SUO: Keeping This List Focused [response #2]
At 10:15 2001-11-02 -0500, John F. Sowa wrote:
>
> In the past, many of Jon A's messages have been criticized as being
> "off topic" or even "off the wall." But he can write clear prose
> when he sets his mind to it.
>
> In particular, I strongly endorse his recent message (copied below),
> which summarizes concerns that many of us have felt about the way
> that the SUO project is being conducted.
>
> First of all, I want to emphasize that the SUMO developers have done
> some very good work by selecting and organizing a large collection
> of axioms that can serve as an excellent resource for ontology
> development.
>
> But unlike some of the SUMO promoters, I do not consider SUMO to be
> an integrated upper ontology, but merely a very useful catalog of
> axioms that could be used to build ontologies. That does not diminish
> the importance of the SUMO work in any way, since such a catalog is
> very valuable. It should not, however, be considered a candidate
> for a standard "upper ontology".
Yes, you are free to disagree with technical topics. However, the main feature about SUMO (and IFF and the conformance discussion) is that standards wording has been proposed. You should propose your own standards wording if you have ideas for improvement.
You've stated many times that you like a "modular" approach, but you have proposed no standards words ... please make a proposal that we can discuss.
> Jim S. has tried to keep the major list free of "discussions that
> don't directly relate to the work of the SUO WG." But the question
> of what "directly relates" depends on one's interpretation of the
> goals. In particular, I would say that the piling up of axioms
> for SUMO is less directly related to "the work of the SUO WG"
> than the question of how those (or any other) axioms should be
> organized.
What "directly relates" is: proposals that suggest/develop/discuss/improve standards words. We are in an IEEE-SA standards development project; our "output" is a standard -- a "document" (can be organized in many different ways); our "development" is the creation of this "document".
> One very serious issue that must be addressed is the question of
> how the SUO relates to the OpenCyc project. For more info,
> see the FAQ list of OpenCyc:
Assuming there were consensus to include this kind of information, this explanation might be an "informative annex" (or maybe a separate document) ... are you willing to make a proposal for these standards words?
> http://www.opencyc.org/faq1.html
>
> They are promising to release 6,000 concept types (with axioms)
> "very soon now." When that happens, there are very serious questions
> of how and whether those concepts and axioms can be shared, integrated,
> or somehow used with SUMO. With the current monolithic structure of
> both Cyc and SUMO, such sharing is unlikely. But with other ways
> of organizing the ontology, such sharing would be much easier.
>
> Discussion of how the ontology should be organized to facilitate
> sharing are, I believe, central to "the work of the SUO WG."
> There are other issues that are also important, and I agree with
> Jon's note below that their proponents should not be silenced
> or exiled to outer Siberia.
Jon A. is free to have his discussion on the ONT list. As you can see from the ONT list, he is getting very little response to his messages (approximately 98% of the messages on ONT are from Jon A.). However, I'm not interested in having this list choked with messages from Jon A. (as he has done in the past). For example, in the recent E-mails from Jon A. on "Inquiry Driven Ontology Development", no one is responding to the list on that topic ... Jon A. has included private responses, which might give the appearance that people are discussing his work on the list, but largely there has been little discussion of his topics.
John S., I think you've pointed out in the past that if Jon A. could improve the presentation of his work, it might get more attention and response. Currently, his comments are getting virtually no attention on ONT (and minimal attention on SUO, as compared to the number of his postings).
I have responses to Jon A.'s E-mail below ...
> John Sowa
>
> > ¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
> >
> > Jim, All,
> >
> > The IEEE SUO WG is chartered to develop a standard, true.
> >
> > The IEEE SUO WG is not chartered to develop a standard, any standard.
> >
> > The presumption of the charter is that it be a good standard, one that
> > has a chance of being voluntarity adopted by the community of practice.
> >
> > I believe that most members of this group share or espouse the aim
> > that the standard we arrive at in the end should be qualified to be
> > a widely respected and internationally supported ontology standard.
> >
> > Many members of this group have serious concerns about the process
> > that will be needed to fulfill this charter or to achieve this goal.
Any process concerns are being addressed by IEEE. We are waiting for their formal response.
> > In particular, many members of the group have serious reservations
> > about the advisability of narrowing our focus in the way that you
> > are suggesting at this time. The state of consensus in the group
> > about how to proceed simply does not justify any degree of narrow
> > exclusivity at this stage. The idea that discussions of ontology
> > should be marginalized in the way that you recommend strikes me
> > as just a little bit strange.
The discussions on this list are for the purposes of developing the standard. If you have standards wording to propose, then please do so.
> > You and Adam Pease have gotten into the habit of directing people,
> > entirely by personal fiat, to leave the discussion for what you
> > judge or pre-judge to be their lack of focus or their defects
> > of relevance. I believe that this is highly inappropriate.
> > If you have specific reasons for your presumed estimations
> > of relevance then all reasonable people will of course
> > want to hear them.
The question is whether or not postings relate to the development of this standard. Proposals for standards words (or *specific* changes or improvements) are welcome. Other discussion that does not directly relate to the development of this standard should be elsewhere ... Jim S. has suggested that you explain your relationship to the standards development (either the existing documents, or a proposal from you on standards wording).
In my opinion, a very large majority of you postings have nothing to do with standards development ... no standards wording has been suggested. If I've misinterpreted your postings, can you point to specific proposals for standards wording?
> > For my part, I believe that it would be healthy for us, conducting ourselves
> > as reflective and critical practitioners are supposed to do, to open up a
> > discussion of how the heck we go about rationally evaluating qualities
> > like relevance and suitability to a purpose, anyway. The exercise
> > would work, not merely to remedy the evident problems in our group
> > process, but also serve the quality of the prospective standard.
> >
> > I think that I can speak for many people when I say that, when we voted
> > for whatever "working documents" that we voted for, that we did so in
> > the spirit of "not blocking inquiry" on a plurality of many fronts.
> > I do not sense that many of us anticipated that we would be quite
> > so quickly punished for the good deeds of our tolerance, nor
> > quite so fastly chained to these non-blocks of inquiry.
There are no "blocks" to "alternative fronts", the focus has always be the same: developing a standard, and contributions that directly relate to that development, i.e., proposals for specific standards wording (additions, changes, etc.).
You are free to make contributions for specific standards wording (for this project). I encourage you to do so.
-FF
_______________________________________________________________________
Frank Farance, Farance Inc. T: +1 212 486 4700 F: +1 212 759 1605
mailto:frank@farance.com http://farance.com
Standards, products, services for the Global Information Infrastructure