SUO: *Date 20 Feb 2002
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
JA = Jon Awbrey
MP = Mike Pool
JA: http://suo.ieee.org/email/msg07864.html
MP: I admit to a little confusion concerning the point you're
trying to make in your last few notes although I've found
the notes interesting and thought provoking.
MP: Are you simply observing how hard it is to explain
how humans communicate and what a difficult problem
"semantic interoperability" is? I think we can all
agree that that is the case.
Partly I'm wrestling with a very old problem, one that I recognized
immediately under these new disguises, and so, yes, the wide end of
the hopper is very wide indeed, drawing on all of the experiences,
positive and negative, that I've ever had in trying to understand
communication, information, and the styles meaning clarification
known as "effective", "operational", or "pragmatic" definition.
But I'm also really trying the best that I can to bring things
to focus on the intelligent-dynamical-systems-engineering types
of tasks that I'm currently dedicated to furthering in practice.
So you could say that I'm really just asking a nitty-gritty technical
question about the "minimal adequate data structures" that are needed
to support the levels of aptly appreciative, comprehension exhibiting
interpretive performances that the ilk of human kindness accomplishes
on a mundane recurring basis.
MP: In fact, if you're arguing that creating a standard ontology
won't suffice to solve these problems, I wholeheartedly agree.
I do not expect to be able to create any respectable ontology,
upper, middle, or lower, without appreciable software support,
and of a sort that we don't yet have available. I think that
I came to this conclusion during my first month in this group:
| Since all of our useful categories and concepts arise from common experience
| and customary practice, as sharpened up by logical analysis and scientific
| method, in effect, at the end of a process of inquiry, what kind of evidence
| would lead us to imagine that a novel term is needed, and how can we build
| this capacity for what the pragmaticians call the "abductive creativity"
| of a language, formal or natural, into the system? Even if we do not
| ask the system to "prove that it is creative", we may want, at least,
| to leave room for novelty on the part of the system's human users,
| with perhaps a little support for marshalling and evaluating the
| requisite type of evidence that a new term is needed.
|
| http://suo.ieee.org/email/msg00648.html
| Wouldn't it be better to deliver a "maintainable body", in other words,
| to build in a bit of automation for this task -- measures of structural
| distress, means of structural amendment, and so on -- to help with the
| almost inevitable reconstruction, re-instruction, and restructuring of
| the ontology, whether major, middle, or minor in scope or in level?
|
| http://suo.ieee.org/email/msg00652.html
But, of course, this conclusion was really being transferred
from earlier experiences with similar sorts of tasks, namely,
trying to build logic-based systems for qualitative analysis
applications in health sciences research and education areas.
There is just no reason to think that hand-crafted knowledge
bases can support practical work in developed fields of this
order, or any other established real practice that I can see.
MP: But isn't this a bit of a strawman? Are people claiming that
instituting a standard ontology will solve all interoperability
problems and further assuming that it is exactly this hoped for
standard upper ontology that reflects the conceptual framework
underlying all human thought and interaction? If so, I join
you in heaping ridicule on the claim or at least share some
of your scepticism.
Heap Ahead:
| 5. Semiotics
|
| The discussion relating to semiotics has been less controversial that
| the other areas discussed in this paper, but it is worth summarizing
| here, because it has resulted in a well defined set of concepts that
| are not available in any other ontology, as far as the authors of
| this paper are aware. This area incited less controversy after
| the group developed some concrete examples of the relevance of
| semiotics concepts to reasoning in intellectual property
| issues and media.
|
| A set of examples has motivated development
| of this section of the SUMO ontology:
|
| · An edition of the printed play Hamlet
| · A copy of an edition of Hamlet
| · A performance of Hamlet
| · A performance of Hamlet captured on video
| · The timeless informational content of the play Hamlet
|
| Cf: "IEEE Standard Upper Ontology: A Progress Report" [sic]
| By: Adam Pease, Ian Niles, Teknowledge
Aside from the "Cargo Cult Semiotics" quality of this entire statement
and the gross inaccuracies of reporting on Working Group activities --
what could be controversial if you ignore all informed opinion? --
we have this precious gem:
"The timeless informational content of the play Hamlet". [sic joke].
Key phrase: "as far as the authors of this paper are aware".
That pretty much says it all.
MP: However, my ambitions for a standard upper ontology are not nearly
as lofty. I see such an ontology simply as a potentially useful
resource that will facilitate (not fully achieve) interoperability
exactly because many people have agreed to use it as the basis for
their ontologies, not because it provides some insight into cognition.
I don't think it will explain how humans think or communicate nor do
I see it as the holy grail of artificial intelligence.
The question about the modular memo is asked in order to
illustrate a particular sort of problem-solving strategy.
To arrive at an operational definition of something like
"effective diambiguation", "interdisciplinary inter-ops",
"transmodular communication", or whatever it may be, try
looking at working examples of what you're talking about,
and then ask what kinds of functions and structures it
would take to iconify them.
MP: Do you have reasons to think that the proposed ontology(ies) are
unsuitable given these less grandiose plans? I'm very interested
in the deeper questions you allude to but since I was never under
the impression that a standard ontology would answer them, I don't
find them particularly damning for the SUO effort.
I think that I answered this above.
In my opinion, our current proposals
are vastly different in both quality
and scope, but even the most valiant
of efforts along the lines of this
homespun à priori axiomatic vein
is not likely to get even as far
as Plato and Aristotle did.
But it's late, and I'll have to say more about this another time.
Jon Awbrey
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