Actions

Ontolog Forum

Revision as of 08:46, 9 January 2016 by imported>KennethBaclawski (Fix PurpleMediaWiki references)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

UoM_Ontology_Standard workshop (Face-to-Face) - Fri 2009-10-30

Topic: Moving the UoM_Ontology_Standard Forward

Workshop Co-chair: Mr. Edward Barkmeyer (NIST), Dr. Frank Olken (NSF) & Mr. Howard Mason (BAE, ISO)

This is a face-to-face workshop for the UoM_Ontology_Standard working group. The session will be focused on getting the work, up to this point, into a draft standard. Other members of the community interested in tracking the progress of this work are welcome to join as observers. Remote participation is supported as well.

  • Location:

National Science Foundation,

Room 1235 (Director's Conference Room),

4201 Wilson Blvd.,

Arlington, VA

Our NSF Host: Dr. Frank Olken

Archives

Session Details

  • Discussions and Q & A:
    • Anyone who wants to speak should raise their hands and/or be recognized by the session chair first, before doing so.
    • Before you speak, please always identify yourself and make sure you can be heard (by both onsite and remote participants)
    • You can type in your questions or comments through the browser based chat session by:
    • or point your browser to: http://webconf.soaphub.org/conf/room/uom_20091030
      • instructions: once you got access to the page, click on the "settings" button, and identify yourself (by modifying the Name field). You can indicate that you want to ask a question verbally by clicking on the "hand" button, and wait for the moderator to call on you; or, type and send your question into the chat window at the bottom of the screen.
    • For remote participants, (when everyone is muted) If you want to speak or have questions or remarks to make, please "raise your hand (virtually)" by click on the "hand button" (lower right) on the chat session page. You may speak when acknowledged by the speaker or the session moderator (again, press "*3" on your phone to unmute). Test your voice and introduce yourself first before proceeding with your remarks, please. (Please remember to click on the "hand button" again (to lower your hand) and press "*2" on your phone to mute yourself after you are done speaking.)
    • thanks to the soaphub.org folks, one can now use a jabber/xmpp client (e.g. gtalk) to join this chatroom. Just add the room as a buddy - (in our case here) <uom_20091030@soaphub.org> ... Handy for mobile devices!
  • Please note that this session may be recorded, and if so, the audio archive is expected to be made available as open content, along with the proceedings of the call to our community membership and the public at-large under our prevailing open IPR policy.

Agenda

Opening & Progress-to-Date ... (i) [ audio ] (1:06:07 ; 7.6MB)

  • 8:30am EDT Welcome, introductions, administrivia - Frank Olken
  • 9:00 goals of the workshop - Howard Mason ... [ slides ]
    • Motivation and market for this work
    • rules for conducting the meeting (if there are many attendees),
    • call for agenda items, adoption of agenda
    • Proposed work plan going forward (preview here ... to be adopted towards end of session) - Joel Bender / Peter P. Yim ... [ slides ] from 2009.10.22

Accomplishments-to-date:

  • 9:30 Agreement on languages - moderator: Ed Barkmeyer
    • proposed is OWL (OWL2 DL), CLIF (owl-ized), and UML class diagrams for presentation/discussion
  • 10:00 Break

Consensus on Scope: ... (ii) [ audio ] (2:07:23 ; 15.0MB)

  • 10:15 Scope issues
    • ref. draft scope statement - Ed Barkmeyer ... [ slides ]
    • units only, or quantities and units models
    • specific named units: SI-only, metric-only, the UCUM set
    • relationship to existing standards - VIM, UCUM, UnitsML, etc.
    • conversion
    • scales
    • measurement and uncertainty
    • specification and tolerance
    • priorities and microtheories
    • other?
  • 12:00-1:00pm EDT - lunch

Moving Ahead to an SDO: ... (iii) [ audio ] (2:15:15 ; 15.0MB)

Content Issues:

  • 1:30 Alignment issues (from base docs) - Pat Hayes
    • basic conceptual concern: what is a measurement unit (a quantity, a magnitude, something completely different)
    • dimensions and dimensional algebras
    • derivation representation
    • conversion representation
    • other?
  • 2:45 break

Workplan & Next Steps:

  • 3:00 next steps
    • adopt a workplan
    • action items, target dates
    • telecon schedule
    • resources
  • 3:30 adjournment

Proceedings

Please refer to the archives above

===IM Chat Transcript captured during the session=== (lightly edited for clarity)

Peter P. Yim: Welcome to the: UoM_Ontology_Standard workshop (Face-to-Face)

- Fri 2009-10-30 (210C) Topic: Moving the UoM_Ontology_Standard Forward

Workshop Co-chair: Mr. Edward Barkmeyer (NIST), Dr. Frank Olken (NSF) & Mr. Howard Mason (BAE, ISO)

Joel Bender: Hello anon!

anonymous1 morphed into Roger Burkhart

anonymous morphed into Silvia Gaio

anonymous morphed into HansPeter_de_Koning

anonymous morphed into Bo Vargas (Raytheon)

Peter P. Yim: 8:50am EDT - session started ...

Peter P. Yim: participants introduced themselves

Peter P. Yim: Frank Olken declared the session open

Peter P. Yim: 8:57am - Howard Mason - presenting our "Goals"

Frank Olken: Frank Olken has jointed the chat room.

Frank Olken: Howard Mason is presenting a brief talk on our goals for the

project.

Frank Olken: Howard mentioned the case of ton - which has multiple

definitions in differents systems of units. Worse, ton has several

different possible dimensionalities: ton as unit of mass, ton as unit of

power (for refrigeration), and ton as unit of energy (as in megatons of

yield for nuclear weapons.

