The 1982 George Forsythe Memorial Lectures
by Jerome H. Saltzer
Stanford University Computer Science Department

Lecture #2, Wednesday, January 27, 1982, at 7:30 p.m., in Jordan 040

Intended audience:  Computer science audience with computer systems interest.

Title: INTRIGUING IDEAS ABOUT COMPUTER SYSTEMS

In the last few years. several interestingly powerful ways
of thinking about computer systems have emerged, sometimes replacinq
a long-standing but murky intuition. Four of these ideas will
be discussed, with examples that illustrate why they seem to
be especially intriguing and potentially important:

     a) hints:  separating fundamental structure from speedup trickery.

     b) write-once:  the log is really the data.

     c) end-to-end arguments: only the applications knows how.

     d) failure models separating algorithm from opinion.