi: no. clif is 'larger' than OWL. RaviSharma: Pat: Does that mean there is a reasonably unique representation for CLIF in OWL? PatHayes: Yes. RaviSharma: Pat: Many thanks HajoRijgersberg: It is no problem incorporatig currencies in the ontology. Start by just defining every currency term (or rather, name). Next, express their definition in terms of other currencies using conditional equations, such as "if calendar time = january 2007 then 1 dollar = 0.50 euro" (or whatever, I don't know the exact rate at this moment). These definitions will/can become independent objects, rather than properties of the particular currencies. RaviSharma: Hajo: you are addressing ontology based on relative values without any standard? HajoRijgersberg: @Ravi: What do you mean with without any standard? I think I don't understand your point... sorry. Could you try to explain? PatHayes: Hajo, i think the point is whether the highlevel concepts apply to them or not. HajoRijgersberg: @Pat: So, probably the high-level concepts (if I understand what you mean) are applicable to currencies too... MikeBennett: @Hajo: The issue of currency is not a technical issue of "how can we make this ontology express that thing", rather it's a set of relationships which can be expressed ontologically, but is related to a more general set of concepts and relationships than those that sit at the top of an ontology of physical units of measure. There is no reason to try and shoe-horn the currency questions into the physical UoM work. MikeBennett: But conversely, if the upper ontology is expressed right, the place where the "Physical UoM" fits in to those concepts. If it doesn't we won't have expressed the more primitive concepts correctly. PatHayes: Hajo, I think that was the point of bringing them up, yes. Quantities of monetary value, etc.. HajoRijgersberg: @Mike: do I understand you correctly that you don't want currencies in the ontology? Why not? MikeBennett: @Hajo at this point I need to know whether the ontology is to cover non physical units or not, because if it is then we need to include those more primitive concepts I'm alluding to. I'm equally happy if this is a physical ontology which does not include currency. Either way, the part of the ontology that expresses measurements of physical quantity, should not include currency. HajoRijgersberg: @Mike: Indeed, a physical module should not contain currencies. MikeBennett: @Hajo we agree. MartinWeber: It's funny that the currency thing comes up again and again in the scope of people talking about units of measure ... MartinWeber: It's a phenomena that changes your measurement readings over time and space and has a great uncertainty attached to it .. Oh and it's not fixed at all MikeBennett: @Martin: that's correct, but it's not simply a vague thing, it's a different kind of thing. HajoRijgersberg: @Martin: What do you mean with "it's not fixed at all"? Are you maybe referring to something I stated? MartinWeber: @Mike: different kind of thing: exactly, currency looks like a unit but has a wholly different philosophy / mechanics behind HajoRijgersberg: What do you mean, Martin? Please explain. PatHayes: again, Ed has made my point. Ravi: we will not address questions of how units are dtermined in practice. "the meter" is assumed to be a continuing entity, even as the method of determining it changes. MikeBennett: @Martin that's right. It can be ontologised. I've written up some notes on this which I will tidy up and put on the EDM Council Semantics Repository site. I'll post a link to the group when I've done that . MartinWeber: @Hajo: I was referring to there is no definite value a certain currency has. There is no "correct" value for one euro e.g.; you always have to include way more context in your "measurement" of a quantity of currency than you need for physical units. I don't doubt that currency CAN be ontologized, but I doubt it will easily fit into a framework for *physical* units. MikeBennett: @Martin exactly so. MartinWeber: feels sort of "everything looks like a nail if you only wield a hammer". MartinWeber: (trying to express currencies in a framework for phys units) RaviSharma: Hajo: I meant to say that the relationships among currencies depend on time and not on any currency being a "gold" standard, unlike time or distance? HajoRijgersberg: OK. What I mean is a currency is like a unit, but indeed, it has a (time) dependent definition (or conversion) in terms of other currencies. So, that's a kind of a more special definition statement than we are used to with units. Right? MartinWeber: ...time, location, the feelings of people involved in trade, public hysteria, propaganda, ... oh and so many more things... RaviSharma: Martin: you have already clarified what i wanted to for Hojo, thanks. HajoRijgersberg: I gave a time-conditional example of dollar and euro. MikeBennett: @Hajo you are right. There are things which are units which do not form