Frank Olken: We are now discussing the base documents.

Frank Olken: VIM is available from BIPM at http://www.bipm.org. It is a

vocabulary for understanding units of measurements. 9:09 AM

Frank Olken: ISO 80000 is a now a successor to ISO 31 (SI units). 9:10 am

Frank Olken: UN/ECE Recommendation 20 is from UN/CEFACT. This is a

recommendation for use in cross border trade. 9:11 am

Frank Olken: QUDT was produced by Top Quadrant, for NASA Ames. 9:12

Chip Masters is now discussing this.

Frank Olken: SWEET is a large ontology created by NASA JPL. Part is

measurement units, in OWL DL based on .... by Unidata. It is in OWL DL.

09:13:00 AM

HansPeter_de_Koning: ISO/IEC 80000 "Quantities and units" will replace

both ISO 31 and IEC 60027. Currently 8 parts are released as

International Standard

Chip Masters: The QUDT draft specification and links to the ontology

files can be found here http://www.qudt.org

Frank Olken: UCUM is being adopted by HL7 and Open Geospatial Information

consortium. Developed by Gunther Schadow (not present).

Frank Olken: There is some argument about UCUM - it only considers units

not quantities. 9:14 am

HansPeter_de_Koning: One of the sources for QUDT was the March 2009

version of the QUDV (Quantities, Units, Dimensions and Values) model for

SysML RTF 1.2, provided by European Space Agency (ESA) to TopQuadrant

anonymous morphed into Nicola Guarino

Frank Olken: UnitsML originally from NIST, now an OASIS project. This is

intended to markup units for xml / html documents. It is an XML schema,

not an ontology.

Frank Olken: UnitsML tries to sort out units, beyond ISO, UCUM.

Frank Olken: Roger Burkhart and HansPeter_de_Koning developed QUDV. They

are on the line. New version is coming soon. Covers units and

dimensions. 9:17 am

Frank Olken: QUDV now has a version as an OWL ontology. Documented on the

OMG wiki site.

Chip Masters: Hans, thanks for clarifying the source SysML source for

QUDT.

Nicola Guarino: Is the shared screen working?

Frank Olken: 9:19am - Nicola, no we do not have the shared screen

working. You need to download slides directly from the web page,

Frank Olken: Correct spelling is HansPeter_de_Koning.

anonymous morphed into Brand Niemann

Frank Olken: Welcome Brand Niemann, we are now discussing the base

documents.

Peter P. Yim: Agenda for the meeting is at:

http://ontolog.cim3.net/cgi-bin/wiki.pl?UoM_Ontology_Standard/Workshop_2

009_10_30#nid22GX

HansPeter_de_Koning: The initial OWL version of SysML QUDV is available

at http://www.omgwiki.org/OMGSysML/doku.php?id=sysml-qudv:qudv_owl

Joel Bender: http://clarkparsia.com/files/pdf/units-owled2008-eu.pdf

Frank Olken: See also Quantities in OWL at http://bit.ly/2wodVR by Bijan

Parsia and Michael Smith, presented at OWLED 2008.

Peter P. Yim: On "other?" document base ... check out MikeDean's input

(http://ontolog.cim3.net/forum/uom-ontology-std/2009-10/msg00108.html#ni

d04) and possibly something VinayChaudri of SRI may be sending us

Frank Olken: 9:27 am - For QUDT see the presentation at

http://bit.ly/47v3gF

Peter P. Yim: now that we realize that QUDV has an OWL ontology available,

and Hans-Peter de Koning has agreed to support us on this effort, our

OWL champions will now comprise of: Rob Raskin, Chip Masters &

HansPeter_de_Koning

Frank Olken: Pat Hayes - CLIF suitable for normative representation,

rather than implementation.

Frank Olken: Ed Barkmeyer notes that we will likely use OWL (OWL Full? OWL

DL?) also.

Frank Olken: Ed Barkmeyer notes the possible use of UML diagrams to help

understand the UoM ontology. 9:32 am

Frank Olken: Ed Barkmeyer notes UML diagrams will not be normative.

Frank Olken: EdBarkemeyer - I am sure we will use OWL 2.0.

Frank Olken: Pat Hayes: we should publish as much as possible in OWL 2.0

DL.

Frank Olken: Pat Hayes: rationale for use of OWL is to get it out to the

world.

Frank Olken: Also use OWL 2.0 Full when necessary.

Steve Ray: Bottom line, the normative version will be in CLIF, with

informative versions available in OWL and UML, right?

Frank Olken: Pat Hayes: The point of the CLIF version is as a reference

for implementors.

Pat Hayes: @Steve: Yes, I think that is basically right.

Frank Olken: Will the anonymous person please change yourself so as to

identify yourself. Click on settings button to do this. .....9:40 am

Frank Olken: Pat Hayes: OWL 2 is definitely better than OWL 1, i.e., more

expressive.

Frank Olken: Pat Hayes: We need to make it clear the users of ontology are

required to implement in CLIF.

anonymous morphed into Mark Rivas

Frank Olken: Howard Mason: CLIF will be normative reference, will also

publish informative version in OWL 2 DL.

Pat Hayes: Frank: NOT required...

Frank Olken: Pat, do you mean that OWL 2 DL is not a required

publication?

Frank Olken: Hans-Peter, are there tools for publication into CLIF, e.g.,

syntax checkers?

Frank Olken: Pat Hayes: yes, there are parsers.

