Newsgroups: comp.ai,comp.ai.philosophy,comp.ai.games,comp.ai.nat-lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!uunet!world!nproco
From: nproco@world.std.com (Daniel Burnstein)
Subject: Re: Loebner Prize $2000 and a medal
Message-ID: <D721Fw.DvG@world.std.com>
Organization: The World Public Access UNIX, Brookline, MA
X-Newsreader: TIN [version 1.2 PL2]
References: <3ls4nh$qkn@hopper.acm.org> <3m5asj$isa@nunic.nu.edu>
Date: Sat, 15 Apr 1995 02:26:20 GMT
Lines: 301
Xref: glinda.oz.cs.cmu.edu comp.ai:29071 comp.ai.philosophy:26911 comp.ai.games:963 comp.ai.nat-lang:3237

It may be that we are intelligent *and* nervous about change?  Some folks 
operate in a dark room and radiate intelligence.  Howard Gardner writes 
about other frames of mind.  

This rage*ing debate brings to mind the story I heard about a sensitive 
professor at Cornell.  He was annoyed by the young lady in the front row 
of his large lecture class who always came to class and never took notes 
- just knitted.  One day, in frustration, he said, "Young lady, did you 
know that some people believe that knitting is an advanced form of 
masturbation."  The young lady replied, without looking up, "You do it 
your way and I'll do it my way."

Dan Burnstein



Robert Epstein (repstein@nunic.nu.edu) wrote:
: Dr. Loebner's reading of Turing's 1950 paper is preposterous.  Turing
: stresses throughout the paper that "limbs" and "sense organs" are 
: utterly irrelevant to an assessment of intelligence, and the 
: truth and elegance of his proposal are self-evident.  
: Turing even mentioned Helen Keller, just as Mr. Fisher does
: in his memo below.  I can assure you that Dr. Loebner 
: appreciated Turing's message when, a decade ago, he 
: first asked me to establish an annual Turing Test.  His recent
: change of mind remains a mystery to me.   Rather than
: trying to argue with Loebner at this point, however, I would urge  
: the programming community simply to sit tight while the Prize
: Committee continues its efforts to reestablish the annual Turing
: Test with a new underwriter.  The Loebner contest -- and I say this
: with a heavy heart -- is now dead.  Please review the memo below, if
: you haven't already seen it.  /re

: From: repstein@nunic.nu.edu (Robert Epstein)
: Newsgroups: comp.ai
: Subject: Loebner Contest in Disarray
: Date: 13 Mar 1995 21:28:41 GMT
: Organization: National University
: Lines: 68
: Message-ID: <3k2de9$ki8@nunic.nu.edu>
: NNTP-Posting-Host: nunic.nu.edu
: X-Newsreader: TIN [version 1.2 PL2]

: I have directed the Loebner Prize Competition since it began in 1990, and
: I have also served as a member of the Prize Committee that has overseen
: the event and issued the rules each year (see the Summer '92 issue of AI
: MAGAZINE, the June '94 issue of POPULAR SCIENCE, or the April
: issue of WIRED for details).  Other members of the Committee have included
: Drs. Joseph Weizenbaum, Oliver Selfridge, Daniel Dennett, and other
: distinguished people in computer science and other fields. 

: Unfortunately, a few weeks ago, Dr. Loebner, the benefactor who offered
: the original challenge award, demanded that the Prize Committee change the
: event in a number of ways.  The Committee rejected his suggestions
: unanimously.  Without exception, the suggestions were self-serving, making
: it virtually impossible for anyone ever to win this contest.  We were also
: concerned about the fact that Dr. Loebner's suggestions were not in the
: spirit of Turing's original proposal.  For example, he now insists that
: programs must respond to "audio & visual" input, but the Turing Test is a
: test of AI exclusively; Turing specifically ruled out sensory feedback
: from the test.  Loebner also now insists that all programs run on-site,
: meaning that supercomputer users, custom hardware users, and poor graduate
: students can never participate in the event.  He also insists that
: entrants be forbidden from using on-line data bases, which we believe will
: mean tremendous setbacks in some of the programming efforts currently
: underway. 

