Re: SUO: ELP's summary of MRW's standards experience

["Pierre Grenon" <pierre.grenon@ifomis.uni-leipzig.de>]

Hi John,

This one is a good one! One of the most amazing of your messages. You probably
have broken a few hearts here and there. Now that the psychological effect is
over, I'd be curious to hear about the arguments behind your apparently
gratuitous claims.

(1) "Cyc upper level is a disaster zone"

Can you give the particulars motivating your opinion?

(2) "The waters surrounding Cyc and SUMO are far more shark infested than
Matthew's"

(I'd give you this one: I think there might be sharks in the golf of Mexico,
never heard about sharks in the Manche.)

(3) "I view OpenCyc, SUMO, and other ontologies as resources that we can pick
up and make more general.  There are certainly enough problems in all of them
that would require research to get a good solution."

It would be quicker to start from positive claims and consistent theories which
are not gratuituosly poopooed. It would be good also to realize that a
principle can not be singled out of a theory. 

(4) "I don't expect the SUO upper levels to resemble the current upper levels
of either SUMO or OpenCyc, but we can profit from analyzing their assumptions
and looking for deeper, broader, and more general foundations."

Assuming these are so bad. Do you really think we can profit more from bad
sources than from good ones?

(5) "I believe it is far safer to start with solid foundations, along the lines
that Matthew is recommending, than to start with the upper levels of either
SUMO or Cyc."

Why in the hell have you put forward your motion which was all about raking the
place with IS ontological bits and ends?

(1, 2, 3, 4, 5): Your argument:
There is a bunch of theories which are really no good for nothing. 
We should look for solid foundations.
If we take good theories and we look how they behave when taken together, we'll
surely find out all we need to know for building a good ontology. 
Corollary: If we look at no good theories and see how bad they are when taken
together surely find out all we need to know for building a good ontology. 
Hence, let's take the crap, solid foundations will emerge. 

This is crazy. When you correct a bunch of exams, you don't look at all of them
and try to build some system of shittiness in order to come up with a basis for
correction. You come up with a sufficiently good basis which will allow you to
judge the merit of each exam. If you need to give grades, then you start
comparing these results. 

What you really are going to push this group toward is to use your very own
Sowasian  implicit digestion of Peirce and maybe Whitehead to judge a bunch of
theories which you don't have proper knowledge of nor interest in. You will
ignore positive distinctions when they do not fit your preexisting theoretical
parse, and possibly push for the institutionalization of your idiotic views. 

I take the liberty to feel scandalized because I still regard this group as
potentially productive of a serious SUO. It serves no purpose to use a registry
of ontologies if you really intend this. What is apparently useful is to know
what an ontological theory should comport and which are the issues they
address. There is more than enough litterature on this which is reliable to
bother with the exegesis of a bunch of theories which you do not even regard as
worthy. 

(6) "It might be possible to build a consistent ontology on DOLCE's
foundations, but it is inconsistent with established usage in science,
engineering, and ordinary language (not to mention SUMO and Cyc as well)."

Indeed, my view on the italian ontology is that it is of course one theory
among other. I disagree with their treatment of constituency and other details,
but OTOH it is plain to me that this theory has the merit of being consistent
if not sound and a good modeling paradigm.

Besides, what is of interest in the work of the italians is that they take
DOLCE as _one_ ontology and are very into buidling libraries of ontologies.

Nevertheless, can you exapnd on the alleged inconsistencies with established
usage in:
-science
-engineering 
-natural languague
-SUMO
-Cyc

(7) "One of [Nicola's] assumption is totally inconsistent with Matthew's"
[criterion of identity in his 4d ontology].

Interesting. For one thing, Matthew's principle is either unlcear or eminetly
controversial. I concur it address the same kind of issue. However, I don't
think you can put them on the same table of dissection for the mere reason that
Matthew's principle is aprt of a 4d theory and Guarino's assumption concerns
continuants which are in Dolce 3D entities. In any event, the respective
distinction they address are not on the top of the ramification of alternative
theories of spatiotemporality. In order to do such comparasion meaningfully you
would hvae had to go a long way down in the ontology with consensus at earlier
steps. 

(8) "Even though I think that Nicola made some seriously wrong assumptions, his
discussions can be valuable, even when we choose to insert "not" into his
assumptions." 

Which are the (other?) serioulsy wrong assumptions you mention? 
 
Very eager to learn why you pushed this group to adopt what you regard as a
bunch of crap which should not be used for any other purpose than creating
confusions from which light will arise? And the details of what is so crappy in
all of these ontologies?

