SUO: Re: article on the pitfalls of metadata
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
John F. Sowa wrote:
>
> Jon,
>
> I partly agree with you, but with a lot of qualifications:
>
> > Expecting people to mark up their own texts is like expecting
> > people to communicate in parse trees or sentence diagrams.
>
> I would compare it with asking people whether some dubious sentence
> is grammatical or ungrammatical. There are three kinds of people,
> who can be characterized by how they respond to such questions:
>
> 1. The great majority, who feel embarrassed and mumble something
> to the effect that they don't remember anything they were
> taught in high-school English.
>
> 2. People who believe everything they were taught in high-school
> English and are ready to make authoritative pronouncements
> on what is supposed to be "correct".
>
> 3. Linguists who were brainwashed by Chomsky into thinking
> that the ultimate source of data for doing linguistics
> is the "native speaker's intuitions".
>
> People in group #1 may speak good, acceptable English, but they
> are hopeless at characterizing it in any metalevel terminology
> of any sort. They would be just as hopeless at using XML as
> any notation taught in high-school, college, or a workshop
> on text annotation.
>
> People in group #2 tend to condemn large numbers of syntactic
> constructions that have been in common use by good writers
> from Shakespeare to the BBC. If they learned XML, they would
> be delighted to do "authoritative" tagging, but the tags of
> one such "authority" would not be consistent with those of
> any other.
>
> People in group #3 are more tolerant in what they consider
> grammatical, but the rules they "intuit" cover less than 80%
> of the sentences used in well-edited books, journals, and
> newspapers. They would argue indefinitely over the proper
> way to tag most of the sentences in any text.
>
> > It's right up there with thinking that people can translate
> > whatever they are doing with natural languages into plain
> > vanilla literal FOL.
>
> There are English texts that could be translated into FOL,
> but only if they had first been written in FOL before being
> translated into English.
>
> I remember reading about a protestant minister who said that
> whenever he was asked to compose a prayer for some occasion,
> he would first write it in Latin. Then he would translate
> the Latin into English. That process gave the text a style
> that is common in traditional prayers, and it put the audience
> into the proper mood for a religious service. (My observation:
> a somnolent, soporific mood that makes it easier to fall asleep
> while sitting in a hard wooden pew.)
>
> John Sowa
John,
You produce, as usual, a variety of interesting points,
that I believe it would be good to continue connecting
the dots between, but most of them are skew to the line
that I was trying to draw out here, so I making another
try at explaning my point.
Upon my recovery from a series of childhood traumas with the
CyberDyneCorp 3600, Fortran 67, and the 36-hr turnaroundtime,
I was fearfully, tremulously, but strangely attracted back to
orbit of the computer, on the renewed hope that it would help
me do things in a variety intellectual realms that I could not
already do by myself. That was 20-25 years ago and I am losing
hope of seeing much of anything along these lines in my lifetime.
The reasons for that dim prospect all seem to lie with the human
tendency to keep on putting the same old whine in each new bottle
that comes along, instead of making the effort to think about the
true potential of the new medium.
One of the main hangovers is analytic philosophy, that continues to promulgate
the tardive delusion that you can conquer significant problems by dividing them
into just two classes, the "already solved in principal" and the "illegitimate".
One of the more surprising things that I've learned in several discussion groups
over the last three years is that even people who feel quite honestly that they
have overcome this philosophy are in many ways still ruled by its despositions.
Who knows, on a bad day, maybe even yours truly.
Just for the record, I have never intended to suggest that it would be a bad thing
to have really good FOL modelers and provers, indeed, getting somesuch was one of
my earliest conscious goals in computing. Recalculant experience may have given
me one or three non-mainstraum ideas about how this can be achieved in reality,
but that's another story.
The illusion that I am trying to dispell is that we will have solved very many of
our most interesting problems, for those who insist on remembering what they were,
even if we reach this most happy FOL.
The very first item in your bibliography is this:
| John F. Sowa, "Thought clusters in early Greek oral poetry",
| co-authored with Cora A. Sowa, Computers and the Humanities 8,
| pp. 131-146, 1974.
|
| http://users.bestweb.net/~sowa/direct/jfsbio.htm
So I know you know what I'm talking about.
After the FOL,
after botryology,
what goes on in poetry?
Jon Awbrey
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o