[FoRK] Kudos for Jeff Hawkins' _On Intelligence_

Jeff Bone jbone at place.org
Wed May 25 20:04:55 PDT 2005

So I've been wrong about Jeff Hawkins three times.

Active Paper and Palm were about the same size back in '93 / '94.  We  
viewed them as a potential "competitor" in what we believed would be  
a rich emerging market for 3rd-party PDA software.  I met Jeff and  
the Palm team a few times over the course of a couple of years and,  
well, misjudged him and his business several times.

Strike One:  "Nobody's going to learn a *new alphabet* just to get  
freakin' text into their PDAs."  ;-)  Four years later I'm entering  
text into my Palm Pilot at nearly 30 words a minute.  Doh!

Strike Two:  "Those dolts, off quixotically writing their own  
platform.  How the hell are they going to compete with the all-star  
cast of General Magic's partners and licensees?  Nope, we're going to  
fill in the huge missing piece (Internet connectivity and apps) for  
Magic Cap and make a fortune while those guys tank."  Two years later  
we'd made a nice up-front payment licensing our stuff to General  
Magic, but Magic Cap had tanked and our huge back-end never  
materialized.  And I was in line at Demo to sign up for one of the  
first Palm Pilots.

Strike Three's a bit more subtle.

I knew of (and perhaps had even discussed briefly) Jeff Hawkins  
fascination with neuroscience.  I considered it to be a fascinating  
but rather pointless avocation rather than a serious calling.  He was  
an engineer / entrepreneur, after all --- *not* a real scientist.  He  
surely wasn't going to contribute anything meaningful so far afield  
of his expertise, and particularly not while trying to rule the PDA  


Serious kudos for _On Intelligence_.  A bit pop / tech-light,  but a  
great glimpse into what may be one of the most important insights of  
our time.  Hawkins has taken his entrepreneurial treasure chest and  
used it to bankroll a neuroscience research institute with one goal:   
figure out *how the brain works* --- i.e. how the structure and  
electrochemical activity of the human brain gives rise to  
intelligence.  And he may have done it.  The book presents what ---  
while a synthesis of prior primary research, etc. --- may be the  
first convincing, comprehensive theory of how the architectural  
microstructure and macrostructure of the cerebral cortex gives rise  
to human cognition, sentience, etc.  Hawkins calls it the "memory- 
prediction framework."

While some of the details may be wrong, the hypotheses presented in  
this book just *feel* generally right.  It's simple, consistent,  
coherent, and maps fairly well to observation.  And --- while this is  
a very apples / oranges sort of thing --- it fits well with my own  
recent experiences with neurally-inspired software architectures,  
learning systems, and so forth.

If you have any interest at all in the human brain and the phenomenon  
of intelligence, read this book.



More information about the FoRK mailing list