COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002
C00005 ENDMK
C⊗;



Gary Knott was a student at Stanford, 1968-1970; he took several of my courses
at that time, and served as my teaching assistant at least once.  He was
clearly one of the two or three most promising in his entering class of Ph.D.
students.  I have continued to follow his work with interest.  His 1972 survey
paper on hashing functions was scholarly and definitive, and I continue to use
it.  His research, while mathematically sound, is always driven by practical
motivations of representation and computation on current real-world systems.
While reading his _curriculum vitae_, I found that one of his papers (CACM, 
Jan.`74) solved a current need of mine.

His writing is clear and well crafted.  I know his teaching skill only by faint
memories of his Ph.D. oral examination, but it seems to me he lectured well.

I have not followed his work in mathematical modeling; a superficial inspection
suggests that MLAB is a valuable piece of software, benefiting from careful
attention to the human engineering.

Temperamentally, he is the sort of person one wants as a colleague: hard working,
humane, and thoroughly sane.  Should you ever need a department chairman, I would
think him a good choice.





				Robert W. Floyd
				Professor of Computer Science
				Stanford University