From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!wupost!micro-heart-of-gold.mit.edu!news.media.mit.edu!minsky Thu Jan  9 10:34:22 EST 1992
Article 2582 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!wupost!micro-heart-of-gold.mit.edu!news.media.mit.edu!minsky
>From: minsky@media.mit.edu (Marvin Minsky)
Subject: Re: Edelman's non-computability non-argument
Message-ID: <1992Jan9.002007.29841@news.media.mit.edu>
Sender: news@news.media.mit.edu (USENET News System)
Cc: minsky
Organization: MIT Media Laboratory
References: <2215@ucl-cs.uucp> <356@tdatirv.UUCP> <61662@netnews.upenn.edu>
Date: Thu, 9 Jan 1992 00:20:07 GMT
Lines: 15

In article <61662@netnews.upenn.edu> weemba@libra.wistar.upenn.edu (Matthew P Wiener) writes:
>In article <356@tdatirv.UUCP>, sarima@tdatirv (Stanley Friesen) writes:
>
>Along these lines, I note with irony that the famed Minsky-Papert proof
>that perceptrons can't recognize connectivity and the like to be support
>for them.  Have you ever seen those illusion pictures involving spiral
>mazes?  Where you can't tell if it's one or two walls, except with your
>finger?  M&P showed that perceptrons model this perceptual illusion, yet
>no one noticed at the time.

I re-drew them for the new edition, this time using a computer to make
sure they had the same areas.

By the way, there is an even funnier illusion on page 5 of
Perceptrons, which scarcely anyone noticed.  I'm not sure what it