: At this point:  (a) Dr. Loebner has set about trying to run the event
: himself, (b) the Prize Committee has submitted a formal protest to the
: Board of Trustees of the research institute that administers the Prize,
: (c) the Prize Committee is seeking a new underwriter for the event, and
: negotiations are going smoothly, (d) most of the programmers who have
: submitted in previous years have indicated that they will not enter the
: event as Loebner now envisions it. 

: Of special concern to the Committee is Loebner's insistence that he
: personally, or that his personal representatives, must be fooled by the
: winning entry in order for him to award the grand prize. The Committee has
: always relied on a double-blind procedure to assure objectivity in
: judging; we find Dr. Loebner's new demand highly objectionable.

: We are confident that the Turing Test will continue as an annual event. 
: Watch for upcoming announcements. 

: Cordially,

: Dr. Robert Epstein, Research Professor, National University
: Director Emeritus, Cambridge Center for Behavioral Studies

:              ___________________             __________________
:             /933 woodlake drive/            /tel 619 436 4400 /
:            /     ________     /            /     ____________/
:           /     /       /    /            /     / 
:          /     /       /    /            /     / 
:         /     /______ /    /            /     /____
:        /cardiff by the sea/            /fax 4490  /
:       /             _____/            /      ____/
:      /              \                /      / 
:     /       /\       \              /      /______________
:    /ca     /  \92007  \            /repstein@nunic.nu.edu/
:   /       /    \       \          /                     / 
:  /_______/      \_______\obert   /_____________________/pstein  
:                                                               


: Hugh Loebner (loebner@ACM.ORG) wrote:
: : Travis Fisher <tfisher@hannibal.wncc.cc.ne.us>
: :    In message  <3llkqf$6h6@nlcnews.nlc.state.ne.us> writes:

: : >loebner@ACM.ORG (Hugh Loebner) wrote:

: : >> In accordance with the requirements of Dr. Loebner as
: : >> published in the June 1994 Communications of the ACM,
: : >> the winner of the Grand Prize must be prepared to deal
: : >> with audio visual input (pattern recognition).

: : >Say What???  If the Turing test is a test of intelligence (it _is_
: : >supposed to be, isn't it) then these recquirements are not a good
: : >idea.  Why not?  Was Hellen Keller intelligent?  The point is,
: : >these requirements are extraneous to the basic intelligence that
: : >the Turing test was originally designed to test.

: : >This Loebner competition, once these rules are in effect, will be
: : >laughable, especially when the on-site requirements are considered.
: : >I would really like to see a PC or a Mac do audiovisual pattern
: : >recognition in realtime, much less do this and _anything_ else at
: : >the same time.

: : >What comes the year after next?  Does the computer have to be in the
: : >form of an android that is indistinguishable from a human?

: : In Message-ID: <3lllg9$785@nlcnews.nlc.state.ne.us> he continues:

: : >blasirs@dunx1.ocs.drexel.edu (Robert S. Blasi) wrote:
: : >>
: : >>In article <3lh3iq$s7c@hopper.acm.org>, Hugh Loebner <loebner@ACM.ORG> wrote:
: : > >
: : > >In accordance with the requirements of Dr Loebner ...[ cut- see above]
: : > > ... (pattern recognition).  However, to
: : > >celebrate the first unrestricted Turing Test, this rule
: : >
: : >       First, can anyone clarify for me when a "restricted" and an
: : > "unrestricted" Turing Test were proposed?
: : [cut]

: : >I believe that the phrases "restricted" and "unrestricted" are intended
: : >to refer to the subject areas in which the computer is expected to be
: : >able to converse.  In a "restricted" Turing test, the computer (and
: : >human confederates) get to choose a specific area for discussion,
: : >while in an "unrestricted" Turing test, any area of conversation brought
: : >up by the judge is fair game.
: : >
: : >As for audio-visual communications, I think this is [word self-censored]
: : >invented by Dr. Loebner because he is getting scared he might
: : >actually have to give out a grand-prize award sometime soon.