part of an ontology of physical units of measure. So Units is more general than Physical units, just as Quantity is more general than Physical Quantity. HajoRijgersberg: Or I should say: a date-time dependent definition (or conversion rule). MartinWeber: there's different currency exchange markets with differenting exchange rates at the same time, too, only differing by location. Think about arbitrage. So yes, there's (way) more necessary context for currencies than for physical units. It's like measuring in a different universe IMO. HajoRijgersberg: OK, these definions (or conversion rules) should be time and location dependent. Do I understand your point correctly now? Good point. MartinWeber: well, I meantioned time and location, but there's more to look at (like system of market -- what defines the value of a currency? relative to what? etc.) RaviSharma: Hojo:there is another semantics implied: take 2 cases, I take a Euro in Europe, it has a certain purchase value based on european perception of value of Euro. Now take that Euro to US, its value will depend on the exchange rate related perception of Euro and this will only be true for free market curencies which are progressively being regulated and sovereign risks are invloved in calculations. Further conversion to government cotrolled currencies will not result in a commercially correct value?? MartinWeber: @Hajo: I think it's outside the scope although parts of the underlying framework decisions might be reused for expressing currencies. But then again, as Mike has stated already, there seem to be currency ontologies out there... HajoRijgersberg: @Martin: OK, so a definition of a currency may be dependent on many variables. So we're going to have quite complex if statements in these definitions (or rather, conversion rules). No problem in theory, I would say. TomRuss: Even within a single currency, you have the issue of time-dependent variation. The entire issue of using inflation-adjusted values for economic analysis recognizes that a dollar in 1980 is not even the same as a dollar in 2010. HajoRijgersberg: @Tom: That's something that an analysis or economic theory or something should take account of. It's not really something for in a units of measure ontology, right? TomRuss: @Hajo. Right I agree that it doesn't belong in units of measure. I was just remarking that the temporal aspect applies even within a single currency. So it does have different behavior. Meters (more or less) have always been the same length. MartinWeber: ...at least a meter won't change if a war breaks out in some region of the world HajoRijgersberg: @Ravi: looks like these if statements are even more complex than I already thought, right? But in theory, still, no problem, I would think? HajoRijgersberg: @Martin: Please show me such an ontology, because I would be surprised. HajoRijgersberg: @Martin: Watch out, there have been many feet... HajoRijgersberg: At least already three in the Netherlands in the past... RaviSharma: Hojo:at a given time you have only relative values among currencies and we have many websites for lookup, but I bet they differ as do the values defined by central banks? HajoRijgersberg: Yes, probably. But that should all be in such an ontology, ideally. I admit, that's a bunge of work. RaviSharma: Hojo: first the trading communities have to agree in a contract as to which are the valid currency exchanges from their perspectives such as LIBOR rate, etc. Then the relationship map will only work for international trade. It is not clear how the internal (intra-country) values will fluctuate or be related to external value of currency, these are interesting models from eco and trade perspectives? MartinWeber: @Hajo: MikeBenett @ 15:07 EST. Try to get hold of his notes... RaviSharma: Hajo:you are aware of ebXRL type of work I suppose, they have to deal with currency conversions? HajoRijgersberg: @Ravi: So it is quite a challenge to specify all that formally in an ontology. But in fact very important work, I would guess, if there are computer systems that could benefit of that. HajoRijgersberg: No, I don't know ebXRL. I'll check it out. Thanks. [ppy comment: Ravi could be refrrring to XBRL here] MikeBennett: Guys, it's simpler than that. Economic value is a thing (with sub types e.g. value in use, value in exchange). The kind of thing it is, we can call "Quality". Quality is something which can be expressed. The expression of the quantity of Economic Value, is Price (or Valuation). That is itself expressed in various mutable ways (percentage, currency-and-number, heuristic e.g. bond yields). HajoRijgersberg: @Martin: What do you mean? RaviSharma: Hajo: Yes certainly it is a great help to develop such "relative-ontology" of currencies to the world of eCommerce and international trade and is very important if effort is backed by resources. MikeBennett: @Pat - [referring to Pat's slide] if you lose the word "physical" in line 