From: Deborah McGuinness (dlm@ksl.stanford.edu)
Date: 11/08/01
agreed that we want to have some smart reasoning backend do more work for us (and
generate at least the daml code and possibly some abstraction as well).
that said, that is job security.... no small design or programming task.
One could imagine however a JARGON or TELI like interface - Jargon was Bill Wood's
limited english translator of KL-ONE from the early 80s and TELI[1] was the bruce
ballard's transportable english language interface also from the mid 80s (We used
it as a front end to kandor in 1985 to have what I viewed as a "database ease" query
language to a DL. We generated a cute demonstration system called kandi that
appeared quite usable to those who understood the limitations of the english front
end.)
neither were rocket science but convenient.
Today, one could make them much more usable with web interface tools.
one could imagine the ability to have nice web guis that help one form classes out
of english-like templates and then have a backend system generate the daml+oil code.
as i said, this is no small design or implementation effort however and
unfortunately, in my opinion, the most useful ones would have a fair amount of user
modeling information in them so as to help present more useful options to users (and
not overwhelm them with the breadth of options in the language).
d
[1] B. W. Ballard and D. E. Stumberger
Semantic Acquisition in TELI: A Transportable, User-Customized Natural Language
Processor
@inproceedings{Ballard:86AC,
AUTHOR = {B. W. Ballard and D. E. Stumberger},
TITLE = {Semantic Acquisition in TELI: A Transportable, User-Customized Natural
Language Processor},
YEAR = 1986,
BOOKTITLE = "Proc. of the 24th ACL",
ADDRESS = "New York",
PAGES = {20-29}
Jim Hendler wrote:
> At 9:08 AM -0800 11/8/01, Deborah McGuinness wrote:
> >I am willing to collect such idioms.
> >Please send me
> >1 - english of a statement you want to make
> >2 - a daml solution (if it is tricky, an english paraphrase first is useful).
> >
> >I have a start at such a collection since I wrote the "tricks of the
> >trade" section
> >of the "how and when to live with a kl-one-like language" [1] paper
> >many moons ago.
> >I generated that from working with users of classic and seeing how
> >and when they
> >were confused in their modeling tasks.
> >That was for a mostly less expressive language than daml+oil
> >(although interestingly
> >enough would have happily been able to say the thing that jeff
> >wanted to say since
> >it did have same-as as a constructor).
> >
> >[1]
> >http://www.ksl.stanford.edu/people/dlm/papers/living-with-classic-abstract.html
> >
> >d
>
> while I think this is useful, I think a human-readable solution isn't
> going to get very far -- we need to figure out new and better ways to
> do this (example, come up with simpler language features and make the
> logic reasoners work a little harder) - just a thought
> -JH
> --
> Professor James Hendler hendler@cs.umd.edu
> Director, Semantic Web and Agent Technologies 301-405-2696
> Maryland Information and Network Dynamics Lab. 301-405-6707 (Fax)
> AV Williams Building, Univ of Maryland College Park, MD 20742
> http://www.cs.umd.edu/users/hendler
--
Deborah L. McGuinness
Knowledge Systems Laboratory
Gates Computer Science Building, 2A Room 241
Stanford University, Stanford, CA 94305-9020
email: dlm@ksl.stanford.edu
URL: http://ksl.stanford.edu/people/dlm/index.html
(voice) 650 723 9770 (stanford fax) 650 725 5850 (computer fax) 801 705 0941
This archive was generated by hypermail 2.1.4 : 04/02/02 EST