COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	To:      Gene Golub
C00004 ENDMK
C⊗;
To:      Gene Golub
From:    Bob Floyd
Subject: Appointment of Alan Perlis

NO, I'm opposed.

I served at CMU with Perlis for three years.  I attended several of his
courses, heard numerous public lectures, etc.  He was popular with students,
because his enthusiasm for the subject was contagious, but on every subject
with difficult technical content (e.g., implementary call-by-name in ALGOL,
George Collins' PM system) he got the details all wrong; he didn't understand
them.  He was on the ALGOL GO committee, but didn't realize (he said, not long
after) that they had incorporated recursion. 

His paper on threaded lists is full of bugs; it led nowhere. His Formula ALGOL
system was full of incompatible features; I never heard of anyone using it.
When we offer affiliation to Herb Grosch, we should offer one to Perlis.
(I could say more).

RWF