From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!haven.umd.edu!darwin.sura.net!Sirius.dfn.de!fauern!unido!mcsun!fuug!news.funet.fi!sunic!seunet!kullmar!pkmab!ske Mon Mar  9 18:32:56 EST 1992
Article 4049 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!haven.umd.edu!darwin.sura.net!Sirius.dfn.de!fauern!unido!mcsun!fuug!news.funet.fi!sunic!seunet!kullmar!pkmab!ske
>From: ske@pkmab.se (Kristoffer Eriksson)
Newsgroups: comp.ai.philosophy
Subject: Re: Intelligence Testing
Message-ID: <6603@pkmab.se>
Date: 24 Feb 92 13:50:42 GMT
References: <1992Feb22.171652.5827@oracorp.com>
Organization: Peridot Konsult i Mellansverige AB, Oerebro, Sweden
Lines: 31

In article <1992Feb22.171652.5827@oracorp.com> daryl@oracorp.com writes:
>>The structure is not complex. It's a table, remember.
>
>I said the *data* is enormously complex! To get a measure, take some
>measure of complexity and estimate the complexity of a single sentence
>of English. For example, you might consider how many bits it takes to
>code it using the best known compression algorithms. Now multiply this
>number by the number of sentences in the table (estimated at
>10^(10,000,000)). That gives a rough idea of the complexity of the
>table.

A technical note on measuring the complexityof the data in the table:
I think you should compress the entire table, before measuring the size,
rather than compressing the average sentence on its own, and multiplying
by the number of entries. Unless the table prescribes totally random
behavior, there will be lots of redundancy between different parts of
the table (different branches of the conversation that the table
prescribes).

An intelligent table will actually be less complex than an unintelligent
random table, although it will still be very complex. An intelligent table
should, using this way of measuring complexity, should be of comparable
complexity as a program or physical apparatus that produces the same
intelligent behavior! (The storage size will be different, though,
obviously.) An algorithm can be viewed as a way to compress the actual
table of all possible outputs.

-- 
Kristoffer Eriksson, Peridot Konsult AB, Hagagatan 6, S-703 40 Oerebro, Sweden
Phone: +46 19-13 03 60  !  e-mail: ske@pkmab.se
Fax:   +46 19-11 51 03  !  or ...!{uunet,mcsun}!mail.swip.net!kullmar!pkmab!ske


