De-Inflection Software?

Richard A. Stauch rstauch at charter.net
Mon Sep 23 15:45:19 EDT 2002


Dear Dr. Dale,

Have you read anything I have written? It sounds to me like you are
replying to things others have written.

Thanks for the warning against copyright theft. I am sure there are some
people who need it (though I wonder what prompts you to accuse
"Evangelical Christians" of being "at the forefront" of this illegal
conduct). As an Evangelical Christian AND a computer professional, I can
safely say I would never do that, nor would I allow anyone on my team to
-- if I ever assemble one -- nor do I know any Evangelical Christians
who would. Nothing I have ever written has advocated such theft, nor
could it be honestly inferred from my writings (though I suppose
anything is possible with the "deconstructionists" among us). Just,
please, try to remember that I started this thread asking if there
already existed some software that would do this (deeper explanation
below); not to steal it, but to buy it to use -- like I do all my books
AND software.

Besides, why would I advocate copyright infringement so publicly? I
could be arrested for stupidity alone.

With all respect, and to be honest, it is amazing to me that so many
obviously bright and learned people can seem to be so opaque when it
comes to advancing a clearly new idea. (New to some, if not "new under
the sun.") Let me try again: This new idea has to do with computer
programming in word morphological transformation, not the regurgitation
of static data from text-string lists. That has been done, and done, and
done again. By definition, therefore, it most assuredly DOES NOT need
"one of the parsed databases already in existence as its foundation."
All it needs is (1) an ability to utilize the rules of word morphology
and (2) tables of the stems and x-fixes used to construct words.

(1) There have to be such rules, or Smyth (for instance) would have had
nothing about which to write. We would not be plagiarizing Smyth or
anyone else, either, as we are talking about codifying rules in program
code, not producing a human-readable book. You cannot copyright the
rules of word morphology, or nobody would dare explain a rule on this
list for fear of copyright infringement.

(2) As for the stems and x-fixes, how can they be copyrighted? One
cannot copyright "SUN-" or "-EDR-" or "-W-" or "-N". So if I had such
software, and I fed it "SUNEDRWN" (as in 4 Mac. 5:1),* it would not look
for that string; it wouldn't have such a table, because that is NOT how
it functions (ergo, no copyright infringement). Rather, it would
(somehow) separate the letters in the word (not to mention accentuation)
and look in its x-fix and stem tables for possibilities. If it found
multiple possibilities (I suppose that could happen), it would simply
report all of those. It would then be up to the end-user to determine
which is most likely right.

* SUNEDRWN just happens to be a noun, picked at random. I could have
picked any word that can be inflected. Also, I don't know if I separated
the word's parts correctly, but that doesn't matter for the purposes of
this post. I am merely trying to explain what I have been talking about,
and how it differs from what others have incorrectly assumed
"De-Inflection Software" means (not trying to prove anything about me).

BTW, the tables I am talking about would not best be constructed from
existing work (CCAT or otherwise). They would be, as I have advocated
before, best compiled by experts from original source documents,
particularly from the Koine Period, but not limited to that. We are
talking about a whole new kind of database for a whole new purpose,
nothing like what you have been working with. That is to say, this is
not your daddy's wheels -- it's a rocket.

I really hope that is clear enough for everybody. I do not expect I will
ever chose to explain it again.

