Ed, (01)
I always appreciate factual corrections. I had heard that
comment attributed to "engineers" for a long time, and the
only attribution I every found was to George Box. After some
Googling, I found many attributions to Deming, but none with
a source. (02)
JFS
>> As George Box said, "All models are wrong. Some are useful." (03)
EB
> For the record, Box wrote this in 1979, which is within the memory of
> the IT world. But mathematical models predate the IT world, and the
> same aphorism was penned by W. Edwards Deming in 1947. (04)
But what is the source? In 1947, Deming was in Japan, where he taught
the Japanese a concept that perfectly fit their philosophy: TQM
(Total Quality Management). I found many citations of Deming for
the above quotation, but none with a date or source. (05)
Among the references was a note you sent to ontolog forum on Sept 18th,
when I was traveling and didn't have much time to catch up on email.
In that note you said, (06)
> So my tenets:
> (1) the world itself is whatever it is.
> (2) our understanding of the world, or any aspect of it, is a 'model'.
> (3) our communication of our understanding is an inferior 'model'.
>
> I'm not sure whether any of those matches John's philosophy. I think I
> may be close to the positions of Pat and Chris. (07)
I completely agree with all three of those tenets. (08)
JFS
>> We could adapt that principle to say "All notations distort
>> the structure of the subject. For various purposes, some are
>> better or more useful approximations than others." (09)
EB
> We could. But it seems to me fairer to say, as semanticists do, that
> the only way to transfer concepts and information from person to person
> is to express them, and no simple means of expression conveys all of the
> concept that the speaker possesses. Even the comprehension of a single
> sentence requires all the trappings of membership in a 'speech community'. (010)
I also agree with this point, but my comment was completely in agreement
with your three tenets. I don't see and didn't intend anything unfair. (011)
EB
> In my experience, there are intrinsically ternary and quatenary verb
> concepts. English grammar supports them with the use of prepositional
> phrases, but listeners use those prepositional phrases in comprehending
> the verb concept. (012)
That's true for verbs like buy and sell, which have 4 slots: (013)
Sam bought the book from Bob for $10.
Bob sold him the book for $10. (014)
But English also has ditransitive verbs that have 3 slots: (015)
Mary painted the wall yellow.
Sue gave the child a present.
Bill considered her generous. (016)
EB
> My point is that we must first distinguish between n-ary database
> relations, which may represent compound statements or relations with
> adverbial modifiers, and conceptually n-ary relations. (017)
Yes. Many relations are constructed by joining two or more
relations with fewer arguments. During the 1970s and '80s,
Ted Codd and others wrote many articles about "normal forms"
with systematic methods for joining and decomposing relations. (018)
> My point is that we must first distinguish between n-ary database
> relations, which may represent compound statements or relations with
> adverbial modifiers, and conceptually n-ary relations. (019)
Yes. There are often very good reasons for defining "views" that
add arguments by joins or delete arguments by projections. (020)
> SQL is a bad example, because it merges n-ary relations, compound
> statements and adverbial modifiers into one grammatical structure. (021)
There are many aspects of SQL that could be much clearer with
a better syntax. But it's not necessary to mark every semantic
feature by a distinct syntactic feature. Sanskrit and Russian
have lots of grammatical markers, but Chinese has almost none. (022)
> So we have a model for what might be done as an RDF extension,
> in much the same way that CGL is a CLIF extension. But we would
> still have to get the target community to agree to understand
> the chosen conventions. (023)
Unfortunately, RDF became a W3C recommendation long before anybody
had a clear idea of what kind of applications it might be useful for. (024)
None of the biggest web companies use RDF. Google will index anything,
but they use JSON for their apps. Twitter, Facebook, and LinkedIn have
huge graph databases, but none of them use RDF. Twitter, for example,
adds billions of graphs per day to FlockDB. (Each graph is much bigger
than a triple.) RDF is too bloated and clumsy to scale to those levels. (025)
IBM doesn't use RDF for Watson. They used XML Schema to define UIMA
(Unstructured Information Management Architecture). They contributed
UIMA and supporting software to OpenNLP, which is an Apache project. (026)
> I am reminded of an observation by Sjir Nijssen that, if you give
> modelers different constructs for similar concepts, half of them
> will use the wrong one half the time. (027)
Yes indeed. That is why it is essential to have a clear idea of the
semantics and pragmatics before freezing the syntax. The first version
of GML was defined in 1969. There was 30 years of experience in using
the *ML family for marking up documents before it was adapted to HTML
for marking up web pages. (028)
But there was zero experience in using that family for knowledge
representation before it became a so-called "standard". (029)
John (030)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (031)
|