: : Response:

: :     Much criticism has been directed at my requirement that the Grand Prize
: : winner be able to deal with audio visual input.  These fall into several
: : categories (1) that Alan Turing did not explicitly discuss A/V, therefore
: : this has nothing to do with a Turing Test, (2) Some individuals, such as
: : was Hellen Keller, are blind, deaf, *and* intelligent, therefore
: : intelligence is not predicated on vision,  (3) that the task exceeds current
: : hardware abilities, and (4) that my requirement is an expedient to avoid ever
: : having to award the grand prize. I will address these seriatim. 

: : 1.  Alan Turing and the nature of the Turing Test
: :      Turing was writing in 1950.  The computer was unknown except to a very
: : few.  Part of his article was intended as an introduction for philosophers
: : as to what a digital computer was.  The rest was primarily an argument in
: : support of the possibility of an "artificial intelligence."  In his paper,
: : Turing mention the "imitation game," not the "teletypewriter game."

: :      The Turing Test is about method, not about content.

: :     The essence of the test is that a computer be indistinguishable from
: : a human in *all* intellectual fields.
: :      Although he wrote of his test using a teletypewriter, Turing
: : concluded with these words: 
: :           "We may hope that machines will eventually compete with men
: :      in all purely intellectual fields.  But which are the best ones to
: :      start with?  Even this is a difficult decision.  Many people think
: :      that a very abstract activity, like  the playing chess, would be
: :      best.  It can also be maintained that it is best to provide the
: :      machine with the best sense organs that money can buy, and teach
: :      it to understand and speak English.  This process could follow the
: :      normal teaching of a child." - A. M. Turing, "Computing Machinery
: :      and Intelligence," MIND, LIX, 1950, p. 460

: :      Consider, now, some "purely intellectual fields." Every IQ test that
: : I can remember being given has had questions of the sort: "What is wrong
: : with this picture?" and "Which illustration of this group does not belong?"
: : and "Which is the next shape in this series"?  I am sure that *you* can
: : think of interesting intellectual questions regarding visual material.
: :      Suppose that we do exclude "audio visual" input, and restrict the
: : input to exclusively teletypewriter communications.  Consider the following
: : 6 line input string from the human judge to the (computer/human):

: : >Please examine the pattern in the next 5 lines and tell me what you see.
: : >.....X...X
: : >......X.X
: : >.......X
: : >......X.X
: : >.....X...X
: :     
: :     Are we to say that this question is forbidden because it does not 
: : consist of words, but rather characters?  For that matter, how many
: : signatures appended to mail and news postings comprise clever images
: : constructed of characters?  Does yours?  These attacks make me feel :-(
: :     I have been involved with computing long enough (since 1966) to
: : remember when line printers were used to print complex pictures constructed
: : of characters with different ink densities [eg the seq [space] . - = x X @].
: :     Someone wrote of the physicist Richard Feynman that other physicists, in
: : doing advanced work in an area, had found that Feynman had anticipated them
: : and "signed the guest book and then moved on."
: :      I think that this is the case here with Turing.  Turing was completing
: : his article. He had described the digital computer and discussed at length
: : the possibility of an artificial intellect. He had propounded the basic logic
: : of the Turing Test.  Like every editor, his was probably complaining that
: : the article was too long.  I think that with the paragraph I quote, Turing
: : was saying in effect: "There's much more that can be written about this.
: : There's heuristic programming. There's the question of sensory data and how
: : to deal with it.  I have discussed only the teletypewriter, but the logic
: : can be extended to other I/O modes.  I know this (that's why I mention
: : 'the best sense organs money can buy') but for now this is enough."
: :      
: : 2.  The intelligence of those without sight and hearing.
: :     I believe that all intelligence is ultimately grounded in sensory
: : input.  Consider what Hellen Keller herself wrote regarding her
: : acquisition of language.
: :           "We [Hellen Keller and her teacher, Anne Sullivan] walked
: :      down the path to the well-house, attracted by the fragrance of
: :      the honeysuckle with which it was covered.  Some one was drawing
: :      water and my teacher placed my hand under the spout.  As the cool
: :      stream gushed over one hand she spelled into the other the word
: :      _water_, first slowly, then rapidly.  I stood still, my whole
: :      attention fixed upon the motions of her fingers.  Suddenly, I felt
: :      a misty consciousness as of something forgotten - a thrill of
: :      returning thought; and somehow the mystery of language was revealed
: :      to me.  I knew then that "w-a-t-e-r" meant the wonderful cool
: :      something that was flowing over my hand.  That living word awakened
: :      my soul, gave it light, hope, joy, set it free!  There were barriers
: :      still, it is true, but barriers that could in time be swept away."
: :      - Hellen Keller, The Story of My Life, Penguin Group Signet
: :      Classic, NY, 1988. p.18

