[seek-kr-sms] OBOE discussion: Ferdinand's version
Joshua Madin
madin at nceas.ucsb.edu
Wed Jul 26 14:03:09 PDT 2006
Hi Sergey and Ferdinando,
Thanks for getting this going again. I have a few comments to make.
> Shawn,
>
> It would be nice to give Ferdinando a chance to defend his version
> against our criticism (after tomorrow morning hi will be
> unavailable for comments).
>
> We were discussing the question “can a measurement be considered as
> a kind of observation”, where I do agree with Ferdinando and you
> don’t. For convenience I paste my last unanswered message below. In
> that message I missed an important argument of yours that “sometime
> we can talk about measurement even without mentioning any related
> observation”.
To me, we need two concepts: (1) a concept that points at the thing
we are interested in (e.g., organism, rock, idea, theory, or even
measurement etc.), and (2) a concept that points at the "aspects" of
that thing that we record (e.g., organism height, rock weight, idea
brightness, theory plausibility, measurement dispersion etc.). What
to label these concepts should be a different argument to what we are
trying to accomplish using this framework.
I agree that measurements could be considered observations, but we
need words to describe the two concepts I outlined above. We could
lump ALL these concepts under the label "Observation", but this is
not practical because there is a lot of information we can get from
keeping them separate. For example, one "Thing" can have multiple
"Measurements", and context should apply at the level of the
"Thing" (e.g., organism hasContext theory). And if it doesn't, then
the Observation should be broken down further into holistic
elements. This is very powerful because it makes sampling design
transparent for aligning and merging data, and helps scaling and
grouping if there is sufficient information.
In his quoted phrase above, I think that Shawn was referring to
Measurement Theoretic approaches, and I agree that in these cases the
measurements become the thing that we focus on (if you focus on the
framework, not what we've labeled our owl classes), and then we look
at a whole suit of different "aspect" of these thing, which I would
argue are also measurements with values, units, etc. You can take
measurements of measurements. This fits in well with the current
framework. I agree that the cardinality should be changed as you
suggest: Measurement isMeasurementFor(1:1) Observation.
> If this is your argument, then it is argument against the current
> version of OBOE, rather then against Ferdinand’s version. In the
> current version every measurement is connected to exactly one
> observation via the property isMeasuredCharacteristicOf and
> therefore the situation you mention is not allowed to exist
> (should we set cardinality [0,1] for properties hasSubject and
> isMeasurableCharacteristicOf, we may cover the case you mentioned).
> In Ferdinand’s version Observation is defined as a general category
> and calling a measurement “observation” is a matter of naming
> convention which is a bit different from the convention you use
> when you assert that “we can talk about measurement without
> mentioning related observation”. Within Ferdinand’s ontology it is
> your assertion, which is invalid- just by definition. I do not see
> any inconsistencies between this definition and acceptable usage of
> the two words in English. Setting cardinality [0,1] on property
> hasSubject ( of Observation) will allow to talk about the
> measurements that have no subject- the situation of which you were
> concerned. What else?
>
>
> As I mentioned, I believe that Entities and Traits/Characteristic
> should be reintroduced to Ferdinand’s version and what currently is
> Observable should become Characteristic. (I do not understand how
> entity per se could be observable, but this is the question for the
> current version)
>
>
Sergey, I agree here. The Observable Class is already gone. I
apologize for not committing the recent version -- I'll do it now.
> I think that the idea of CompoundObservation and AtomicObservation
> provides much more natural better mechanism for characterizing
> relation between the columns of the table than what we have in the
> current version . ( I do not ask anymore about characterizing
> relation between the rows. Let spatial and temporal context do this
> job as many of you agree)
I think that CompoundObservation is redundant and serves no real
use. If an "Observation" has more than one "Measurement" then it's a
CompoundObservation, right? Why should we add another layer of
complexity by making the annotator have to say this, when it can just
be easily inferred..?
