Dear Matthew and Barry, (01)
MW> I don't see a sharp distinction between science and
> engineering/applied science. Science uses engineering to
> construct and conduct experiments, and engineers do
> science on how their artifacts work. (02)
The basic distinction between pure and applied science is
whether the purpose of the study is to satisfy the scientists'
curiosity or whether the results are intended for some product
or service that people will pay for. (03)
There's also a distinction between experimental and theoretical
scientists. Either kind may be doing pure or applied science,
but the theorists formerly used nothing more than pencil & paper
(or chalk and blackboards). The experimentalists may need a
great deal of engineering to build their equipment. (04)
For example, the enormous equipment for exploring subatomic
particles requires an enormous amount of engineering to
construct. Some of the experiments performed with that
equipment (such as searching for the Higgs boson) is purely
to satisfy curiosity about ultimate nature of matter. It
has no known application. But others might be commercially
valuable to businesses and/or governments. (05)
MW> This [cladistics] is just a matter of labels being
> assigned by perhaps different communities at different
> times to designate different (or the same) classes. (06)
That's more to it that that. Most of the labels are the same
for the same sets of entities. But many of the axioms (or
definitions) have changed. It shows how new discoveries and
theories in science change the definitions and axioms. (07)
JFS>> The latest trains on the New York subway conform to the
>> mechanical and electrical interfaces of a century ago. (08)
MW> Well this is one of those arbitrary things like using
> singular or plural terms. Once you've made the decision
> there is little benefit in changing it. (09)
As science progresses, there may be many benefits from
different options, such as monorails, A/C current instead
of the D/C current of the subways, etc. There are stories
about vacuum-tube computers that were still being used into
the 1980s because the applications were too critical to be
abandoned while switching computers. (010)
JFS>> But I want to emphasize that those standards are largely at
>> the low levels of ontologies. (011)
MW> For example? (012)
Every example I mentioned: subways, computer emails, etc.
Those are very low-level domain-specific categories. (013)
BS> At the cutting edge, yes; 97% of what lies beneath (that the
> earth is a planet, that the sun is a star) remains remarkably
> constant. (014)
Yes, but this indicates several points: (015)
1. A great deal of what we think we know is true to a good
approximation. (016)
2. But nobody knows the limits of application of various
generalizations or what facts we think we know will
turn out to be falsified over unexpected new domains. (017)
3. The old words continue to be used, and their referents may
also be the same. But the definitions are different. (018)
For example, the definitions of planet and star have been
changing. Pluto has been downgraded to 'dwarf planet',
and astronomers have found many objects that blur the
formerly clear distinction between stars and planets.
They discovered a new category of 'brown dwarf', which is too
big to be a planet, big enough to get hot enough to give off
light, but not big enough to initiate nuclear fusion. (019)
BS> International Standard System of Units, anyone?
> No use in science for that, i would imagine. (020)
For engineering, yes. And for the engineering necessary
for experimental science, yes. But for theory, scientists
write equations like E=mc**2 without any consideration of
the units for measuring the quantities. The actual choice
of units is usually irrelevant theoretically. (021)
In fact, many theoreticians will simplify the equations
by choosing a system of coordinates in which the speed of
light c happens to be exactly 1. Then Einstein's famous
equation becomes E=m. (022)
For quantum electrodynamics, they take the next step
of choosing h/(2*pi) = 1, where h is Planck's constant.
That choice plus c=1 is impractical for engineering units
of measure, but it simplifies many equations. (023)
BS> I find it remarkable that so many intelligent voices
> from the computational/engineering side should be arguing
> so vehemently against this experiment (which is not,
> remember, being advanced as a model for OOR). (024)
If the OBO is going to be a limited experiment for a narrow
domain, then I wouldn't have any objection to the organizers
doing anything they pleased. But in that case, it would be
even more important for the OOR to be much less restricted. (025)
John (026)
_________________________________________________________________
Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
Community Files: http://ontolog.cim3.net/file/work/OntologySummit2008/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2008
Community Portal: http://ontolog.cim3.net/ (027)
|