SICP & the Dumbing Down of American Comp Sci 
Author Message
 SICP & the Dumbing Down of American Comp Sci

Hello,

   I have been working in the American software industry
for 20+ years. One book I have come to admire greatly is
"The Structure and Interpretation of Computer Programs"
by Abelson, Sussman, & Sussman or better known to the
initiated simply as SICP!!! This book is like an onion.
Each time you read it you peel off another layer. This
is my third time around!!!

   I think it is a great pity that the academic world seems
to have caved into the  pragmatism dictated by industry and teaches
courses on C++, Java,
etc. I'm not sure  how to counteract this trend. Recently I was in the
Standford U. bookstore. Maybe I wasn't in the right
place and  I missed all the "neat"
SICP-like books because all I saw was the "how-to-make-better"
widgets using C++, Java etc. instead of Computer Science books.

   SICP rightly makes the claim in the beginning of the
book that they will teach the reader Scheme as the book
unfolds. Could C++ be used as basis for a similar book??
I think not. I have written in ANSI C for more than 10
years. Thus I believe my point is the more poignant.

    What do others think??

Regards,

Vasili N. Galchin

Sent via Deja.com http://www.*-*-*.com/
Before you buy.



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci

Posting this sort of thing to comp.lang.{functional,dylan,scheme} is
preaching to the converted, I think.

Matthias

--

Kyoto University, Research Institute for Mathematical Sciences
(remove underscores in mail address; they are there to fight spam)



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci
Hi Vasili.

Quote:
>    I think it is a great pity that the academic world seems
> to have caved into the  pragmatism dictated by industry and teaches
> courses on C++, Java,

C++ and Java are OK, certainly better than Cobol, fortran and Basic.

Java may well provide a route in for functional programming languages (MLj
and Kawa) via the Java virtual machine.  It gets the compiler guru's off the
hook for writing GUI interfaces and other system specific stuff.  For
example, after twelve months of intermittent attempts, I still can't make
the Win32 library for GHC work, but MLj works straight off.  Other than MLj,
the closest I have gotten is to use and extend the Moscow ML Open GL
package, but I haven't been able to set up a proper mouse/keyboard event
loop yet (I'll keep trying though).

Quote:
> etc. I'm not sure  how to counteract this trend.

Simple - use functional programming languages in your workplace and stand up
for your beliefs where it counts.  If that fails, use guerilla tactics!

I administer my workplace's contract and software licencing system (amongst
other tasks).  I use SIOD (Scheme), Javascript and Access 97 VBA (about
2-3000 lines of each) on Solaris, Irix and Windows NT, to automate that
system and publish data on our intranet.  I kept my use of SIOD secret for
about six months (I'm a manager who supervises the Systems Administrator so
I can get away with it - programming provides me with some light relief in
the workplace), but in the end had an argument with my boss about it.  He
accused me of trying to make myself indispensible by using Scheme which he
claimed noone else knew.  After we cooled off, it turned out that Lisp had
been successfully used several years before in our company to convert
Fortran into C (prior to my arrival) and he admitted that Lisp was a good
language!

I still use SIOD here, but I'm thinking of changing over to MLj, Moscow ML,
or one of the Haskell's to replace SIOD for CGI scripts and web site
maintenance.  I hate ML's syntax, but MoscowML and MLj are easier to
install, recompile and use than the Haskell compilers GHC and NHC (under
Windows at least, which is my OS of choice).  Harlequin's Personal Dylan,
Clean and O'Caml are all nice, but I would feel vulnerable with any of those
choices because of the vendor specific nature of those beasts.

SIOD is so small that it is great for sneaking onto your work system, by the
way.

Cheers

Mike Thomas

PS. Who cares if Comp Sci in the US dumbs down?  I live elsewhere.



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci

Quote:
> Posting this sort of thing to comp.lang.{functional,dylan,scheme} is
> preaching to the converted, I think.

Yes, but he needs political rather than religious instruction I think!


Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci

Quote:

>PS. Who cares if Comp Sci in the US dumbs down?  I live elsewhere.

