What surprises me is that Dijkstra was very much into program proof of correctness in which case he should have liked the things that can be done in this regard in APL. And now J; see for example the second code section in http://www.jsoftware.com/jwiki/Essays/Birthday_Problem or in http://www.jsoftware.com/jwiki/Essays/88_Hats . With other languages, esp. languages available at the time (1975) he made his statement, such things are more difficult. e.g. you get entangled in things like the for-loop and the loop index i, for an expression such as (365^n) %~ (!n) * n!365 where n is a vector.
Of course having a goto throws a big monkey wrench into the (proof of correctness) works, but one could have a subset of the language which excluded the goto. ----- Original Message ----- From: Devon McCormick <[EMAIL PROTECTED]> Date: Tuesday, November 6, 2007 8:04 Subject: Re: [Jchat] why did Dijkstra dislike APL so much? To: Chat forum <[email protected]> > That's a good guess though in the essay where he says this - > http://www.cs.virginia.edu/~evans/cs655/readings/ewd498.html - > he does not > elaborate. It's a little odd that he would be so harsh on > APL when in the > same essay he makes other statements with which most APLers > would agree: > "The tools we use have a profound (and devious!) influence on > our thinking > habits, and, therefore, on our thinking abilities." > "It is practically impossible to teach good programming to > students that > have had a prior exposure to BASIC: as potential programmers > they are > mentally mutilated beyond hope of regeneration." > "The use of COBOL cripples the mind; its teaching should, > therefore, be > regarded as a criminal offence." > > I find it interesting that he talked a lot about the practice of > programming but did not do it; this might help explain his blind spot > toward APL. > > > On 11/5/07, Raul Miller <[EMAIL PROTECTED]> wrote: > > > > On 11/5/07, metaperl.j <[EMAIL PROTECTED]> wrote: > > > He seems to think it is a programming technique "of the > past" --- what > > > specifically is he referring to? > > > > I believe he disliked APL's right arrow (branch to line number > > computed by expression on the left) and the absence of control > > structures (if statements, while loops, that sort of thing). ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
