One of the reasons I chose the Forth virtual machine for my initial
architecture for the VIEWTRON "videotex" nation-wide rollout (that didn't
happen for reasons I described in response to "Why Didn't The Internet Take
Off In 1983? <https://slashdot.org/comments.pl?sid=2702791&cid=39217853>")
is that the modems of that era -- even those provided by the VIEWTRON
beta-ISDN deployment in Miami circa 1982 -- were so limiting that in order
to support the lazy (on demand) dynamically downloaded graphics primitives
(think a Turing-complete SVG protocol with distributed execution) it was
important to maximize the compression thereof.  It also helped that the
Western Electric NAPLPS graphics protocol videotex terminal provided by our
AT&T partner had just enough ROM that I could recode NAPLPS in Forth and
support programmable graphics in the tiny amount of RAM left over.  That's
because the Forth VM is vastly more parsimonious than today's VMs.  I
_also_ helped that this tiny interpreter was _so_ parsimonious that it
could be moved into silicon (even gate-array ala NOVIX).  I was in talks
with everyone from Bell Labs to Xerox PARC to Paul Baran's foray into mass
markets, about doing just that ASAP.  The PARC folks liked the "tokenized
Forth graphics communication protocol" idea enough that apparently it
motivated some of them to pursue something similar out of the University of
Utah that became postscript
<https://web.archive.org/web/20020610215620/geocities.com/jim_bowery/psgenesis.html>.
I know they weren't thinking about it at the time I visited them in 1982
and discussed it with them.  I had only mentioned the communications
protocol as a side comment since I was interested in Forth as an assembly
language for distributed Smalltalk execution using David P. Reed's thesis
(which decades later became the basis of Alan Kay's Croquet "Teatime"
distributed virtual reality synchronization).  In any event, all this
points to something pretty important about your approach to "AGI", which is
that Chuck Moore's genius inspired me to look not only at program size
minimization as a means of maximizing channel capacity on early networking
architectures, but at minimum complexity Turing machines for the execution
thereof.  This is probably why my work on neural networks several years
later focused my training algorithms on pruning the networks to achieve
generalization and then, 16 years later, to propose a compression prize
that Matt Mahoney informed me had theoretic support in Algorithmic
Information Theory, pointing me toward the "UI" (Universal Intelligence aka
"AGI") work by Marcus Hutter, thence The Hutter Prize.

Now, why am I going down memory lane with you, oh my fellow acolyte of
Chuck's genius?

Because I believe you have the right instincts down there somewhere buried
in your subconscious and would plead with you to try your hand at The
Hutter Prize <http://prize.hutter1.net/>.  Even if you don't win, your
effort could well be more instructive than your website and posts here.

On Mon, Dec 13, 2021 at 9:14 AM A.T. Murray <[email protected]> wrote:

> Just as war is too important to be left to the generals, so also
> Artificial Intelligence is too important to be left to the Computer
> Scientists. Imagine the chair of Computer Science at some university
> commanding a graduate student, "Go over to the Classics Department and spy
> on the
>
> https://ai.neocities.org/AILA.html -- Latin AI course
>
> for me. Bring me all the course materials you can get your hands on."
>
> It will be an interdepartmental turf-war on campus, just like World War
> Two, but with more at stake. That retrospectively minor war was broken down
> into
>
> the European theater of war and
>
> the Pacific theater of war.
>
> Now the Mentifex AGI Project is broken down into specific theaters.
>
> In the Latin Theater of Mentifex AGI, we have yesterday updated the
> webpage
>
> http://ai.neocities.org/Concursus.html -- a concourse of Latin weblogs.
>
> The Latin title of the page is "Concursus Omnium Bonorum," a phrase from
> Marcus Tullius Cicero which means "The cooperation of all the good men."
> Now is the time for all good men to... oh, never mind. What matters here is
> that the sly trickster Mentifex is using the snobby, effete intellectual
> Classics Community as a back-door or "Trojan Horse" (sure, a hacker term,
> but the Classics people came up with it first by a few thousand years)
> entry into the trillion-dollar forefront of Artificial General
> Intelligence.
>
> En garde, Elon! Choose your weapon, Marvin! C'est la guerre, SAIL.
>
> http://www.mail-archive.com/[email protected]/msg08516.html -- AGI
> archive.
>
> Mentifex
> --
> http://ai.neocities.org/LaThink.html -- mind-module for AI thinking in
> Latin
> http://cyborg.blogspot.com/2019/05/redux.html -- converting Latin AI into
> Russian AI
> http://la.wikipedia.org/wiki/Intellegentia_artificialis -- AI in Latin
> Wikipedia
> http://www.amazon.com/dp/B08NRQ3HVW -- AI in Ancient Latin book at Amazon
>
>
> *Artificial General Intelligence List <https://agi.topicbox.com/latest>*
> / AGI / see discussions <https://agi.topicbox.com/groups/agi> +
> participants <https://agi.topicbox.com/groups/agi/members> +
> delivery options <https://agi.topicbox.com/groups/agi/subscription>
> Permalink
> <https://agi.topicbox.com/groups/agi/Tfcec9bac2a105db6-M6d1f424d217956a450dd216f>
>

------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tfcec9bac2a105db6-M421ed619c1d78faa6df62c51
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to