Frank Olken: Ed Barkmeyer: There is a CLIF mail exploder to standards/tool

developers.

Frank Olken: Joel Bender: people will gravitate to the document in the

language they know.

Frank Olken: Howard Mason: In SC4 that would be text files or xml.

Frank Olken: Steve Ray: This will help adoption (multi-lingual versions).

Steve Ray: The advantage is that people can stay in their comfort zones

regarding development environments.

Frank Olken: RogerBurkhardt: QUDT also uses Object Constraint Language.

Peter P. Yim: 9:50am EDT - review / discussion on work-in-progress (draft)

Roger Burkhart: The QUDV model uses the OMG Object Constraint Language

(OCL) in combination with UML class diagrams to express consistency and

derivation rules such as dimensional analysis. The greater

expressibility of Common Logic could be important to express such

internal constraints.

Joel Bender: As a follow up to my comment, I forgot to complete the

thought. When a document is presented in more than one language it

multiplies the amount of work that is needed to keep everything

consistent, and there is a danger that some constraint cannot be

represented in one or more of the languages. This is pretty obvious when

stated, but isn't always followed through very well depending on how

well the committee participants cooperate. Just a note of my personal

anxiety as the process continues.

Frank Olken: We are getting ready to resume the discussion here shortly

-10:21:00

HansPeter_de_Koning: @Joel: I fully agree. Ideally we should have an

automated way of generating alternative informative specifications,

avoiding dependence on human transformations.

Pat Hayes: @Joel: I agree this is an issue we should be aware of. Just

baldly publishing several 'versions' would not do the job.

Peter P. Yim: 10:25am EDT now going into "Scope issues"

Frank Olken: Ed Barkmeyer is speaking on the Scope of the Units of Measure

Ontology. see his slides linked from the meeeting agenda.

Frank Olken: First issue units of measure only, or also quantities. UCUM

has not quantities. NIST believes we need quantities.

Nicolas Rouquette: In the OMG, there is a specification called Query,

Views & Transformations, QVT, which provides support for specifying

mappings of an ontology to/from different representations in, e.g., UML,

OWL, RDF, etc... This approach is what the Ontology Definition Metamodel

(ODM) uses to specify the mappings amongst UML, OWL, RDFS and Topic

Maps.

Frank Olken: Pat Hayes, Peter P. Yim, Frank Olken agree that we need

quantities.

Frank Olken: Gunther Schadow wanted only units - but he is not on the

call.

Frank Olken: Nicola Guarino concurs with units + quantities, also wants

reference frames for coordinates, etc.

Frank Olken: Howard Mason: by the model of quantities you mean to include

dimensionality. Ed Barkmeyer the VIM talks about quantities,

measurements, units. See David Leal's

Frank Olken: See DavidLeal's UML diagrams on his web page.

anonymous morphed into JamieClark

Frank Olken: Ed Barkmeyer - SI is about scalar quantities. Do we restrict

scope to scalar quantities.

Frank Olken: We need to talk about vector quantities to differentiate

between work and torque.

Frank Olken: HansPeter_de_Koning: we have a model of quantities which

allows tensor measurements.

Frank Olken: HansPeter_de_Koning: start with scalars and do tensors

later.

Frank Olken: Nicolas Rouquette: Differentiating between torque and work is

non-trivial.

Frank Olken: Howard Mason: We need to cover scalars, we will need to

extend to vectors and tensors eventually. We should start with scalars,

but not preclude vectors and tensors.

Frank Olken: EdBarkemeyer: simply create base types for vectors but do

not explicate them further.

HansPeter_de_Koning: Terminology: tensor of rank 0 is scalar; rank 1 is

vector; rank 2 is matrix; rank > 2 is higher order tensor

Frank Olken: Pat Hayes: We should be careful about saying that the

ontology defines things.

Joel Bender: What is the chance that some user of this work will pick the

wrong label, or build a derived work, that uses the wrong class? What is

the consequence of picking the wrong one?

Frank Olken: Ed Barkmeyer: We are drifting in substantive issues, not just

scope matters. I want to return to what the scope of the ontology.

Frank Olken: Ed Barkmeyer: We clearly need to address scalars, possibly

vectors, ... tensors.

Frank Olken: Ed Barkmeyer: particular quantity is a property of a specific

physical object.

Frank Olken: Ed Barkmeyer: ref to Pat Hayes, two stick both 30 cm long ==>

these are two different particular quantities.

Frank Olken: Ed Barkmeyer: amount of length of the two sticks is the same

if both sticks are 30 cm. Do we need notion of particular quantities.

Frank Olken: Howard Mason: we do not need particular quantities.

Frank Olken: HansPeter_de_Koning: we only need abstract quantities, not

particular quantities.

Frank Olken: Joe Collins: we need both notions.

Frank Olken: Ed Barkmeyer: quantity kind is sometimes referred as

dimension.

Frank Olken: Pat Hayes: If these particular quantities are understood

strictly, e.g., "mass-of". Two different protons cannot have the same

mass. Are we talking

Frank Olken: ChipMasters's: comment on bullet 2. The degree to which we

need to capture the distinction between scalar, vector, tensor is

dependent on how detailed we want

Frank Olken: to model physical laws.

Frank Olken: JamieClark: A rough consensus on abstract quantities, but we

will need some notion of particular quantities.

Frank Olken: Pat Hayes: The issue we cannot get rid of is that of

particular measurements (with errors uncertainties).

Frank Olken: Pat Hayes: purely philosophical issue of what we want to

ontologize. Particular quantities are useless in the ontology.