>
> I also like the idea of ObservationSpace. This neatly generalizes
> numeric measurements and qualitative observation, such as color of
> something, state of health etc. The problem of correct description
> of numeric measurement versus nominal observations was surfaced in
> the messages of Matt, Bob Morris and Jessy. I think that the
> current version of OBOE has much bias towards numeric measurement,
> while Ferdinand’s version has a reasonable generic framework to
> deal with both.
I'm not sure about ObservationSpace, but I agree that the idea of
MeasurementSpace would be useful. Take a look at the new version
that will be in the cvs in about 5 minutes and see what you think.
There has been an attempt to better integrate qualitative/categorical
measurement types.
>
> Most likely you disagree with most of these points. If you could
> explain in some details why you disagree (when it is the case)
> before tomorrow , then perhaps Ferdinando will have chance to reply
> on your critics.
>
>
> Cheers,
>
> Sergey
>
>
>
>
>
> ----------------------------------
>
> Sergey Krivov, Research Assist. Professor,
>
> Computer Science Dept. & Gund Inst. for Ecological Economics,
>
> University of Vermont; 617 Main St. Burlington VT 05405
>
> phone: (802) 656 0380
>
> From: seek-kr-sms-bounces at ecoinformatics.org [mailto:seek-kr-sms-
> bounces at ecoinformatics.org] On Behalf Of Serguei Krivov
> Sent: Thursday, June 22, 2006 3:58 PM
> To: 'Shawn Bowers'
> Cc: seek-kr-sms at ecoinformatics.org
> Subject: Re: [seek-kr-sms] OBOE discussion: current version
>
>
> > I am a bit confused here -- and maybe Josh, you can clarify. My
>
> > impression is that OBOE isn't currently changing in this way.
>
>
> Shawn,
>
> One of the problem discussed yesterday was how to capture both
> normal measurements and nominal ones. Obviously the current version
> of OBOE has bias towards normal measurements and it is not clear if
> it would deal satisfactory with the nominal measures, that is -
> classifications. The majority who wrote on the subject consider
> those qualitative observation as important and worth of our
> attention. Josh was trying to sum up the yesterday discussion in
> one way and I thought that the summary he made is worth of
> discussing. I failed to underscore the word "proposed" while
> talking about Josh's summary. I apologize.
>
>
> > > So we have to remove attributes (roles, properties) measurement
> (0:n),
>
> > > count(0,n) classification (0,1) from Observation.
> Alternatively, we may
>
> > > refuse the idea of having Observation as superclass of
> Classification,
>
> > Count
>
> > > and Measurement.
>
> >
>
> > I wouldn't say "refuse" the idea -- I would say that modeling
> measurement
>
> > (and its various types) as observations has some problems --
>
>
> So far I do not see what those problems are (besides those that I
> have mentioned myself).
>
>
> I think the reason to
>
> > distinguish measurement and observation is because they are actually
>
> > different things. In the general setting, one can have observations
>
> > independently of measurements, and measurements independently of
>
> > observation -- suggesting these are fundamently different entities.
>
>
> I wish we were living in a Plato's world , where concepts have real
> existence. In such world all ontologies would exist in the same
> manner as physical object exist and our job would be just to record
> them on paper the way we do it with biomass, color etc. I think
> that Scholastics did believe in this nice reality, but positivists
> came and spoiled everything...
>
>
> I think there is nothing that makes observation and measurement
> fundamentally different, at least in this world. For a moment I
> thought that may be I am the only one who think so. Then I typed in
> Google " define: measurement" and I see many definitions; and some
> of them at the middle/bottom of the page look as:
>
>
> --An observation that reduces the amount of uncertainty about the
> value of a quantity. In the balanced scorecard, measurements are
> collected for feedback....
>
>
> -- A logical rule for assigning numbers to observations to
> represent the quantity of a trait of characteristic possessed
>
>
> So, there are other people who confuse observation and measurement
> in some way. But still, it would be nice we all vote on the question:
>
>
> Is it linguistically correct to consider Measurement as a kind of
> Observation?