Unfortunately most computing products come from the states.
This includes MacOS, Windows, original Unix (but not Linux), netscape, etc.

Both MacOS and Windows remained ugly kludges for so long because
the relevant companies (Apple and MS) had an anti-academic bias
and ignored existing knowledge (e.g. Unix).

Michael



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci



Quote:
> Both MacOS and Windows remained ugly kludges for so long because
> the relevant companies (Apple and MS) had an anti-academic bias
> and ignored existing knowledge (e.g. Unix).

I had always thought that Apple took a positive approach to research and
development (I guess that's different from pure academia), witness the
pirating of the WIMP concept from PARC, and the introduction of Dylan (slyly
reintroducing some on topic subject matter).   They just charge too much in
Australia for their products.

I note that Microsoft seems to be hiring some of the leading lights in the
Haskell compiler writer community.

Cheers

Mike Thomas.



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci

Quote:

>> Both MacOS and Windows remained ugly kludges for so long because
>> the relevant companies (Apple and MS) had an anti-academic bias
>> and ignored existing knowledge (e.g. Unix).
>I had always thought that Apple took a positive approach to research and
>development (I guess that's different from pure academia), witness the

That was later - and last I heard they'd closed down various research groups.

Quote:
>pirating of the WIMP concept from PARC, and the introduction of Dylan (slyly
>reintroducing some on topic subject matter).   They just charge too much in
>Australia for their products.
>I note that Microsoft seems to be hiring some of the leading lights in the
>Haskell compiler writer community.

Again, this is much later - I'm referring to the early days.
Unfortunately MS research seems aimed at incorporating more features
into windows. For example, in order to fix basic problems with bad user
interface design they're looking at incorporating natural language
processing technology.

With respect to Haskell (and returning to this newsgroup) it's
natural for MS to be interested in Haskell - they're probably
running out of ways to make Windows 2000 larger and slower.
Incorporating GHC ought to do the trick ... and once they've
re-written MS Word in Haskell it should slow down and bulk up enough
to force people to upgrade PCs again. Of *course* you need a
septium to do basic word processing.

Michael
PS :-)

Quote:
>Cheers
>Mike Thomas.



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci
Hi Michael.

Quote:
> With respect to Haskell (and returning to this newsgroup) it's
> natural for MS to be interested in Haskell - they're probably
> running out of ways to make Windows 2000 larger and slower.
> Incorporating GHC ought to do the trick ... and once they've
> re-written MS Word in Haskell it should slow down and bulk up enough
> to force people to upgrade PCs again. Of *course* you need a
> septium to do basic word processing.

> Michael
> PS :-)

Actually, if you stand back a bit from a screen with a page of Haskell and a
page of VBA, they both look the same!


Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci
Hello,

   I have been working in the American software industry
for 20+ years. One book I have come to admire greatly is
"The Structure and Interpretation of Computer Programs"
by Abelson, Sussman, & Sussman or better known to the
initiated simply as SICP!!! This book is like an onion.
Each time you read it you peel off another layer. This
is my third time around!!!

   I think it is a great pity that the academic world seems
to have caved into the  pragmatism dictated by industry and teaches
courses on C++, Java,
etc. I'm not sure  how to counteract this trend. Recently I was in the
Standford U. bookstore. Maybe I wasn't in the right
place and  I missed all the "neat"
SICP-like books because all I saw was the "how-to-make-better"
widgets using C++, Java etc. instead of Computer Science books.

   SICP rightly makes the claim in the beginning of the
book that they will teach the reader Scheme as the book
unfolds. Could C++ be used as basis for a similar book??
I think not. I have written in ANSI C for more than 10
years. Thus I believe my point is the more poignant.

    What do others think??

Regards,

Vasili N. Galchin

Sent via Deja.com http://www.deja.com/
Before you buy.



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci


Quote:



> > Both MacOS and Windows remained ugly kludges for so long because
> > the relevant companies (Apple and MS) had an anti-academic bias
> > and ignored existing knowledge (e.g. Unix).