Frank Olken: Ed Barkmeyer: Physicists think in terms of particular

quantities.

Frank Olken: HansPeter_de_Koning: we have will have URI for quantities,

quantity kinds (dimensionality).

Frank Olken: Nicola Guarino: Pat Hayes seems to want to get rid of

particular quantities.

HansPeter_de_Koning: To be precise the definitionURI for a kind of

quantity or a unit will refer to the ISO/IEC 80000 normative definition

Frank Olken: Pat Hayes: one can speak about length of meeting without

reference to a particular measurement.

Pat Hayes: Nicola is correct

Frank Olken: Howard Mason: we have identified an issue, the extent we

connect to external standards for quantities.

Frank Olken: Ed Barkmeyer: We certainly need abstract quantities, unclear

about particular quantities.

Frank Olken: Pavithra: quantity is not an object, it is an attribute. It

has types.

Frank Olken: Howard Mason: as designed measurement, as manufactured

measurement.

Nicolas Rouquette: Agree with Pavithra; I think that NicolaGuarino's

question could be stated as follows:

Pat Hayes: I agree with Pavritha also.

Frank Olken: Ed Barkmeyer: kind of quantities = dimensions. The nature of

the thing being measured.

Pat Hayes: Some kinds of Q may be distinguished on other criteria than

dimension.

Frank Olken: Ed Barkmeyer: quantity kinds can be subtyped: length can be

height, depth, width, ....

Frank Olken: I prefer dimensionality to quantity kind.

Frank Olken: Chip Masters: we need both ..., dimensionality presumes a

system of units ....

Nicolas Rouquette: 1) Temperature of a Person: this is a general property

in the sense that a Person is a general concept. 2) Temperature of

Nicola Guarino is a specialization of the property: Temperature of a

Person. 3) Temperature of Nicolas Rouquette is a distinct specialization

of the property: Temperature of a Person. 4) We can then further

specialize the property to narrow the context in which we want to talk

about such quantities as properties of things in some context. 5) A

measurement model (In the sense of VIM) can impose additional

constraints on the context in which we can say that a quantity property

is measurable and then talk about a measurement as another kind of

property about a property quantity which is a property of something.

Frank Olken: HansPeter_de_Koning: dimensionality presuppses choice of

base unit ....

Frank Olken: EdBarkemeyer: count is another quantity kind.

Frank Olken: Ed Barkmeyer: we need to differntiate between quantity kind

and quantity role (length vs. height, width, ...)

HansPeter_de_Koning: According to VIM kind of quantity is NOT the same

as dimension - Dimension of a (kind of) quantity is the product of

powers of base quantities that you have selected for your system of

quantities

Frank Olken: Nicola Guarino: Two different quantities might have same

dimension.

Frank Olken: Howard Mason: quantity kinds and dimension kinds

Frank Olken: Ed Barkmeyer: dimension = quantity kind role

HansPeter_de_Koning: The VIM terms are "kind of quantity" and "quantity

dimension"

Pat Hayes: @Nicolas: "specialization of a quantity" isn't a very useful

notion, as it has no way to be cashed out in any theories of quantity

relations. The temperature of Nicola (at a time, as determined by an act

of measurement) is not a specialization of temperature, it is a *value*

of temperature. It is not a property at all, but an actual temperature.

The property *temperature of person* is formally a set of pairs <x

y>where x is a person and y is a temperature. We have to allow

temperatures in this ('abstract') sense to *exist*, and when we do, they

suffice to say all that we want to say.

Nicolas Rouquette: @PatHayes: "Temperature of Nicola" is a Tensor; this

property is not tied to a particular context. We can specialize this

tensor, e.g., to refer to the "Temperature of Nicola on Oct. 30, 2009"

which isn't a measurement either.

anonymous morphed into NSF-venue

Frank Olken: Dimension seems overloaded. Dimension in physics seems to

mean quantity kind. Dimension in engineering is a role (e.g., height,

width).

Frank Olken: JamieClark: we should defer further discussion to

substantive phase of the project.

Frank Olken: Nicolas Rouquette: We need to tie our concepts to standards,

standard termionologies, e.g., VIM.

Pat Hayes: @Nicolas: OK, you beat me. I have no idea what you are talking

about. HOwever, *temperature of nicola* is certainly not a tensor in

CLIF, OWL or any ontology formalism I know of.

Chip Masters:

http://books.google.com/books?id=pIlCAAAAIAAJ&printsec=frontcover&source=gbs_v2_summary_r&cad=0#v=onepage&q=&f=false

Frank Olken: Chip Masters: dimensionality = systems dimension (dependent

on systems of measure)

Nicolas Rouquette: I said that, ideally, the UOM should really be an

ontology of VIM and nothing else. VIM has the benefit of having been

thoroughly vetted and reviewed in the scientific community for,

literally, hundreds of years.

Joe Collins: "quantity dimension" is well defined, "dimension" is not

Frank Olken: Peter P. Yim: We should stick to the VIM as closely as possible.

Frank Olken: Ed Barkmeyer: Chapter I conflates two notions of quantity

(Ch. I of VIM). VIM was written by physicists not engineer.

Frank Olken: Ed Barkmeyer: We will to model systems of quantities.

Nicolas Rouquette: Ed: could you specifically point to where VIM is

ambiguous or conflicting about the notion of quantity?

Frank Olken: Ed Barkmeyer: Will we model any other systems than SI?

Nicolas Rouquette: QUDV in SysML 1.2 allows you to define your own system

of units, whether it is a subset of SI, a superset, overlaps with SI or

is completely different.