>
>
> Also most of the definitions of measurement given on that page
> consider measurement to be quantitative, that is assigning numbers…
> We need to deal with qualities as well, and packing those
> qualitative observation inside framework of measurement with a
> special unit is a big stretch. (My understanding that here Matt
> agreed with Ferdinando) Obviously, observing sex of an individual
> {male ,female} does not remotely resemble measurement. So, how do
> we then incorporate those qualitative (nominal) classifications
> into OBOE.
>
>
> As Josh’s summary clearly demonstrates, Measurement, Count and
> Classification have common attributes. So, why do not we have a
> common superclass for them. If it is not Observation or
> AtomicObservation, then what it can be?
>
>
> -sergey
>
>
>
>
>
>
>
>
>
>
> >
>
> > Also, we need to be very careful about talking about data
> structure when
>
> > we talk about oboe. Not all day is tabular -- and so it is a
> mistake to
>
> > design oboe to only work with data formatted this way.
>
> >
>
> > -shawn
>
> >
>
> >
>
> > > In the pdf files, where Josh explicatde examples given
>
> > > by Matt and Mark the attribute C (which, as Shawn explained
> to me
>
> > means
>
> > > hasMeasuredCharacteristic) was used to create n-ary relations.
> If we
>
> > refuse
>
> > > measurement(0:n), count(0,n) classification (0,1), then we have to
>
> > refuse
>
> > > hasMeasuredCharacteristic too. If we don't, then Classification
> and
>
> > Count
>
> > > and Measurement will inherit this attribute and we again land
> in the
>
> > same
>
> > > mess. So how do we now create n-ary relation? In Ferdinando's
> proposal
>
> > > Observation can be CompoundObservation and AtomicObservation.
> Later on
>
> > > Ferdinando took back idea of CompoundObservation, but then what
> do we
>
> > have
>
> > > instead, may be we just need a better name?
>
> > >
>
> > > 4. Personally I like the idea of how Ferdinando's proposal takles
>
> > dichotomi
>
> > > between complex (compound) observations and the other things
> such as
>
> > Count,
>
> > > Measurement and Classification. I like his idea of
> ObservationSpace.
>
> > What I
>
> > > do not like is that Observables can be both Entities and
>
> > Characteristics. I
>
> > > think this is an overkill: Entity (thing, object) and
> Characteristic
>
> > > (property) are two fundamental phylosophical categories. Further
>
> > > generalisation over fundamental phylosophical categories
> warants a big
>
> > mess
>
> > > and missunderstanding. I will gaet back later with my comments on
>
> > > Ferdinando's proposal
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > > _____
>
> > >
>
> > > From: seek-kr-sms-bounces at ecoinformatics.org
>
> > > [mailto:seek-kr-sms-bounces at ecoinformatics.org] On Behalf Of
> Joshua
>
> > Madin
>
> > > Sent: Wednesday, June 21, 2006 3:57 PM
>
> > > To: seek-kr-sms at ecoinformatics.org
>
> > > Subject: Re: [seek-kr-sms] OBOE discussion: current version
>
> > >
>
> > >
>
> > >
>
> > > Based on the comments this morning I have redrawn the core of
> oboe for
>
> > > discussion (attached pdf). It seems to me that the unit-at-all-
> cost
>
> > > framework will greatly simplify what we are trying to deliver for
>
> > improving
>
> > > data integration, but, as Ferdinando said, this framework will
> be hard
>
> > to
>
> > > justify and may cause problems down the line. In the attached
> ontology,
>
> > I've
>
> > > tried to divide the different notions of "measurement". All
> this does is
>
> > > restrict the properties that can be used on different types of
>
> > observation.
>
> > >
>
> > >
>
> > >
>
> > > I've also included "hasProcedure" as a properties that acts on
>
> > Observations.
>
> > > This can also act on Measurements due the the subsumption
> hierarchy
>
> > shown in
>
> > > Figure A. I think that this is what Ferdinando meant, but I'm
> not sure.
>
> > >
>
> > >
>
> > >
>
> > > Cheers. BTW: I just received new emails from Ferdinado and Matt
> -- but
>
> > I'll
>
> > > send this anyway.
>
> > >
>
> > > Josh
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > > >1. Observable is either Entity or Characteristic (at the moment).