: :   If one reads Keller's autobiography (which I strongly recommend) one will
: : find consistent references to olefactory, tactile and thermal sensory
: : references. I have not made these requirements for the Loebner Prize, but I
: : do not disparage anyone who, also establishing a Turing Test prize for the
: : developent of AI, required them.  In requiring A/V input, my feeling is not
: : so much that I am *extending* the range of the Turing Test to include A/V
: : input as I am in *limiting* it to exclude other, important sensory input.
: :      There was an article in the NY Times not long ago about a blind
: : biologist who, by tactile means, was able to classify shells and
: : describe features (such as predation by other organisms) that were
: : invisible to the sighted.  Here too we have an example of one out of the
: : range of "all purely intellectual fields" and again it is based on
: : a sensory (tactile) input.
: :      I did not require tactile, olefactory, and other sensory inputs
: : because I did not want to make the test too difficult.

: : 3.  That the task exceeds current hardware capabilities is not ipso
: : facto an argument that the task is not worthwhile.  It may serve as
: : a guide and as an incentive to the design of new hardware.  
: :     Among the reasons that I sponsored the Loebner Prize was to act as an
: : incentive for the development of AI.  I knew that the task was difficult.
: : In order to reward workers for efforts in a task that I knew would take many
: : years, I stipulated that an annual award be given for that entry which was
: : best in any year.
: :    Rest easy, Travis Fisher.  It may be that the requirement exceeds
: : current hardware capabilities.  Nevertheless, take comfort from the fact
: : that this impediment applies equally to all.  If you enter the contest
: : you have only to create the best entry (given current hardware) and
: : you will win $2000. and a medal and have your name noted as one in a
: : list that will eventually, I hope, result in the development of an
: : artificial intellect.  

: : 4.   My AV requirement is not an attempt to avoid paying the $100,000
: : prize.  It is entirely based upon my belief that (i) intelligence has a
: : sensory component and (ii) that it is congruent with the Turing Test as
: : Alan Turing envisioned it.
: :      (a)  If I did not intend to award the prize I would not have made
: : the pledge.  To state otherwise is to defame me.
: :      (b)  I am supporting the operation of an annual contest as well as 
: : providing a $2000 prize per year to the annual winner. The sooner someone
: : wins the contest, the less my expenditures will be.
: : -------------------------------------------------------------------------------
: : Hugh Loebner, Pres.  Crown Industries, Inc. 155 N Park St, East Orange NJ 07017
: :           Mfgs of Roll-Up Plastic Lighted Portable Disco Dance Floors
: :                and Other Essential Products for Home and Industry
: :                        Crown Industries is *proud* to be a
: :        Corporate Sponsor of The Loebner Prize for Artificial Intelligence
: : Tel:Business 1-800-GO-CROWN                  Personal Voice Mail 1-500-EGO-TRIP
: :                          e-mail  loebner@acm.org
:                                                               