Frank Olken: Ed Barkmeyer: We will model systems of quantities. Will we

model any other systems of measurement within this ontology.

Nicolas Rouquette: Similarly, QUDV in SysML 1.2 allows you to define your

own system of quantities; there is no constraint that says that one has

to use all of ISQ.

Nicolas Rouquette: (ISQ = Int. System of Quantities, which is part of

ISO/IEC 80000)

Frank Olken: Chip Masters: we certainly want to model SI, perhaps other

systems if folks need them. Also possibly use other systems to

illustrate concepts from ontology.

Peter P. Yim: Hans-Peter, you are putting music onto our phone line ...

please do not put your phone on hold

Frank Olken: Some from European Space Agency has put us on hold and is

paying music. Please do not do this.

HansPeter_de_Koning: Apologies! I had a call on my second line...

Frank Olken: EdBarkmeyers: We are discussing systems of quantities (not

yet systems of units).

Peter P. Yim: thank you, Hans-Peter

Pat Hayes: @Hans-Peter: it was very entertaining.

Frank Olken: NicolasRoquette: QUDV can handle multiple systems of

quantities.

Frank Olken: Some systems of quantities (SI) use current as a base

dimension and then charge = current * time. Other systems use charge as

base dimension, and current = charge / time.

Peter P. Yim: @NicolaGuarino - could you document the point you just made on

this chat board, please

Frank Olken: Howard Mason: To what extent we cover other systems of

quantities than SI?

Frank Olken: Chip Masters: Charge = sqrt of force (via Coulomb's Law)

Frank Olken: EdBarkmeyers: Do we cover derived quantities as come

computation over base quantities? Does anyone disagree?

Nicolas Rouquette: I think we need to review the SysML QUDV in the

context of this discussion. We already covered the problems of other

systems of units/quantities and the support for dimensional analysis,

coherence and derivation.

Frank Olken: Chip Masters: I disagree, this would requite an ontology of

operators.

Frank Olken: Frank Olken: I agree with Ed.

Frank Olken: Pat Hayes: We could have an incomplete model of derivations.

Frank Olken: HansPeter_de_Koning: Derivations are simple.

Frank Olken: NicolasRoquette: Both VIM and QUDV include derivations of

derived units.

Frank Olken: NicolasRoquette: VIM and QUDV differentiate between quantity

kinds and dimenisonality.

Nicolas Rouquette: @Frank: My name is spelled Rouquette, not Roquette.

Frank Olken: Howard Mason: Derived units are within scope. Details to be

determined.

Frank Olken: Pavithra: record system of units explicitly.

HansPeter_de_Koning: The SysML QUDV contains a full OCL algorithm that

specifies how to automatically derive the quantity dimension for any

(kind of) quantity that is defined within a system of quantities. The

system of quantities defines its base quantities. One individual system

of quantities can represent the ISQ (International System of

Quantities).

Frank Olken: Ed Barkmeyer: We need to model systems of units explicitly.

Frank Olken: Ed Barkmeyer: Which units go into ontology. Clearly need SI

base units? do we add joules? what about metric prefixes? do we add all

of these derived units? on do we rely on a library of derived units?

Frank Olken: Joe Collins: cgs units are not part of SI. Include SI named

units, metric prefixes?

Frank Olken: HansPeter_de_Koning: separate out derived units

Frank Olken: Pat Hayes: I agree - put derived units in a library, not core

ontology.

HansPeter_de_Koning: To be precise I would separate the basic concepts

in a base ontology, then create a second ISQ/SI ontology the imports the

base ontology and adds the most important ISQ/SI quantities and units

Frank Olken: Ed Barkmeyer: there are libraries out there ... Do we assume

that libraries will become published extensions? What about UCUM? But

they are not ontologies ....

Frank Olken: Howard Mason: Scope will include how to do extensions.

Frank Olken: Howard Mason: rule based registry or explicit choice

maintenance authority?

Frank Olken: EdBarkemeyer: Need a model of unit derviation.

Frank Olken: I agree with Ed on unit derivation.

Frank Olken: Pat Hayes: we will not formalize real arithmetic in OWL,

likely not in Common Logic.

Frank Olken: HansPeter_de_Koning: We specify explicilty dimensionsal

analysis in QUDV - including derivation of derived units and quantities.

Frank Olken: Roger Burkhart: We only do simple derivations, expect we will

need to support unit conversion.

Frank Olken: Ed Barkmeyer: Are scales within scope: ratio scales (length,

time) Absolute scales (mass, temperature)?

Frank Olken: Yes, I think so.

Frank Olken: Ed Barkmeyer: What about Rockwell Hardness?

Frank Olken: HansPeter_de_Koning: We included absolute scales in QUDV. It

is essential for many engineering applications.

Frank Olken: Pat Hayes: should ontology include general notion of scales

and situate SI within this framework?

Frank Olken: Nicola Guarino: We will need scales.

Frank Olken: Howard Mason: We cannot avoid scales.

Frank Olken: Nicola Guarino: What about inverse properties such as

resistance and conductance ...

Peter P. Yim: @Nicola - can you give us the name of the book you cited

again, please

Nicola Guarino: Albert Tarantola: Elements for Physics: Quantities,

Qualities, and Intrinsic Theories

Frank Olken: Ed Barkmeyer: we need a general framework for scales.

Ravi Sharma: Nicola Guarino: The answer lies in physics and not in the

units alone, as there could be different ways of measuring conductance

and also resistance and it need not always add to unity as there are

errors in measurements and different micro processes are invloves.

