SUO: Re: Missing Ingredients
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
MI. Note 2
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
[Consiliating a couple of previous notes]
JA = Jon Awbrey
JS = Jim Schoening
MA = Murray Altheim
Murray,
I am guessing -- on good information -- that you think my last set
of thoughts on this subject a trifle extreme. I assure that they
are my well-considered judgment after 3-plus years of familiarity,
not only with each of the documents since its inception date, but
with the conduct of each of their proponents in this working group.
http://suo.ieee.org/IFF/
http://ontology.teknowledge.com/
http://www.cyc.com/SUO/opencyc-ontology.txt
http://www.tc184-sc4.org/wg3ndocs/wg3n1328/lifecycle_integration_schema.html
In many ways, however, the sharpness of my last remarks would be blunted
so long as this working group thoroughly acknowledges that these starter
documents are just that -- at the starting gates, somewhat before the gun,
so to speak, and scarcely within sight of the finish line, since I do not
think that the SUO race can be viewed as a 50-yard dash, by any stretch of
the metaphor. Indeed, I have put remarks on record that we really ought to
have 10 or 20 more such "starters" to really get started. Viewing my last
remarks in this light, I would have the sting of their critique directed
not so much at the starter documents, per se, but at the very idea that
the lot of them are anywhere near the condition that anyone should bet
their life on them.
With that clarification in mind, I think that it might be good service
to review the main characteristics of each of these runners, as I see
them from my place in the stands, and then others may say what they
see from their vantage points.
But that will be the work of several days, so I will break for now.
Jon Awbrey
JS: There are many ingredients needed to brew a successful standard.
We have a few ingredients, including some starter documents,
a diverse group of experts, a charter from an accredited
Standards Developing Organization (SDO), and a forum
to conduct work.
JS: I see the key missing ingredient as: 'A growing number of organizations
attempting to reuse any of the starter documents, and willing to submit
suggested changes.'
JS: If we had this ingredient,
JS: 1. The suggested changes would help us improve the documents, and also
build consensus, assuming these individuals joined the SUO WG.
JS: 2. The improvements would matter to at least one organization that is
actually building something that needs the improvement. This is
important, because the quantity of these improvements is hopefully
feasible to handle, whereas, the quantity of 'improvements for the
sake of improvement' is infinite and would prevent completion and
balloting of a document.
JS: 3. As the number of users grows, the document moves closer to a
market-accepted de facto standard. IEEE approval will not be
enough. Some level of market adoption must be achieved.
JS: So, how do get organizations to attempt to reuse any of our documents?
Or, if they are doing so, how do we get them to submit their suggested
changes?
JS: Any thoughts?
JA: Some thoughts:
JA: The missing ingredient in all of the current starter documents,
as would be obvious to any expert that you might consult from
any field of expertise beyond the hothouse flowers of AI/KR,
is any measure of respect for what is already considered to
be absolutely standard undergraduate textbook knowledge in
any of the affected fields, much less any appreciation of
the methods of inquiry that are necessary to continue
acquiring and updating that knowledge.
JA: Unless and until these defects begin to be seriously addressed
by this working group, the SUO starter documents will continue
to be a joke, if not utterly retched brew, to any responsible
professional from any discipline who tries to review them.
MA: My reply alludes to an ongoing discussion I've had with John Sowa
and others not about the logic, semantics or epistemology of the SUO
or other ontological systems, but about technical adoption. Since
that seems to be one of the big issues you mention, I'll reiterate
what I've said to John, and something Peter Becker, Finnegan Southey
and others have echoed, namely, that what we need as implementors
is the rubber-meets-the-road documents. TO be able to see how the
mathematical or logical models relate directly to the real world.
MA: The mathematical models are all necessary and part of what I would
expect people like John, Pat, and others to create (they're the
experts), but it doesn't suit the purposes of implementors. It's one
of the reasons why I asked Jon to help me understand LBase. Hand
waving won't do, nor will demanding that implementors learn to
be proficient in logic. It's a tragedy that we all don't get that
education, but there's not point in telling a software engineer to
do so. If the mountain won't move to Mohammed...
MA: What has been lacking (IMO, as an implementor, markup expert, and
relatively new researcher to this field) is that rubber-meets-the-
road model, and also the pedagogical documentation that would allow
some small proportion of the 719,515 registered sourceforge users to
*understand* what the SUO is and does, why it's important, etc. I
know that Pat has gone to some effort in writing up LBase, and I
am thankful for that. But more is needed.
MA: I'm hoping that I can be somewhat of a bridge between communities,
and I'm also hoping I'm not alone. Not everyone can drop their
career and start a Ph.D. program. One of the first things I did
back in 2000, before we'd even finished the XML Topic Map spec, was
to attempt to translate Cyc into XTM. I had no qualifications, and
very little understanding at the time of what I was doing -- it was
obviously a learning experience. What is needed are the kinds of
documents that would have enabled me at that time to understand
what the heck all this stuff is about, and I found very little
that would help me. I've spent the last two years studying in the
hope that I'll gain enough understanding to be one of those bridges.
MA: What would be helpful would be to look at the kinds of models
and documentation, software APIs, etc. employed within the
organizations you're trying to target, and endeavour to have
that material written. If you're trying to hit the web market,
those 719,515, you'd need UML models, Java APIs, and probably some
XML schemas. I realize this kind of thing is probably beyond the
scope of the SUO project -- my point is that there are two islands,
and we all need some bridge building, in both directions.
MA: Long enough message, time for me to hit the hay. (almost 5am here)
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o