[net.lang] MIT Tunnel Vision

richw@ada-uts.UUCP (11/12/85)

I recently received some E-mail which simply made my day.
Having been subjected to a few too many years of MIT, it
was very satisfying.  I'm presenting it here to present
a different viewpoint about Lisp and it's place-of-worship.
The author's name has been omitted since this was E-mail
originally; if only I could sign my name to it...

-- Rich Wagner
--------------------------------------------------------------------
>>    "Structure and Interpretation of Computer Programs",
>>     by Abelson, Sussman, & Sussman
>>
>> "A data object that has a type that can be recognized and tested
>> is said to have _manifest_ type."
> 
> My response has nothing to do with the original subject; I just
> couldn't resist the opportunity to knock MIT tunnel-vision (deservedly,
> I think).  One of the biggest problems with the CS program at MIT
> where, of course, Abelson and Sussman^2 wrote that book, is the
> "Everything was invented here" syndrome.  Sometimes, when the outside
> world simply *can't* be ignored anymore it takes on the form "Everything
> was invented here *first*".
> 
> That text is somewhat horrifying.  It does a number of things very
> well.  I would recommend using it (BUT NEVER ALONE!) in a course where
> you'd like to get across the concept of abstraction and the distinction
> between function and form.  But the horrible thing is that they imply
> (through omission, largely) that what is in that book is *all there is
> to know*.  They never say "This is one way to do it".  Instead they say
> "This is THE way to do it."
> 
> The MIT preoccupation with interactive/interpreted languages shows up
> in the passage you used as a reference.  It is simply never mentioned
> in the text that WHEN you try to figure out an object's type can be of
> tremendous importance.  For the vast majority of people in CS, the
> distinction between "static" evaluation and "runtime" evaluation is of
> enormous value.  Yet Abelson, Sussman and Sussman blithely ignore all
> that because it's not LISP or a LISP derivative.  As you have found
> out, this is a disservice to their readers and most especially to
> their students, many of whom will never leave the MIT community, but
> academic inbreeding is another flame...