/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␈↓ 
⊃April 18, 1979 




␈↓ ↓H␈↓Earl Sacerdoti
␈↓ ↓H␈↓SRI International
␈↓ ↓H␈↓333 Ravenswood Way
␈↓ ↓H␈↓Menlo Park, CA 94025

␈↓ ↓H␈↓Dear Earl:

␈↓ ↓H␈↓        David␈αWilkins␈α
is␈αnow␈α≡nishing␈α
his␈αPhD␈αthesis␈α
entitled␈α␈↓↓Using␈αPatterns␈α
and␈αPlans␈α
to␈αSolve
␈↓ ↓H␈↓↓Problems␈αand␈αControl␈αSearch␈↓␈α
under␈αmy␈αdirection␈αwith␈α
much␈αadditional␈αadvice␈αfrom␈αHans␈α
Berliner
␈↓ ↓H␈↓of␈αCMU.␈α I␈αthink␈αthe␈αthesis␈αwill␈αrepresent␈αa␈αsubstantial␈αadvance␈αin␈αAI,␈αbecause␈αit␈αapplies␈αpattern
␈↓ ↓H␈↓matching␈α∂to␈α∂a␈α∂more␈α∞di≠cult␈α∂heuristic␈α∂problem␈α∂than␈α∂has␈α∞yet␈α∂been␈α∂treated,␈α∂namely␈α∂chess␈α∞tactical
␈↓ ↓H␈↓positions.␈α The␈αone␈αprevious␈αattempt␈α-␈αAlbert␈αL␈αZobrist␈αand␈αFrederick␈αR.␈αCarlson␈αJr.␈α  -␈α
was␈αnot
␈↓ ↓H␈↓very␈αsuccessful,␈αbecause␈αthe␈αnotion␈αof␈αpattern␈αit␈αused␈αwas␈αtoo␈αlimited.␈α Wilkins's␈αpatterns␈αmatch␈αa
␈↓ ↓H␈↓larger␈αpart␈αof␈αhuman␈αthought␈αprocesses␈αand␈αinvolve␈αtime␈αas␈αwell␈αas␈αspace.␈α The␈α
extended␈αnotion
␈↓ ↓H␈↓of␈αpattern␈αused␈αin␈αthe␈αthesis␈αis␈αWilkins'␈αidea;␈αI␈αonly␈αcontributed␈αthe␈αidea␈αthat␈αthe␈αuse␈αof␈αpatterns
␈↓ ↓H␈↓in chess requires such extension.

␈↓ ↓H␈↓        Wilkins␈αhasn't␈αhad␈αteaching␈αexperience␈αbeyond␈αserving␈αas␈αa␈αTA␈αhere␈αand␈αat␈αIowa␈αState,␈αin
␈↓ ↓H␈↓each␈α∞case␈α∞giving␈α∞a␈α∞few␈α∞lectures.␈α∞ However,␈α∂he␈α∞is␈α∞well␈α∞organized␈α∞and␈α∞quick␈α∞in␈α∂conversation␈α∞and
␈↓ ↓H␈↓writes reasonably well, so I don't think he will have di≠culty in teaching.

␈↓ ↓H␈↓        He is pleasant and co-operative and strongly motivated to doing good research.


␈↓ ↓H␈↓Sincerely,



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