Frank Olken: Ed Barkmeyer: What about nonlinear scales - logarithmic, e.g

sound intensity in decibels.

Frank Olken: Ed Barkmeyer: what about scales such as rockwell hardness.

Frank Olken: Pat Hayes: It is just a partial order.

Frank Olken: Actually, it is a total ordering.

Frank Olken: Ed Barkmeyer: Do we model unit conversions?

Frank Olken: Evan Wallace: Yes, otherwise we are wasting our time.

Frank Olken: HansPeter_de_Koning: We do it, it is simple.

Frank Olken: Steve Ray: Chip Masters was concerned with the mathematical

operators needed.

Ravi Sharma: Evan Wallace: Yes or else there will be no communication

between the different measurement systems.

Frank Olken: chipMasters: We all want unit conversions.

Frank Olken: Chip Masters: we need to model logarithmic functions.

Nicolas Rouquette: Earlier, someone expressed a concern about scoping how

much of "scales" do we want to tackle. I think that focusing first on

the notions of scales for which we can provide value-added reasoning

support (e.g., Hans-Peter mentioned automated conversion) is a good way

to force ourselves to limit the scope to what we can reason about.

Frank Olken: Howard Mason; no dissent on need for units conversion

modelling within the ontology.

Frank Olken: Ed Barkmeyer: do we model particular quantities and

measurements? in first draft?

Ravi Sharma: Chip Masters: The scale of conversion or accuracy does not

matter but affects accuracy of measurement whether linear, log, and

often with limits including singularities.

Frank Olken: Ed Barkmeyer: The physicists are excited about particular

measurements.

Frank Olken: Howard Mason: This threatens to take us into the realms of

other standards.

Frank Olken: Pat Hayes: We need particular quantities for the standard to

be useful. Perhaps we can partially specify this.

Frank Olken: Howard Mason: Include in scope some discussion of measured

values ...

Frank Olken: Ed Barkmeyer: TC 213 does tolerance and uncertainty. Do we

model tolerances? This has significant commercial significance.

Nicola Guarino: I have to go now, bybye everybody. Nice discussion!

Frank Olken: Steve Ray: We need to differentiate measurement uncertainty

and specification tolerance (descriptive, vs. prescriptive).

Frank Olken: Steve Ray: Can we avoid prescriptive notions of tolerance?

Frank Olken: Steve Ray: Tolerances and measurement uncertainty are

separable issues.

Frank Olken: I favor defering issues to tolerances.

Frank Olken: Evan Wallace: This is an artificial distinction. this is

problematic for commerce.

Frank Olken: HansPeter_de_Koning: We should limit discuss of tolerance.

Frank Olken: Howard Mason: We will not consider tolerances in first

release.

Frank Olken: No consensus about measurement uncertainty.

Frank Olken: Ed Barkmeyer will not include either measurement uncertainty

or tolerance within first version.

Frank Olken: EdBarkemeyer: Should we divide the ontology into modules?

Yes? Unclear, how?

Pat Hayes: For the record, tolerance is easy, but uncertainty and

probability is new territory for formalization in ontology languages, so

we risk being too ambitious.

Frank Olken: OWL 2 is working on modularization. CLIF?

Frank Olken: Pat Hayes: Yes, CLIF has modularization scheme, including

restricting scope of existential quantities.

Frank Olken: Howard Mason: We want modularization.

Frank Olken: Pat Hayes: module import is transtive

Frank Olken: Ed Barkmeyer: Explicit microtheories - possibly inconsistent

with each other?

Frank Olken: Pat Hayes: Names can mean different things within

metatheories. This is risky within a standard.

Frank Olken: Pat Hayes: You can get something similar by subscripting

names with contexts in common logic.

Frank Olken: Pat Hayes: We do not want full CYC microtheories - e.g.,

multiple meanings for names within microtheories.

Pat Hayes: metatheory//microtheory

Frank Olken: EdBarkmeyers: Should we include guidance for how to do

extensions to this standard? Yes !!!

Pat Hayes: Yes, as far as we can.

Frank Olken: Ed Barkmeyer: relationship to other standards efforts? UCUM?

UnitsML? Other?

Pat Hayes: HOw can a mere mortal like myself get hold of a readable copy

of iso 80000 ?

Nicolas Rouquette: You can find VIM and various publications related to

ISO/IEC 80000 here:

Nicolas Rouquette: http://www.bipm.org/en/publications/

Frank Olken: HansPeter_de_Koning: Use NIST document on treatment of

English units as example for how to do extensions.

Frank Olken: Jerry Smith: When we get to 80 percent, publish!

Frank Olken: Nicolas Rouquette: We need to know where the repository will

how, implications for intellectual property.

HansPeter_de_Koning: NIST document SP 811

http://physics.nist.gov/Pubs/SP811/

Frank Olken: Howard Mason: This closes scope discussion.

Peter P. Yim: 12:29pm - lunch break

Joel Bender: Is the conference line staying open?

Frank Olken: We will resume at 1:15 PM, 17:15 PM UK, 18:15 Europeans time

- i.e., 45 minutes.

Peter P. Yim: 1:19pm - back in session

Pat Hayes: Ed: do we have a referenceable summary of what we agreed this

morning?

Joel Bender: http://docutils.sourceforge.net/rst.html

Frank Olken: We have resumed the meeting.

Frank Olken: We are now discussing the standardization strategy.

Frank Olken: It appears that we use OASIS as the base Standards

Development Organization.

Frank Olken: OASIS will accommodate a variety of file formats including

xml, xhtml.

