Joe Collins wrote:
> I do think that we must deal with representing uncertainty in measurement of
>quantities.
> I bring this up because Pat Hayes wrote:
>
>
>> I would strongly urge that we NOT do this, whatever the SI standard says;
>>
>
> From an information science perspective, all decision processes have error
> rates, i.e., uncertainties. Making measurements involves making decisions:
>the
> metre-stick may be continuous, but the increment marks are discrete and I
>have to decide what value to choose.
>
> From a scientific perspective, when we talk without any notion of
>uncertainty
> about measurements of properties that are not observably discrete, we are
> arguably talking nonsense. Even the counting of discrete properties is
>subject to error.
>
> While it might not be common, everyday practice to explicitly refer to
> uncertainty, it is usually tacitly assumed.
> (01)
I think you and I are largely in agreement. The problems are: (02)
(1) uncertainties proceed from measurement mechanisms and measurement
rollups. Trying to define 'uncertainty' properly in a formal ontology
becomes very involved. That is why the GUM was published -- to explain
to scientists and engineers how to think about uncertainty and how to
write down what is relevant to the measurements they made. (03)
(2) Yes, there is a demand for expression of uncertainty. If we go only
one step down that path, we will have a model that has a nearly
undefined slot for "uncertainty", which means that any two uses of it
are likely to be semantically incomparable. If we go further, it will
take a few years to get a correct ontology in the area. (04)
(3) Many commercial uses of quantity values implicitly or explicitly
involve tolerances, and the tolerances determine the requirements on the
the actual measurement. So in many commercial uses, the concept of
'tolerance' is the dominant concept and it affects the uncertainties in
interesting ways. (05)
(4) We don't want to confuse tolerance with uncertainty in the model. A
specified quantity value has tolerance (and usually ignores uncertainty,
sometimes with serious consequences); a measured quantity value has only
uncertainty. An abstract ideal 'quantity magnitude' has neither. It is
the use or source of the quantity value that acquires the tolerance or
uncertainty property. (06)
So satisfying all the would-be users with a consistent, useful and
rigorous formal ontology will be a longer and more complex task, and may
produce an ontology that none of the intended customers understands.
This is why John Sowa suggests we think about stopping with a minimal
ontology, and let these diverse usage communities produce the
appropriate microtheories. (07)
> For example, when I ask for 1 kg of bologna at the grocery store, I'm pretty
>sure they assume I don't mean 1.00 kg or 1.000 kg or 2 kg.
> (08)
But the grocer's scale reads 1.0, 1.1, etc. and the needle is between
1.0 and 1.1. What does that mean? Is it OK if he presumes that it is
accurate enough to satisfy both of you that he is honestly selling you 1
kg? Once you start adding 'uncertainty' ideas, we bring in the question
of when the scale was last calibrated. Neither of you wants to think
about that. (09)
And when you asked for the kilo, what tolerance did you assume? 1 kg
+/- 50 g? or did you mean at least 1 kg; 956g isn't really acceptable? (010)
And when a scientist states the measured residual mass of a bolt as 1.00
kg after chemical cleaning, what does he mean? And if he states the
uncertainty as ".0048 kg", what does that mean if you are measuring
rates of corrosion? (011)
An ontology that allows the attachment of uncertainties by another
ontology developed by people who care and understand it doesn't burden
the grocer with concerns he doesn't understand and doesn't want you to
have. Conversely, the scientist doesn't need an ontology that deals
with tolerances and legal fulfillments and ignores uncertainties.
Unfortunately, some engineers need both (and only some of them realize
that). We do not want to take on the ontological resolution of
tolerances meets uncertainties -- there are several communities of
experts that deal with that in their measurement domains.. (012)
> It is certainly standard practice to cite uncertainty in communicating
>measured quantities, and it's pretty well defined in the SI.
> (013)
Yes, in the abstract, 'uncertainty' is well-defined. It is the
expression of uncertainty that is interesting. Presumably we would
include the concept so that uncertainties can be expressed. Would you
be happy with a 'quantity-value has-uncertainty simple-quantity-value'
property (with the basic SI definition = the maximum amount by which
'quantity-value' may differ from the actual value)? What about
conditions, e.g. "at standard temperature and pressure"? (014)
The standard practice of stating uncertainties usually appears in text
that states the measurement mechanisms, or implies the use of some
common measurement processes and equipment, and the conditions that
affect the measurement. So the kind of background for the expression
that is described in the GUM is implicit in the actual statement, but
often explicit in the surrounding text. The simple property above has
none of that context. (Of course, the fact of the statement of an
instance of the property may only appear in an ontological context that
guarantees its proper interpretation. You hope, until some other
ontology links to it.) (015)
I want to make clear that I think this is an issue we need to resolve.
I'm just worried that we are being seduced into thinking we can produce
something this is useful by failing to perceive the slope we have
stepped onto. (016)
-Ed (017)
"We must strive to make things as simple as possible, but no simpler."
- Albert Einstein (018)
--
Edward J. Barkmeyer Email: edbark@xxxxxxxx
National Institute of Standards & Technology
Manufacturing Systems Integration Division
100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528
Gaithersburg, MD 20899-8263 FAX: +1 301-975-4694 (019)
"The opinions expressed above do not reflect consensus of NIST,
and have not been reviewed by any Government authority." (020)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/uom-ontology-std/
Subscribe: mailto:uom-ontology-std-join@xxxxxxxxxxxxxxxx
Config/Unsubscribe: http://ontolog.cim3.net/mailman/listinfo/uom-ontology-std/
Shared Files: http://ontolog.cim3.net/file/work/UoM/
Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?UoM_Ontology_Standard (021)
|