Dear Matthew and Peter, (01)
I think we're mostly in agreement, but I'd like to emphasize
some critical details. (02)
MW> I expect that biologists are using ontologies in much the
> same way as engineers do: as a common language which can be
> used to communicate with with minimum ambiguity. (03)
Yes, but most practicing biologists are basically engineers:
they apply established principles and paradigms to some aspect
of medicine, agriculture, etc. (04)
At the level of pure research, all the words and principles,
even the oldest and most firmly established, are subject to
radical revision and redefinition. For example, the old
textbooks classified fish, amphibians, reptiles, birds, and
mammals at the same level of the tree of life. (05)
But the more recent work in cladistics led to a major overhaul
of the tree at the most basic levels. Now amphibians are
placed on a branch of the fish clade, reptiles are on a
branch of the amphibian clade, and birds and mammals are
on two different branches of the reptile clade. (06)
Most of the lower branches of the tree are still connected
at the same points, and there is little change for most
biological engineers. It is still correct to talk about
fish, amphibians, reptiles, birds, and mammals as coherent
groups. (07)
But there has been a major restructuring of how the trees
are derived -- including the inheritance hierarchy. For
example, we now know that the same gene encodes the plan
for the photoreceptors of the human eye, the eye of a
fruitfly, and the light-sensitive spots of a marine worm. (08)
MW> Agreement as to [an ontology's] use for communication
> within a community is necessary for it to achieve its
> purpose. (09)
Yes, but all such agreements are subject to revision, even
at very basic levels. Engineers don't change the terminology
as rapidly as pure scientists, largely because they apply
aspects of science that are not at the forefront of research. (010)
The latest trains on the New York subway conform to the
mechanical and electrical interfaces of a century ago.
And many programming standards were determined by the size
of the punched card on the 1890 Hollerith machines --
including the fact that today's email handlers wrap the
line at 72 characters, leaving 8 columns for a line id
on an 80-column card. (011)
MW> There are standards at two levels here:
> - standards for ontology development,
> - standard ontologies. (012)
I agree. (013)
But I want to emphasize that those standards are largely at
the low levels of ontologies. The standards for the tracks
and voltage of the NY subway were set before anyone knew
anything about relativity or quantum electrodynamics. (014)
PB> I believe truth to be a relative term in both space and time... (015)
Truth doesn't change -- the universe doesn't change to satisfy
our whims. But what we think we know is constantly changing. (016)
John (017)
_________________________________________________________________
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/ (018)
|