The simple bottom line of that dimension of questions about Newton is that 
Larry 
Tesler has wide scope and equally wide and deep abilities, and that the Dylan 
folks in Cambridge did a very good job in trying to make a new language "along 
the lines of Lisp".

Regarding "frames" etc., I think what computer people were likely to know and 
use and how to think in the 80s was possibly wider and deeper than today.

Also, take a look at SK8, which was yet another interesting system done by 
Apple 
Advanced Development (aka "research")

"Frames" themselves came from Marvin Minsky's amalgam of Pat Winston's Thesis' 
knowledge structures and object oriented programming ideas from PARC and his 
own 
lab (Hewitt).

Newton was a project that allowed too many people's opinions to hold sway. Many 
of the clashing opinions were good on their own, but didn't fit with some of 
the 
other ideas. I wouldn't know how to start describing the twists and turns this 
took.

More recently Alex Warth of Viewpoints (with Stephen Murrell) did a very nice 
Prolog in about 9 or 10 lines of OMeta + about 90 lines of JavaScript for the 
semantics. So this particular approach to quasi-Horn logic is relatively simple 
and quite pretty in a dynamic language.

Cheers,

Alan






________________________________
From: Duncan Mak <[email protected]>
To: Fundamentals of New Computing <[email protected]>
Cc: Alan Kay <[email protected]>
Sent: Wed, March 30, 2011 9:46:20 AM
Subject: Re: [fonc] visual environments created by present/former VPRI staff


Hello Alan,
On Wed, Mar 30, 2011 at 11:47 AM, Alan Kay <[email protected]> wrote:

At Apple, besides Fabrik and Playground (several versions), there was MacPal 
and 
Constructo
>There was (unimplemented) "the hopping curriculum" which used a visual syntax
>Tableau was yet another "before-after" visual production language (which 
>eventually became StageCast). All these go back to SNOBOL, and AMBIT (and its 
>graphical manifestation AMBIT-B at Lincoln Labs).
>

After reading the little bit of history on the Dylan/Ralph/Newton system that 
was done at Apple in the early 90s, I have wanted to find out more about it for 
quite some time.

Mikel Evins about the Lisp-based Newton OS
http://lispm.dyndns.org/news?ID=NEWS-2004-08-14-1 

http://en.wikipedia.org/wiki/SK8

Particularly the bit about frame systems:

It had a frame system used to implement a knowledge base that stored things 
like 
prototypical Person and Place and Date objects, and supported fuzzy 
comparisons. 
The so-called "IA" ("Intelligent Assistant") subsystem used these frames so 
that 
it could guess things like the fact that when you write "Joe" you probably mean 
"Joseph Smith" or "Josephine Baker" from your address book. 

Recently, I started reading the chapters on Logic Programming in SICP and tried 
to learn more about Prolog and Frames. It surprised me to see something so 
"theoretical" used in a consumer-facing product, even though it ultimately 
failed.

Alan, do you know more about these systems, and how they were put together?

Z-Bo, sorry if I'm hijacking your thread.
-- 
Duncan.
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc

Reply via email to