Browse Prior Art Database

PLANNER IMPLEMENTATION PROPOSAL TO ARPA 1972-1973

IP.com Disclosure Number: IPCOM000128142D
Original Publication Date: 1971-Dec-01
Included in the Prior Art Database: 2005-Sep-15
Document File: 11 page(s) / 43K

Publishing Venue

Software Patent Institute

Related People

Hewitt, Carl: AUTHOR [+3]

Abstract

The task objective is the generalization and implementation of the full power of the problem solving formalism PLANNER in the next two years. We will show how problem solving knowledge can be effectively incorporated into the formalism. Several domains will be explored to demonstrate how PLANNER enhances problem solving.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 8% of the total text.

Page 1 of 11

THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.

MASSACHUSETTS INSTITUTE OF TECHNOLOGY A. I. LABORATORY

Artificial Intelligence

December 1971 (Memo No. 250

PLANNER IMPLEMENTATION PROPOSAL TO ARPA 1972-1973

Carl Hewitt

This proposal was written at the Artificial Intelligence Laboratory, a Massachusetts Institute of Technology research program supported in part by the Advanced Research Projects Agency of the Department of Defense and monitored by the Office of Naval Research under Contract Number N00014-70- A-0362-0002.

Reproduction of this document, in whole or in part, is permitted for any purpose of the United States Government.

PLANNER Proposal

Task Objectives

The task objective is the generalization and implementation of the full power of the problem solving formalism PLANNER in the next two years. We will show how problem solving knowledge can be effectively incorporated into the formalism. Several domains will be explored to demonstrate how PLANNER enhances problem solving.

Current Status

Sublanguages of the PLANNER formalism have been implemented at M.I.T. We would now like to have the full power of the formalism available to us. The restrictions of the sublanguages we have now are cramping many of the current applications and discouraging other new applications. The work is currently being pursued on two time-shared computers: a Honeywell 645 and a Digital Equipment Corporation PDP-10. Pursuing work on two systems instead of one is somewhat unusual an so we will give a detailed justification. PLANNER should be independent of any particular computer series or time-sharing monitor. By implementing on two systems we insure that this is true in practice as well as theory. MULTICS and ITS are currently the best systems available for PLANNER. Both systems are currently being upgraded so they

MIT A.I. Laboratory Page 1 Dec 01, 1971

Page 2 of 11

PLANNER IMPLEMENTATION PROPOSAL TO ARPA 1972-1973

will probably still be the best systems in a few years. M.I.T. has signed the contract for the follow on processor for MULTICS. The Stanford A. I. Laboratory is constructing a proto-type for the follow on processor for the PDP-10. Both follow on systems propose to have large amounts of fast random access memory and even larger virtua1 address spaces. If either one comes to fruition, then PLANNER can be applied to problems with large data bases. PLANNER uses hash coding so that the time that it takes to retrieve a fact is essentially independent of the size of the data base PROVIDING there is enough fast random access memory. One danger to be avoided in this approach is duplicating work on the two systems. So far the interaction between the two implementations has been extremely fruitful.

The following is a detailed comparison of MULTICS and ITS for the purpose of implementing PLANNER: Fast Random Access Memory: MULTICS has 400K and ITS has 250K. For acceptable response it should be the case that the working sets of all the us...