00100	Dear Mr. Evans:
00200	
00300		The opportunity to learn what IBM is thinking about with  the
00400	M.I.T.  group  was very valuable to me, and I hope what I have to say
00500	in return will be interesting to you and of use.
00600	
00700		First of all, it is very important to me  that  IBM  make  as
00800	good  computers  as  possible.   The  fact that IBM did not implement
00900	really powerful time-sharing system with the 360  meant  that  no-one
01000	else did either.  Therefore, I will do what I can to help.
01100	
01200		I  am  writing  directly to you rather than to George Raidin,
01300	because I fear that he will find  my  concerns  somewhat  beside  the
01400	point  given the problem he has undertaken to solve.  This is because
01500	I think that the problem he has undertaken to solve, that of creating
01600	a  high  level  machine  language  structure,  is beside the point in
01700	meeting the computer needs of the  next  decade.   In  particular,  I
01800	think   it   is  not  the  main  issue  in  determining  whether  the
01900	applications of computers will grow enough  so  that  the  market  in
02000	dollars will expand rather than contract.
02100	
02200		My  only direct comments on the proposed FS structure are the
02300	following:  Its complicattion makes  blunders  possible  which  might
02400	force  a  user to get around the intended data and program structures
02500	and build a new structure on top of it that is not really  of  higher
02600	level  but  merely has the slightly different specifications required
02700	for the user needs.  While this is  a  danger,  I  think  it  can  be
02800	avoided by having enough smart people go over the proposed structure.
02900	Incidentally, I think my fellow computer  scientists  will  be  quite
03000	favorably  impressed,  because  the  idea  of  going  to a high level
03100	machine structure is in accordance with public opinion among computer
03200	scientists  in  so  far  as  this  group can be said to have a public
03300	opinion.
03400	
03500		Now for my own opinion as to what the most important issue
03600	is.  This is the creation of a universally usable time-sharing
03700	system.  The problem has the following aspects:
03800	
03900		1. All applications must be programmable in the system with
04000	acceptable efficiency.  This includes both airline reservation systems
04100	and on-line debugging.  Acceptable efficiency does not
04200	require that greater efficiency cannot be obtained by starting from
04300	scratch.  Instead it means that the advantages of the universal system
04400	outweigh the advantages of starting from scratch just as people program
04500	in FORTRAN even though a faster machine language program could
04600	be written.