/LMAR=0/XLINE=3/FONT#0=BASL30/FONT#1=BASI30/FONT#2=BASB30/FONT#10=BAXM30/FONT#11=ZERO30/FONT#3=STA200/FONT#4=NGB25
␈↓ ↓H␈↓␈↓βS␈↓∧ Artificial Intelligence Laboratory, STANFORD UNIVERSITY, Stanford, California 94305

␈↓ ↓H␈↓∧Telephone 415 497-4430␈↓ 
-July 7, 1980 




␈↓ ↓H␈↓Professor Jack Schwartz
␈↓ ↓H␈↓Courant Institute
␈↓ ↓H␈↓251 Mercer Street
␈↓ ↓H␈↓New York, N.Y.  10012

␈↓ ↓H␈↓Dear Jack:

␈↓ ↓H␈↓        Enclosed␈α≡nally␈αis␈αthe␈αancient␈αmemorandum␈αI␈αreferred␈αto.␈α I␈αcouldn't␈α≡nd␈αit␈αin␈αmy␈αown␈α≡les,
␈↓ ↓H␈↓which␈α
were␈αdisrupted␈α
by␈α
the␈αmove␈α
of␈α
the␈αAI␈α
Lab,␈α
but␈αfortunately␈α
Don␈α
Knuth␈αwas␈α
able␈αto␈α
retrieve
␈↓ ↓H␈↓a␈αcopy.␈α He␈αhad␈αalso␈αwritten␈αto␈αStrachey,␈αasking␈αabout␈αthe␈αrelation␈αbetween␈αhis␈α1959␈αICIP␈αpaper
␈↓ ↓H␈↓and␈αmulti-terminal␈αtime␈αsharing,␈αand␈αhe␈αsent␈αme␈αa␈αcopy␈αof␈αthe␈αreply,␈αa␈αcopy␈αof␈αwhich␈αis␈αenclosed.
␈↓ ↓H␈↓In order not to lose the memo again, I have had it put in the computer.

␈↓ ↓H␈↓        Here are a few comments on the memo.

␈↓ ↓H␈↓        1.␈α∞Someone␈α∞thought␈α∞the␈α∞date␈α∞should␈α∞be␈α∞January␈α∞1960,␈α∞but␈α∞I␈α∞forget␈α∞why.␈α∞ It␈α∞was␈α
certainly
␈↓ ↓H␈↓shortly␈α∞after␈α∞it␈α
became␈α∞known␈α∞that␈α∞IBM␈α
would␈α∞produce␈α∞a␈α∞transistorized␈α
version␈α∞of␈α∞the␈α∞709␈α
but
␈↓ ↓H␈↓probably before its naming as the 7090.

␈↓ ↓H␈↓        2.␈αI␈αdon't␈αknow␈αwhy␈αI␈αsaid␈αthe␈αideas␈αweren't␈αespecially␈αnew,␈αexcept␈αpossibly␈αthat␈αI␈αhad␈αbeen
␈↓ ↓H␈↓advocating␈α∪them␈α∩since␈α∪I␈α∪came␈α∩to␈α∪M.I.T.␈α∩ in␈α∪1957.␈α∪ I␈α∩may␈α∪have␈α∩interpreted␈α∪the␈α∪presence␈α∩of
␈↓ ↓H␈↓boundary registers in the STRETCH architecture as evidence of an IBM time-sharing project.

␈↓ ↓H␈↓        3.␈αAs␈α
Strachey's␈αletter␈α
mentioned,␈αthe␈α
word␈αtime-sharing␈αwas␈α
"in␈αthe␈α
air",␈αand␈α
I␈αthink␈αI␈α
must
␈↓ ↓H␈↓have␈αmisidenti≡ed␈α
what␈αI␈α
meant␈αby␈αtime-sharing␈α
with␈αwhat␈α
became␈αknown␈αas␈α
multi-programming.
␈↓ ↓H␈↓This␈α⊂is␈α⊂why␈α⊂I␈α⊂complained␈α⊂about␈α⊂the␈α⊂lack␈α⊂of␈α⊂a␈α⊂comprehensive␈α⊂treatment␈α⊂rather␈α⊂than␈α⊂claiming
␈↓ ↓H␈↓novelty.␈α∞ I␈α
was␈α∞probably␈α
trying␈α∞to␈α
avoid␈α∞claims␈α
of␈α∞novelty␈α
in␈α∞order␈α
to␈α∞try␈α
to␈α∞persuade␈α∞Morse␈α
to
␈↓ ↓H␈↓commit the Computation Center to time-sharing.

␈↓ ↓H␈↓        4.␈α∞I␈α∞didn't␈α∞believe␈α∂that␈α∞boundary␈α∞registers␈α∞could␈α∞be␈α∂added␈α∞to␈α∞an␈α∞existing␈α∂machine␈α∞design
␈↓ ↓H␈↓without␈α
slowing␈α
it␈α∞more␈α
than␈α
IBM␈α
was␈α∞likely␈α
to␈α
agree␈α
to.␈α∞ This␈α
proved␈α
mistaken,␈α
and␈α∞when␈α
we
␈↓ ↓H␈↓asked␈α⊃IBM␈α∩for␈α⊃modi≡cations␈α⊃in␈α∩support␈α⊃of␈α∩time-sharing,␈α⊃they␈α⊃had␈α∩no␈α⊃objections␈α∩to␈α⊃Teager's
␈↓ ↓H␈↓proposed form of boundary registers.

␈↓ ↓H␈↓        5.␈α∞My␈α∞memo␈α∞also␈α
doesn't␈α∞mention␈α∞inhibition␈α∞of␈α
input-output␈α∞instructions␈α∞in␈α∞user␈α∞mode.␈α
 I
␈↓ ↓H␈↓forget␈α
when␈α∞this␈α
was␈α∞identi≡ed␈α
as␈α∞part␈α
of␈α∞the␈α
required␈α∞package,␈α
but␈α∞STRETCH␈α
didn't␈α∞have␈α
it,
␈↓ ↓H␈↓although it did have boundary registers.

␈↓ ↓H␈↓        It␈α∂occurs␈α⊂to␈α∂me␈α∂that␈α⊂there␈α∂may␈α∂be␈α⊂an␈α∂M.I.T.␈α∂proposal␈α⊂to␈α∂NSF␈α∂for␈α⊂time-sharing␈α∂research
␈↓ ↓H␈↓submitted␈αabout␈α
1960.␈α Perhaps␈α
the␈αNSF␈α
people␈αcould␈α≡nd␈α
it.␈α The␈α
people␈αinvolved␈α
would␈αhave
␈↓ ↓H␈↓been Morse and/or Teager and/or Dean Arden and/or me.
␈↓ ↓H␈↓αProfessor Jack Schwartz␈↓ ¬|July 7, 1980␈↓ 
nPage 2␈↓ 



␈↓ ↓H␈↓Best regards,



␈↓ ↓H␈↓John McCarthy
␈↓ ↓H␈↓Director
␈↓ ↓H␈↓Professor of Computer Science

␈↓ ↓H␈↓cc: Jack Blackburn