Frank Olken: After OASIS standard would be forwarded to ISO (or possibly

W3C).

JamieClark: Or both; issue of where to submit will be for the committee

once the have a final OASIS Standard product.

JamieClark: Every submission has a time and strategy tax, though, so

that'll be easier to evaluate once underway. In any case, OASIS makes

those submissions.

Frank Olken: Ed Barkmeyer: Unhappy at the prospect of falling into the

clutches of the W3C.

Frank Olken: Pat Hayes: I am guessing that W3C will pass.....

Frank Olken: Howard Mason: We want to avoid going through more than one

standardization process.

Frank Olken: Joel Bender: during the standards development process the

draft documents will be available on the TRAC server ...

Frank Olken: JamieClark: We will need a copy of the standards drafts on

the OASIS server (even the working documents).

Frank Olken: JamieClark: OASIS also has site for email for standards

development.

Frank Olken: JamieClark: Many XML based projects have run aground on XML

tools issues.

Frank Olken: Nicolas Rouqette: How will you coordinate with OMG on QUDV?

I would like to avoid duplication of work with SysML, the creation of

similar but different standards.

Frank Olken: Howard Mason: The committee will establish coordinating

processes

JamieClark: OASIS rules permit use of properly administered outside

tools for hosting functions we don't carry out internally.

Frank Olken: NicolasRoquette: Andrew Watson at OMG is the right person

for coordination of OMG and OASIS work.

Frank Olken: JamieClark: Show the SysML the charter, ask if they want to

be involved with the creation of the UML model for UoM ontology.

JamieClark: Joel: Talk to our Mary McRae, she's the authority on

approval of TC use of outside resources. And, as it happens, a CMS

expert. Mary.mcrae [at] oasis-open.org

Joel Bender: Thank you.

Frank Olken: Roger Burkhart: I do not see problems of coordinating the OMG

and OASIS work.

Frank Olken: Roger Burkhart: I chair the SysML revision group.

Frank Olken: Roger Burkhart: Original SysML submitters gave very

permissive license.

Frank Olken: Nicolas Rouquette: We have used ODM mappings to translate

to/from UML, Owl.

Frank Olken: Howard Mason: No dissent on use of OASIS as the vehicle for

this ontology standard.

Frank Olken: Howard Mason: Let us start discussion of the charter of the

standards group.

Frank Olken: Howard Mason: Proposed name: QUOMOS: Quantity and Units of

Measure Ontology Standard

Frank Olken: Howard Mason: Do we include usage rules in the standard?

Frank Olken: Various: no

Nicolas Rouquette: Ed, are you saying that you don't like SysML? I'm

choked!

Joel Bender: (there is a quite a bit of discussion that is difficult to

hear on the conference call)

Frank Olken: I think we should avoid business rules - as too politically

sensitive.

Frank Olken: Goal is electronic open-access document.

Frank Olken: OASIS: non-asssertion regime. Membership in the TC waives

your rights to content of standard.

Frank Olken: Intended users: development of information models

Frank Olken: Ed Barkmeyer: Also document markup developers

Frank Olken: Also data exchange markup developers.

Frank Olken: Evan: You skipped over a section on "dimensions". We need to

be clear about quantity kind.

Frank Olken: Language for conducting business: English.

Frank Olken: Various standards to coordinate: UnitsML, BIPM, ISO 80000,

Frank Olken: Evan Wallace: I do not see Recommendation 20 on here. I am

concerned that it might be constraining. I am glad to see it omitted

here.

Frank Olken: Do we need a heartbeat - a regular working draft

publication?

anonymous1 morphed into Hajo Rijgersberg

Frank Olken: JamieClark: forward standard to ISO in charter?

Frank Olken: consensus: No.

Frank Olken: Draft title of standard: Quantity and Unit of Measure

Ontology Standard (QUOMOS)

Frank Olken: NIST (is a member of OASIS) and will support the standard.

Frank Olken: Peter P. Yim (is an individual member of OASIS) and will support

the standard development.

Frank Olken: BAE will suport the standard and is a member of OASIS.

Frank Olken: NicolasRoquette (JPL) will support.

Frank Olken: NSF is not a member of OASIS.

Frank Olken: Roger Burkhart (John Deere) is not a member.

Frank Olken: Eurostep may support this.

Frank Olken: OMG is not a member of OASIS.

Frank Olken: NIST, JPL, BAE?, CMU?, Eurostep, LBNL?, --- we need to get

approval of primary OASIS members.

Frank Olken: DOD DISA is a member of OASIS, could also endorse the

standard.

Frank Olken: Schedule for first meeting?

Frank Olken: Ed Barkmeyer: I am estimating mid-January, 2010 for first

meeting. Mostly teleconference.

Nicolas Rouquette: Bye.

Frank Olken: JamieClark: First meeting is likely to be just a

teleconference.

Peter P. Yim: @JamieClark - OASIS should request from ISO a copy of the

latest ISO/IEC 80000 standard for the purpose of this development (to

Mike Smith of ISO) with the understanding that this will be put into a

password protected shared file workspace for this working group

Frank Olken: Howard Mason: OASIS needs to request of Mike Smith a copy of

ISO 80000 for purposes of the ontology std development.

Peter P. Yim: above suggested by Howard Mason

Frank Olken: We will resume in 5 minutes.

Frank Olken: We are reconvening now.

HansPeter_de_Koning: I am back on-line and in the audio conference

Frank Olken: We are trying to schedule a teleconference (perhaps a part

of Ontolog Forum) to discuss QUOMOS project.

