Chaumont Devin, devil at lava.net, wrote on 3 Feb 2000:
>Panlingua: Chapter 14, The Typeless Link,
>by Chaumont Devin, February 3, 2000.
>>>Some Reasons for a Change in my Working Hypothesis about Panlingua.
>>In late 1999 it became increasingly evident that a major upgrade to the
>theory of Panlingua was in order. I have been uneasy for some time about
>the phenomenon of link type. What were link types, and precisely how did
>they relate to other linguistic phenomena? In short, I had the feeling
>that this was a loose end that needed to be tied down.
[...]
http://www.geocities.com/Athens/Agora/7256/mind4th.html Mind.Forth PD AI
(which can be downloaded and run easily without any knowedlge of Forth)
has a very similar arrangement to what Chaumont Devon has outlined here:
>> eat
> ^
> |
>cats->*->mice
>>(cats eat mice)
>>This same verb semnod that links to "eat" also links to "eats", "ate",
>"eaten", and "eating".
>After all, does a nerve fiber in a brain, acting as an associative tag,
have a "type"? Probalby not! The fiber merely carries a signal.
Where it comes from, and where it goes, are what matters.
Out in the open AI projects: Panlingua; Mind.Forth... Any others?