SUO: Re: Lifecycle Integration Schema -- Matthew West
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
LIS. Discussion Note 38
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
Matthew,
I began my consideration of your LIS proposal in this way:
| I will make this thread independent of my other questions,
| starting at the top and trying to think about how I would
| formalize the LIS ontology or data model in logical terms,
| asking for clarification as I go. I always try to formalize
| things in layers -- if there is a layer that I can formalize
| in propositional calculus (pure boolean functions and so on),
| then I will do as much as I can on that layer first; if there
| is a layer that I can handle in ordinary naive set theory, then
| I will do that first, putting off questions about the exact axioms
| until they come up, not to mention nonwellfounded set theory, which
| nobody should imagine is going to make life one bit easier for them.
Let me now explain a bit more as to what
I mean by "formalizing in layers" (FIL).
It's much like the grand old strategy of
stepwise refinement, which is very often
described as a top-down form of analysis,
but the "top" is jarring if you consider
that it begins with "zeroth order logic",
(ZOL), even "negative order logic" (NOL),
a term that I will explain later, all of
which would most definitely be ranked as
being at the bottom of the logical order.
Of course, we have long had the problem in our work that
what is called "upper" is also called "foundational", so
maybe it would be better to drop these up and down terms,
if we but knew which way they would fall if we drop them.
At any rate, I have already given one example of what it looks like
when you mine the ZOL course of a given lode of ontological subject
matter to some degree of provisional completion. That is the work
I did on John Sowa's "Top Level Categories", as may be found here:
EEE. Examples! Examples! Examples!
01. http://suo.ieee.org/email/msg09667.html
02. http://suo.ieee.org/email/msg09668.html
03. http://suo.ieee.org/email/msg09670.html
04. http://suo.ieee.org/email/msg09672.html
05. http://suo.ieee.org/email/msg09677.html
06. http://suo.ieee.org/email/msg09689.html
07. http://suo.ieee.org/email/msg09693.html
08. http://suo.ieee.org/email/msg09696.html
09. http://suo.ieee.org/email/msg09699.html
10. http://suo.ieee.org/email/msg09706.html
11. http://suo.ieee.org/email/msg09713.html
12. http://suo.ieee.org/email/msg09719.html
13. http://suo.ieee.org/email/msg09725.html
14. http://suo.ieee.org/email/msg09734.html
15. http://suo.ieee.org/email/msg09738.html
16. http://suo.ieee.org/email/msg09744.html
17. http://suo.ieee.org/email/msg09796.html
18. http://suo.ieee.org/email/msg09841.html
19. http://suo.ieee.org/email/msg09878.html
20. http://suo.ieee.org/email/msg09908.html
21. http://suo.ieee.org/email/msg09941.html
22. http://suo.ieee.org/email/msg09950.html
23. http://suo.ieee.org/email/msg09956.html
24. http://suo.ieee.org/email/msg09959.html
25. http://suo.ieee.org/email/msg10016.html
26. http://suo.ieee.org/email/msg10029.html
27. http://suo.ieee.org/email/msg10045.html
28. http://suo.ieee.org/email/msg10054.html
29. http://suo.ieee.org/email/msg10056.html
30. http://suo.ieee.org/email/msg10068.html
31. http://suo.ieee.org/email/msg10073.html
32. http://suo.ieee.org/email/msg10094.html
33. http://suo.ieee.org/email/msg10117.html
34. http://suo.ieee.org/email/msg10254.html
35. http://suo.ieee.org/email/msg10261.html
36. http://suo.ieee.org/email/msg10296.html
37. http://suo.ieee.org/email/msg10336.html
38. http://suo.ieee.org/email/msg10342.html
By way of constant reference to its SUO-work-relatedness, many will
recall that John's TLC was absorbed whole hog into the folds of the
"Suggested Upper Merged Ontology" (SUMO) from Teknowledge & Company.
My initial translation of TLC into the ZOL formalism that I know as
"Cactus Language" (CL_0, or CL Sub Zero), took me precisely one day
from the time that Jim Schoening first assigned the task to convert
various ontologies into sundry languages for the sake of comparison:
WLTU. What Language To Use
01. http://suo.ieee.org/email/msg01949.html
02. http://suo.ieee.org/email/msg01956.html
I apologize for repeating things that many people here already know,
but there is persistent tendency for other people here to criticize
the work-relevance of information that they never even plan to read.
As far as my current effort on the LIS test goes, however, I have only
gotten as far as e-mitting the following tiny bit of ZOL formalization:
( thing ,( abstract_object ),( possible_individual ))
Using the following labels, T = thing, A = abstract_object, P = possible_individual,
this string expression parses into a graphical syntax that abstractly looks like so:
A P
o o
T | |
o--o--o
\ /
\ /
@
Roughly speaking, if taken as an assertion, this says that every <thing>
is either an <abstract_object> or a <possible_individual>, but not both.
At any rate, much less at this rate, I still have a lot of work to do!
After coffee ...
Jon Awbrey
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o