>
> > > Characteristic has only one subclass Dimension, which defines
> the set of
>
> > > base quantities such as length, weight, etc. , Dimension
> includes only
>
> > > things measured in quantities. Thus at the moment we are missing
>
> > > specification for observations of such characteristics as
> color, smell,
>
> > > taste or anything which is measured in qualitative scale.
>
> > >
>
> > > This is a question that has come up a lot recently and really
> needs to
>
> > be
>
> > > confronted with some good examples. The idea was that nominal
>
> > measurements
>
> > > would just be given unit "name" and a characteristic, such as
> "red".
>
> > This
>
> > > would mean having these characteristics in an extension
> ontology such as
>
> > a
>
> > > "classifiation ontology" (which would plug into OBOE's
> charactersitic).
>
> > >
>
> > > I don't think this is right. Simply, the values of that
> observation come
>
> > > from a finite set of color classes (or instances). Not a
> measurement, if
>
> > we
>
> > > define measurement as comparison with a reference unit (meter
> of tree)
>
> > using
>
> > > an abstract unit for the dimension (meter for length).It is
> ameasurement
>
> > if
>
> > > we define measurement to encompass assigning a class to an
> observable in
>
> > a
>
> > > context as the result of measuring it. I'd rather call it a
>
> > > "Classification", subclass of Observation and siblings of
> Measurement.
>
> > And
>
> > > we could have "Ranking" as subclass of Classification, where
> classes
>
> > must
>
> > > have an ordinal relationship. But stretching the definitionto
> make it
>
> > fit in
>
> > > the unit-at-all-costs framework and giving the characteristic
> the role
>
> > of
>
> > > subsetting the value space doesn't sound right at all. This was
> the
>
> > thought
>
> > > behind proposing an explicit value space.
>
> > >
>
> > > Ordinal measurements may not be as easy to deal with. It might
> work in
>
> > the
>
> > > same way as above, but use the unit "rank". However, the ordinal
>
> > ontology
>
> > > would need to contain constructs that deal with "direction" or
>
> > "magnitude".
>
> > > For example, "high" is distinct from and of greater magnitude than
>
> > "low".
>
> > > This ontology would have to be able to deal with arbitrary
> numbers of
>
> > > levels, similar to the way we dealt with Observation in OBOE
> for coping
>
> > with
>
> > > experimental design. The idea was to remove these kind of
> things (i.e.,
>
> > > characteristics) from the core ontology because the way that
> people want
>
> > to
>
> > > use them are so variable.
>
> > >
>
> > > Similar concerns,plus one:I don't think the ordinal
> relationshipbetween
>
> > > classes such as {high,medium, low} has much of a chance to be
> captured
>
> > in
>
> > > OWL. Nor I think it should be, as you don't do much with it in
> workflows
>
> > > unless it'sa realnumeric scale (whose ordinal properties are
> also not
>
> > > expressed in OWL, so why bother?).If really necessary, we
> couldmake
>
> > > suchclassificationhierarchies subclasses of"Rank" and use
>
> > anumericproperty
>
> > > for ordering suchvalues, but all the logic necessary to do
> anythingwith
>
> > it
>
> > > remains outside OBOE.
>
> > >
>
> > >
>
> > >
>
> > > ----------------------------------------------------------
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > > Our definition, if I remember correctly, was :Observation is a
>
> > statementthat
>
> > > an Observable has been observed. I think more than this is
> going to
>
> > > colorOBOEwith restrictions it does not need to have.By the
> way,we model
>
> > the
>
> > > result of the observation, not the process of the observation,
> and the
>
> > > result is not an event.To annotate a dataset we don't need to know
>
> > anything
>
> > > about the measurement except its results.
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> > >
>
> _______________________________________________
> Seek-kr-sms mailing list
> Seek-kr-sms at ecoinformatics.org
> http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/seek-
> kr-sms
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mercury.nceas.ucsb.edu/ecoinformatics/pipermail/seek-kr-sms/attachments/20060726/b44f1c29/attachment-0001.htm
More information about the Seek-kr-sms
mailing list