[Date Prev] [Date Next] [Thread Prev] [Thread Next] Indexes: Main | Date | Thread | Author

Re: [ba-unrev-talk] further research and thinking related to unrev-ii archive

On Wed, 17 Jul 2002, Paul Fernhout wrote:    (01)

> Interesting paper -- especially coming from a practical experienced
> based analysis of various issues (like mentioning the lack of a standard
> way to reference a reply in email and the impact on analysis). I like
> the idea of tools that provide ways for some work to be automated and
> other to be done or refined by people. By the way, didn't see any
> reference to license issues as another practical problem. :-)    (02)

One can only grind so many axes at once...we chose a particular
one in this paper.    (03)

> Thanks for sharing the paper. It displayed fine in OpenOffice, although
> you might consider RTF, HTML, or a non-proprietary format to maximize
> long term readership.    (04)

That's in progress. There's a home base for the papers as well as
the archives of the online discussion that led to the revision.
It's not clear if that stuff is public so I did not post the
URLs. That issue will be clarified soon.    (05)

> While it was all interesting, I didn't see one point that screamed
> "revelation about the nature of knowledge representation and
> discourse...". Which part was that? Did you mean your paper's insight:
> "In other words, each document reflects at least one concept, and the
> clusters potentially represent not only conceptually related documents,
> but also a category or set of categories for which the concepts have not
> yet been identified. " Or did you mean it more in your writing: "The
> existence of the associative connections is the locus of discovery, not
> the definition of these connections"?    (06)

That's where the archive of the discussion may make some of that
clear. But, in the meantime, if you compare the older version of
the paper with the newer version of the paper, you'll see that we
changed our strategy. The first version simply explains what we
were doing, the second version explains why.    (07)

The why is articulated somewhat in:    (08)

  http://www.bootstrap.org/lists/ba-unrev-talk/0206/msg00097.html    (09)

but really comes down to a fundamental disagreement with the PORT
position that knowledge must be (and can be) represented in
formal structures. We disagree with that. Human discourse is not
formal and the associative connections you can make by having,
talking about and thinking about discourse are, as you point out,
the locus of discovery. The place where learning happens.    (010)

Formal representations, in a sense, are big classification
systems. Classificatory structures model closed worlds. They can
be extremely complex and thus can model an extremely complex
world, but they are, when all is said and done, closed. Thus
_new_ discovery is limited. Rehashing is the mode.    (011)

Human discourse is more like categories: the intension and
extension is flexible. New discoveries can be made because groups
can grow, shrink, or be created.    (012)

> I found it interesting that on a practical basis messages without a MIME
> Content-Type of text/plain content were tossed. Always kind of suspected
> such messages were trouble (and avoid them)... Still, this shows how
> hard it is to try anything that defines a new standard beyond plain text
> (like making links more embedded somehow).    (013)

Indeed.    (014)

> I'm not on line much these days, but thought your great contribution
> deserved further applause.    (015)

Thank you very much. It's very nice to have your comments.    (016)

Chris Dent  <cdent@burningchrome.com>  http://www.burningchrome.com/~cdent/
"If you assume that there is an instinct for freedom, that there are
opportunities to change things, that hope is possible, then hope may be
justified, and a better world may be built. That's your choice.'' N.Chomsky    (017)