1, I suspect this all works at the higher level we were talking about. EdBarkmeyer: The concern I have with Pat's model is the relationship between scales and quantity values. In VIM 3.25m is a quantity value. If "meter" is a function, does that mean that 'measurement unit' will be a 'function' in CLIF? PeterYim: I believe we might have to defer some of the technical discussion to the future QUOMOS TC and move forward (with this meeting) RaviSharma: Peter, Chip, Howard, Steve and Ed and other speakers, participants many thanks. Ravi HajoRijgersberg: @Ravi: That's great. We should do something with that! MikeBennett: Is the call over? My line dropped. I didn't want to distract with questions of currency other than to know whether I should participate in this work PatHayes: mike, we extended the time. MikeBennett: Thx. dialling back in [[HansPeter_de_Koning]]: Sorry I just lost my audio connection, will try to reconnect ASAP PatHayes: hans-peter, lets take this offline, i think we agree but i would like to know more detail. [[HansPeter_de_Koning]]: I am back in conference PeterYim: @Chip - kindly post a message and the reference link(s) to the [uom-ontology-std] list, per your presentation (just a reminder) EdBarkmeyer: I think Chip is right about the "logical union" of QUDV and QUDT. Chip Masters: @Peter: Yes, I will post everything to uom-ontology-std as soon as we have updated the site. [[HansPeter_de_Koning]]: @Pat: I will send an e-mail on the concepts of free vector and torsor and the relations to scale Howard: for those who plan to be joining the OASIS QUOMOS TC, please indicate MartinWeber: *going to be involved at least to link to UnitsML TC* (I'm a member via NIST) PatHayes: i will be involved, once i set up membership. SteveRay: I am actively working my membership on the TC with OASIS and CMU's primary member. EdBarkmeyer: I will be involved as an observing member until we have NIST management agreement on our role. Frank Olken: I expect to participate in the OASIS QUOMOS TC. The primary LBNL representative to OASIS HajoRijgersberg: @Peter: I'm afraid my organization isn't a member of OASIS. What can I do? PeterYim: @Hajo - that's one of the things we may be discussing, as the next item on the agenda (issues relating to participation) EdBarkmeyer: Hajo: You should talk that over with the OASIS membership folk. There are personal memberships, and you may be a member for some other organization with which you are affiliated by whatever agreement you have with them. HajoRijgersberg: @Ed: I also have a relation with Vrije Universiteit Amsterdam, but they are no member either. HajoRijgersberg: @Ed: Who are that, the OASIS membership folk? JimDisbrow: @Ed - Is the U.S. Department of Energy a member of Oasis? or how would I find out if it is? MartinWeber: @Jim: go to http://oasis-open.org. Click "Members" work yourself through all the sponsors/contributors/liasons... EdBarkmeyer: @Hajo: if you go to the website, www.oasis-open.org, there is a "membership questions" page, and a means of creating an email to the people who can help. MikeBennett: My firm (& client) are not members of OASIS. If the scope is physical measures I won't have much to offer. If I can help offline with these broader conceptual issues I'd be glad to put my input in via existing members. HajoRijgersberg: The organizations I am affiliated to are no members. I'm checking out the website right now for possibilities to become an individual member. JimDisbrow: If the US Goverment is a member, or if I can join via NIST's memership, I'll be involved. The $300 Individual membership fee is an economic barrier. But I'm interested. PatHayes: sounds good. Frank Olken: The primary LBNL representative to OASIS has approved my membership in the TC. I believe we are down to paper shuffling and my creation of login/password with OASIS. HajoRijgersberg: $300? Yes, that's a barrier. EdBarkmeyer: @JimDisbrow: the OASIS membership is by agency. But we can talk about other relationships. PeterYim: those having issues with OASIS membership could best try to get it resolved by emailing "Scott McGrath" with a copy to Howard and Peter PeterYim: @Hajo - let's address your issues offline (again, please email Scott cc Howard and Peter) SteveRay: Great meeting. Nice job of chairing, Howard. Thanks very much. PatHayes: i have to leave now. bye all. PeterYim: @Hans-Peter - let's address your issues offline too (again, please email Scott cc Howard and Peter) ... important, we need you as one of the OWL editors HajoRijgersberg: Thanks everyone! PeterYim: great meeting ... thank you all! JimDisbrow: @Frank - who is the LBNL representative? PeterYim: @Jim - not sure if Frank is still online ... the LBNL rep could be Mary Ann Piette JimDisbrow: @Peter - thanks PeterYim: @Jim - suggest you ping Frank offline just to make sure. Thanks for joining us today. PeterYim: bye, everyone ... Thanks again! - session ended 12:57pm PST - //