Frank Olken: Yim: we will have teleconference to finalize QUOMOS charter

on Nov. 19, 2009. Thursday - see developing session details

at: http://ontolog.cim3.net/cgi-bin/wiki.pl?UoM_Ontology_Standard/ConferenceCall_2009_11_19

Frank Olken: Howard Mason: We need all contributions, and names of

sponsoring persons, organizations within OASIS by Nov. 16, 2009.

Frank Olken: We now need a convenor for the mid-January 2010 - First OASIS QUOMOS TC meeting.

Howard Mason, Ed Barkmeyer, Frank Olken, Peter P. Yim are candidates.

[ subsequent post: 14-Jan-2010 is a Thursday, and could be a candidate date

for that meeting. -PeterYim ]

Frank Olken: We now need a list of deliverables to go into the charter.

Frank Olken: Ed Barkmeyer: A bunch of modules, each in CLIF, derived OWL 2

DL, UML pictures, explanatory English text.

Frank Olken: Modules will be quantities, units, and scales. Optional

units on measurement uncertainty, tolerances.

Frank Olken: I may be able to participate as a representative of OASIS

member LBNL.

Frank Olken: Deliverables: we will produce xxxx initially. We may produce

xxx modules later.

Frank Olken: We will start with SI base, and extension mechanism.

Frank Olken: Ed Barkmeyer: Quantities, Units of Measure, Scales, SI base

units, Derived Units, ... modules

Frank Olken: Maybe also a module called Dimensions.

Frank Olken: A core set of modules covering quantities, units of measure,

scales, SI base units, Derived Units, Dimensions, and Extension

mechanism.

Hajo Rijgersberg: How about measures? And how about quantity kinds? Are

they regarded as separate concepts or as classification of quantities?

Peter P. Yim: (about an hour ago) Hajo Rijgersberg sent in his input about

scope (and more) in a message at:

http://ontolog.cim3.net/forum/uom-ontology-std/2009-10/index.html

Frank Olken: Eachmodule shall include: CLIF, OWL @ DL,

HansPeter_de_Koning: I think my line is muted from your side?

Hajo Rijgersberg: I'm sorry, people, I should have sent it earlier.

Something went wrong with starting time interpretation here...

Joel Bender: (it is very difficult to hear, there is still an office

conversation obliterating the conference)

Frank Olken: Hans Peter, We seem to be getting background noise from your

phone.

HansPeter_de_Koning: I will try to reconnect...

Frank Olken: Howard Mason: The TC will plan to meet every 2 weeks.

Frank Olken: The standard will be known as Quantities and Units of

Measure Ontology Standard (QUOMOS). There was consensus on this.

Frank Olken: Who is editing the charter? Howard Mason can finish the week

of Nov. 16. Ed Barkmeyer can mark up the wiki in the meantime.

Hajo Rijgersberg: Why restrict the title of the standard to quantities

and units? There is so much more. Doesn't the term "unit" cover what can

be called "the domain of units"?

Pat Hayes: Hajo, I think the title isnt meant to be proscriptive, only a

general indication.

Frank Olken: Dimensions are within scope, just not in the title.

Frank Olken: Any other items of business?

Pat Hayes: However, QUODMOS is kind of cute....

Frank Olken: Howard Mason: We are adjourned.

Peter P. Yim: Ed Barkmeyer suggest we poll everyone on their OASIS membership

status. Peter to put request on the uom mailing list ... we don't want

to lose anyone!

Peter P. Yim: Great session ... thank you everyone ... audio recording and

chat transcript will be posted tomorrow.

Peter P. Yim: Appreciations to Frank Olken and NSF for hosting us today!

-- session ended: 2009.10.30 - 15:30 EDT --

  • Further Questions & Follow-up: - please post them to the [ uom-ontology-std ] listserv
    • if you are already subscribed, post to <uom-ontology-std [at] ontolog.cim3.net>
    • (if you are not yet subscribed) you may subscribe yourself to the [ uom-ontology-std ] listserv listserv, by sending a blank email to <uom-ontology-std-join [at] ontolog.cim3.net> from your subscribing email address, and then follow the instructions you receive back from the mailing list system.
  • Mark your calendars and come join us at the next virtual session for this working group - UoM_Ontology_Standard/ConferenceCall_2009_11_19
  • we shall look forward to those who are interested to support this effort to consider joining us at the developing OASIS QUOMOS TC - watch out for the announcement of its formation from both OASIS and ONTOLOG

Audio Recording of the Workshop Sessions

  • To download the recorded segments of the workshop, click on the respective audio link under the archives section
    • individual recording segment are linked to from each of the agenda items above
    • the playback of the audio files require the proper setup, and an MP3 compatible player on your computer.
  • Conference Date and Time: 30-Oct-2009 8:50am~3:30pm EDT
  • Total Duration of Recordings: 5 Hour 29 Minutes - (i) 1:06:07 ; (ii) 2:07:23 ; (iii) 2:15:15
  • Total Recording File Size: 37.6 MB (in mp3 format) - (i) 7.6; (ii) 15.0; (iii) 15.0 MB
  • suggestions:
    • its best that you listen to the session while having the respective slide presentations (when available) opened on your desktop in front of you. You'll be prompted to advance slides by the speaker.
  • ... *my apologies for the "echos" in quite a portion of the audio recording ... I was away from my normal recording equipment and settings, and evidently, still has issues doing it right while on the road. Anyhow, I am glad the audio archive is still quite intelligible! =ppy/2009.11.01

Resources


For the records ...

How To Join Remotely (while the session is in progress)

Attendees

On-site Participants

Remote Participants