From: clayton stirling bartholomew (c.s.bartholomew@worldnet.att.net)
Date: Thu Nov 19 1998 - 14:37:31 EST
This is getting off topic so I will converse in private after this post.
Randy LEEDY wrote:
> I understand your well-founded cautions. My wording "widest possible
> acceptance" implies that there is a wideness of acceptance that would
> be impossible, on other words, that whatever may be done will be
> unacceptable to some. But many if not most of us, I think, would view
> as untenable a syntactical model that does not allow us to say with a
> considerable degree of consistency among ourselves at least that this
> nominative noun is the subject of that verb (distinguishing between
> subject and predicate of a copula is one obvious problem area that
> warrants a provision for several optional taggings) or this adjective
> modifies that noun, etc. I am not yet ready to abandon my conviction
> that a database such as I propose would be a great help to the large
> majority of us even though none of us would recognize it as perfect.
> However, I'm willing to be dissuaded if a strong enough case can be
> put up.
>
Randy,
I am certainly not trying to dissuaded you!!! I have been harping on the need
for this kind of database for several years.
If you can come up with a set of functional categories that are sufficiently
discoupled from the morphological categories then I would applaud your
efforts. What I mean by this is simple. If a constituent functions as the
subject of the main verb it should be given the same function tag without
regard to its lexical/morphological characteristics. This will allow the user
to search for subjects of the main verb which would include pronouns, nouns,
participles, . . . etc.
I think that an attempt to build a database of semantic functional tags might
be a bit on the ambitious side. But if you stick with syntactic functional
categories, like limits-the-main-verb or limits-a-substantive, etc., you will
probably be on safer ground. This sort of database would require two kinds of
information, the syntactic functional tag and a relational pointer connecting
the two or more constituents that participate in the relation defined by the
tag.
Needless to say, the developer of this model needs to be a three headed
monster (a joke). He needs to be a linguist, a NT Greek scholar, and a
software engineer with a specialty in information modeling. If no such there
headed monsters are easily available then you need to build a team which can
communicate well. In the world where I used to work this would have involved
face to face contact five days a week over several months to work out the
details of such a complex problem. Lots of writing on the white board, lots of
false starts, lots of design reviews. The major problem in dealing with this
in e-mail is the lack of graphics. These kinds of problems are solve by
drawing diagrams not by writing paragraphs.
Having said all of this I am very much in favor of seeing this effort go
forward. Have you chatted recently with some of the folks in Vancover/Portland
about parallel efforts that might be going on there?
-- Clayton Stirling Bartholomew Three Tree Point P.O. Box 255 Seahurst WA 98062--- B-Greek home page: http://sunsite.unc.edu/bgreek You are currently subscribed to b-greek as: [cwconrad@artsci.wustl.edu] To unsubscribe, forward this message to leave-b-greek-329W@franklin.oit.unc.edu To subscribe, send a message to subscribe-b-greek@franklin.oit.unc.edu
This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:40:08 EDT