COMMENT ⊗   VALID 00003 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.DEVICE XGP
C00003 00003	.PORTION MINSKY
C00006 ENDMK
C⊗;
.DEVICE XGP

.FONT 1 "NGR25"
.FONT 2 "NGB30"
.FONT 4  "BDI25"
.FONT 5  "NGR20"
.PAGE FRAME 70 WIDE 55 HIGH
.AREA TEXT LINES 4 TO 55
.AREA HEADING LINES 1 TO 3
.TURN ON "↓_π{"
.TURN ON "⊗" FOR "%"
.MACRO B ⊂ BEGIN VERBATIM GROUP ⊃
.MACRO E ⊂ APART END ⊃
.TABBREAK
.COMPACT
.SELECT 1
.PORTION MINSKY
.BEGIN FLUSH RIGHT
Doug Lenat
Artificial Intelligence Laboratory
Computer Science Department
Stanford University
Stanford, California 94305
415-329-1031, 415-497-4971
{date}
.END
.B


Professor Marvin Minsky
Artificial Intelligence Laboratory
Massachusetts Institute of Technology
545 Technology Square
Cambridge, Mass. 02139


.E
.SELECT 2
Dear Marvin,

.B

.E
.INDENT 6

	Here is the long version of the BEINGS paper, as you requested, plus
the latest draft of the short IJCAI paper. The latter has been almost completely
rewritten, and perhaps warrants reexamination. 

Only the appendices of the long document are worth scrutiny; the initial
sections need to be cleaned up.
Despite their length, bear in mind that the appendices
are but tiny excerpts (e.g., the CF dialogue in its entirety would occupy an
extra 300 pages.)

If you have any specific questions, requests for the whole PUP6 code,
complete target programs, complete
dialogues, etc., I'll be happy to send you more material.
The overall worth per pound drops significantly, of course.

	When my math-understanding ideas get written up as a formal
thesis proposal (a few weeks), I'll mail you a copy.  Meanwhile, thanks
for your interest.

.B

					Sincerely yours,

.E