Best 
Pierre

> Dear Matthew and Eric,
> 
> I selected some excerpts from your recent exchange,
> about which I'd like to add a few comments:
> 
> >MW: As I see it we are in the early stages of this.
> > And as you know I talk in terms of years rather than months.
> 
> And the 10-year time frame of EPISTLE is probably reasonable.
> 
> >MW: I would see merging things that are accidentally different
> > as important. I also see recognising real differences as
> > important too.
> 
> Yes.  And until those "real differences" can be subsumed by a
> more comprehensive theory, they will have to remain as options.
> Later, we hope, they might be shown to be special cases of
> some GUT (Grand Unified Theory). 
> 
> > [ELP]... But I claim that perhaps the most fundamental and
> > important part of an OO or ontological design is its
> > relational subset. 
> >
> > MW: In fact EXPRESS has its own implementation environment,
> > but in EPISTLE we have as much looked for a relational
> > implementation....
> >
> > MW: We don't have a standard database implementation.
> > It is a point for competitive advantage or different 
> > approaches for different purposes.  The rule is that the
> > physical model should support the conceptual data model.
> 
> When you use logic, the differences between a relational DB
> and an OODB vanish.  An ontology is a *conceptual model*,
> which should be neutral with respect to any *data model*.
> 
> > [ELP]... The problem I see here is an unwillingness to do
> > the hard work of getting down to the bedrock so we can build
> > solid foundations....  That's why I'm suggesting starting
> > with an existing ontology.  
> >
> >MW: Its really not far beneath our feet, and some is exposed.
> > The bits I've found are First Order Logic, Set Theory (ies),
> > Category Theory, Mereology, and you need to add a theory of
> > individuals. Mine is 4 dimensionalism. It says we exist in
> > a space-time manifold, we have temporal as well as spatial
> > parts, and if two things have the same spatiotemporal extent,
> > they are identical.
> 
> This is a good statement of a good starting point.  The main
> problem with existing ontologies is that the main candidates
> don't have solid foundations.  In fact, Doug Lenat is the first
> to admit that the Cyc upper levels are not very important.  I
> agree with him -- primarily because I think that the Cyc upper
> level is a disaster zone and that a better foundation would
> simplify much of the lower levels.
> 
> > [ELP] You want far more that the simple initial foundation
> > than I want.  You want to load it up with axioms.  You want
> > n^2 mappings between existing ontologies.  You don't like
> > the existing candidates, so you seem to want to jump out
> > into untested waters.  You are truly brave.
> >
> >MW: I just want to start with the foundations, and build it
> > up incrementally.
> 
> I completely agree with Matthew on this point.  I believe that
> the waters surrounding Cyc and SUMO are far more shark infested
> than Matthew's.
> 
> >MW: In the mean time we should discuss the foundations for
> > a 3D ontology so we can have a principled and grounded base
> > for that. I think Nicola's paper(s) and Ted Sider's book are
> > good places to start.
> 
> You have to be very careful with Nicola's DOLCE.  One of his
> fundamental assumptions is totally inconsistent with Matthew's
> principle "if two things have the same spatiotemporal extent,
> they are identical."
> 
> As an example, Nicola assumes that a vase and the lump of
> clay that constitutes the vase are two distinct individuals.
> In principle, his ontology can have any number of distinct
> individuals collocated in exactly the same spatiotemporal
> extent.
> 
> It might be possible to build a consistent ontology on DOLCE's
> foundations, but it is inconsistent with established usage in
> science, engineering, and ordinary language (not to mention
> SUMO and Cyc as well).
> 
> > [ELP] I hope we agree that SUO is doing something much more
> > complex and much less proven.
> > 
> > MW: No, I just see it as taking the next incremental step
> > from where we were.
> 
> I agree with Matthew.  I believe it is far safer to start
> with solid foundations, along the lines that Matthew is
> recommending, than to start with the upper levels of either
> SUMO or Cyc.
> 
> >MW: Development is doing something specific - developing a
> > product, or supporting a project. Standards work is
> > developing stuff that can be reused later. Often by picking
> > up experience from projects and making it more general,
> > sometimes by noting a problem and doing research to get
> > to a good solution. We are in the latter category here.
> 
> I agree.  And I view OpenCyc, SUMO, and other ontologies as
> resources that we can pick up and make more general.  There
> are certainly enough problems in all of them that would
> require research to get a good solution.
> 
> >MW: Ted Sider's book and Nicola's paper talk about some of
> > the issues where you need clarity about your position.
> > Unclarity leads to ambiguity which leads to different
> > interpretations being possible, both in extending
> > the ontology and holes in inferencing over it.
> 
> Yes.  Even though I think that Nicola made some seriously
> wrong assumptions, his discussions can be valuable, even
> when we choose to insert "not" into his assumptions.
> 
> > [ELP]... but if you completely threw out any trace of the
> > previous work (including the experienced gained) - never mind.
> >
> >MW: No of course not. I only mean that over a 10 year period,
> > by the time we finished you could not see much if anything
> > of the start point.
> 
> Yes.  I don't expect the SUO upper levels to resemble the
> current upper levels of either SUMO or OpenCyc, but we can
> profit from analyzing their assumptions and looking for deeper,
> broader, and more general foundations.
> 
> >MW: it is really being disciplined about managing an issues
> > log.  During ballot cycles/ review periods issues against
> > the standard can be raised. These must be logged. You then
> > work diligently to resolve those issues, seeking agreement
> > of those who raised them that they have been resolved, or
> > noting that this is too difficult at the present time.
> > This visibility inspires confidence, and means people feel
> > engaged rather than ignored. operated properly it is a very
> > good way to stimulate an improvement culture.
> 
> An excellent summary.  And a registry is an important adjunct
> to such a log:  the log is organized chronologically, and the
> registry is organized hierarchically, but both of them should
> have cross-references to one another.
> 
> John Sowa
> 
-- 
Pierre Grenon
IFOMIS Uni Leipzig
Haertelstr. 16-18
04107 Leipzig
http://people.ifomis.uni-leipzig.de/pierre.grenon/
pgrenon@ifomis.uni-leipzig.de
phone: 49(0)351971672
fax: 49(0)3519716179
?V'Pš0