Sincerely,
Richard Allan Stauch
Long Beach, CA
http://www.rstauch.com
 

> -----Original Message-----
> From: Dr. Dale M. Wheeler [mailto:dalemw at multnomah.edu]
> Sent: Monday, September 23, 2002 8:30 AM
> To: Biblical Greek
> Subject: [b-greek] Re: De-Inflection Software?
> 
> B'Greekers:
> 
> As someone who has been working with parsed original language
electronic
> texts for about 25 years now, permit me to make a somewhat crass
> observation, which I'm sure some of you'all won't like...
> 
> The "De-Inflection" software, as you call it, sounds to me like it
needs
> one of the parsed databases already in existence as its foundation,
> especially if it is being written by someone who doesn't know Greek
(or
> Hebrew).  One simple example would be irregular verb principle
parts...it
> would probably take someone without a knowledge of Greek several
hundred
> man hours at least just to be able to track down all the various
irregular
> verb forms and their alternates (eg., DEIKNUMI vs DEIKNUW) just in the
NT;
> unless of course one had one of the GNT parsed databases, in which
someone
> who does know Greek had invested the time to do the work for you.
> 
> A bit of history...the CCAT texts were originally parsed by programs
> written by Dick Whitaker, et.al., who were experts in the languages
and
> programming over 25 years ago, and I'm still fixing the mistakes made
by
> the parser (and the grad students and others even less qualified who
tried
> to fix the problems, anomalies, etc.).  If those guys couldn't solve
all
> of
> the problems with programming, I would have serious doubts that anyone
> else
> can.
> 
> As far as I know EVERY single GNT, LXX, and HMT parsed database comes
with
> a very clear license agreement about usage, and the use you'all seem
to be
> describing is clearly in violation of those agreements.  I have always
> found it very strange that people are quite scrupulous when it comes
to
> using/mis-using material that is in print, but feel perfectly free to
use
> electronic material, as if it were in some different category.  In
fact,
> the license agreements for the electronic texts are even more
restrictive
> than for print materials.  Evangelical Christians seem to be at the
> forefront of ignoring the biblical principle that a workman is worthy
of
> his wages, and thus have misused, misappropriated, etc., etc., etc.,
these
> text apparently without a second thought.
> 
> So here's a suggestion...rather than trying to reinvent the wheel, why
not
> contact the copyright holders of these already existing databases and
> offer
> to help their ongoing work on these texts.  There is still a lot of
work
> to
> be done on the texts, and there are all kinds of additional data that
can
> be added to these texts in order to make them even better research
> tools.  Some of the auto parsers, etc.,  discussed in this thread
would be
> easy to build with the parsed databases already in existence, and
would
> make excellent add-on tools for people trying to learn the language or
for
> people doing serious research. And with those things easily done, you
> could
> then apply your skills to assisting someone who really knows the
languages
> in developing new add on tools and data and really make a significant
long
> term investment in original language studies which will have an impact
for
> generations to come.
> 
> The one exception to most of the above is James Tauber who has been
> working
> with the CCAT parsed GNT for several years, I believe with the
permission
> of the copyright holders, UPenn and Bob Kraft.  I've tried several
times
> to
> get James to work with me, but he prefers to go it alone
apparently...I'm
> not trying to criticize him by saying this, since he is certainly free
to
> do as he pleases....I just know that if I'd had him working with me on
the
> GNT and LXX over the past several years, we could have made a lot more
> progress with those texts, because he is quite knowledgeable and
skilled
> in
> these areas; I fear this was a poor attempt at a compliment...
> 
> If anyone is interested in making such a contribution of their
talents,
> here are the people to talk to:
> 
> For the GNT:  Prof. Paul Miller at The GRAMCORD
> Institute:  scholars at gramcord.org
> For the LXX: Dr. Bernard Taylor for CATSS:  <taylorb at earthlink.net>
> For the HMT: Dr. Kirk Lowery for Westminster: <klowery at wts.edu>
> 
> 
> Blessings...
> 
> Dale M. Wheeler, Ph.D.
> Editor, the GRAMCORD Morphological Greek New Testament
> Co-Editor, the Groves-Wheeler Westminster Hebrew Morphology
> Associate Editor, the UPenn/CCAT/CATSS Morphological LXX
> 
> 
>
***********************************************************************
> Dale M. Wheeler, Ph.D.
> Research Prof., Biblical Languages          Multnomah Bible College
> 8435 NE Glisan St.                                  Portland, OR 97220
> V: 503-2516416                                 E: dalemw at multnomah.edu
>
***********************************************************************
> 
> 
> ---
> B-Greek home page: http://metalab.unc.edu/bgreek
> You are currently subscribed to b-greek as: [rstauch at charter.net]
> To unsubscribe, forward this message to leave-b-greek-
> 145275F at franklin.oit.unc.edu
> To subscribe, send a message to subscribe-b-greek at franklin.oit.unc.edu
> 






More information about the B-Greek mailing list