Date: Mon, 23 Jul 2001 12:45:59 -0700
From: | Eric Armstrong, eric.armstrong@eng.sun.com |
Reply-To: unrev-II@yahoogroups.com |
To: | unrev2, unrev-II@yahoogroups.com |
Subject: | Lee's Nodal system |
I spent a couple of hours with Lee the other day,
going over the architecture and capabilities of
Nodal. Much of what we discussed has yet to be
incorporated into his paper at....
http://www.ai.sri.com/~leei/OHS/NODAL-WhitePaper.htm
But the bottom line is that I am convinced his NODAL mechanism is indeed a better design for the KRNL (Knowledge-Repository Node Layer), which I had been trying to construct. I eventually bailed out when the complexity grew overwhelming. But Lee isolated the defects, simplified the thing enormously, and at the same time added drastically new concepts based on Groves and abstract data modeling.
The result, I truly believe, is a superb undertaking that will serve as solid foundation for the Engelbart-inspired tools we have separately envisioned -- and one that will go a long way towards helping them work together.
The ambitious aim of his undertaking is to create a node-based "file system" that can be used by any application in the same way they use current file systems -- but which provides granular addressability, versioning, and all the rest to smarter applications.
In other words, it is stone-dead killer.
Sincerely,
Eric Armstrong
eric.armstrong@eng.sun.com