> I had always thought that Apple took a positive approach to research
and
> development (I guess that's different from pure academia), witness the
> pirating of the WIMP concept from PARC, and the introduction of Dylan
(slyly
> reintroducing some on topic subject matter).   They just charge too
much in
> Australia for their products.

> I note that Microsoft seems to be hiring some of the leading lights
in the
> Haskell compiler writer community.

     ^^^^^^^^^^^^ E.g. who???

     Vasili

Quote:

> Cheers

> Mike Thomas.

Sent via Deja.com http://www.deja.com/
Before you buy.


Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci

Quote:

> Hello,
>    I have been working in the American software industry
> for 20+ years. One book I have come to admire greatly is
> "The Structure and Interpretation of Computer Programs"
> by Abelson, Sussman, & Sussman or better known to the
> initiated simply as SICP!!! This book is like an onion.
> Each time you read it you peel off another layer. This
> is my third time around!!!
>    I think it is a great pity that the academic world seems
> to have caved into the  pragmatism dictated by industry and teaches
> courses on C++, Java, etc.

I think part of the reason for this is because students these days are a
lot less willing to take "what is good for them". Twenty years ago students
were probably less demanding and more willing to take on trust what they
were told by their professors in general. In Computer Science we have the
additional factor that twenty years ago, students coming in to do a degree
in the subject probably wouldn't have touched a computer before, and wouldn't
have had much idea about the subject. Now most potential students at least
*think* they know about the subject, and think it's all about training in
current popular systems. If you take what they will dismiss as a "too
academic" approach, you will lose the best students to a neighbouring
university which makes its selling point "we teach the language they use in
industry".

Matthew Huntbach



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci

Quote:

>If you take what they will dismiss as a "too
>academic" approach, you will lose the best students to a neighbouring
>university which makes its selling point "we teach the language they use in
>industry".

Oh.  So you teach mostly COBOL, then, right?  :-)
--
Bob Jarvis
Mail address hacked to foil spammers!
Remove "ob" from address to reply


Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci


Quote:

> > Hello,

> >    I have been working in the American software industry
> > for 20+ years. One book I have come to admire greatly is
> > "The Structure and Interpretation of Computer Programs"
> > by Abelson, Sussman, & Sussman or better known to the
> > initiated simply as SICP!!! This book is like an onion.
> > Each time you read it you peel off another layer. This
> > is my third time around!!!

> >    I think it is a great pity that the academic world seems
> > to have caved into the  pragmatism dictated by industry and teaches
> > courses on C++, Java, etc.

> I think part of the reason for this is because students these days
are a
> lot less willing to take "what is good for them". Twenty years ago
students
> were probably less demanding and more willing to take on trust what
they
> were told by their professors in general. In Computer Science we have
the
> additional factor that twenty years ago, students coming in to do a
degree
> in the subject probably wouldn't have touched a computer before, and
wouldn't
> have had much idea about the subject. Now most potential students at
least
> *think* they know about the subject, and think it's all about
training in
> current popular systems. If you take what they will dismiss as a "too
> academic" approach, you will lose the best students to a neighbouring
> university which makes its selling point "we teach the language they
use in
> industry".

> Matthew Huntbach

Matthew,

  I agree part what you say. However, I think greed plays
big part. Both on the part of industry and software engineers.
E.g. I read article in San Jose Mercury News criticising
Silicon Valley companies for only now doing E commerce and
showing lack of real innovation. They compared to a company
started by MIT student doing (I forget!) I think natural language
product. Also, American education is too pragmatic. No mathematics.
E.g. in writing in C, C++, with no garbage collection always
memory leaks! Also big correctness problem. Maintainability
 (which is related to correctness) is big problem; also
managing complexity of software! All related. Sorry my English is
not so good!

Vasili

Sent via Deja.com http://www.deja.com/
Before you buy.



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci


Quote:

> Posting this sort of thing to comp.lang.{functional,dylan,scheme} is
> preaching to the converted, I think.

> Matthias

> --

> Kyoto University, Research Institute for Mathematical Sciences
> (remove underscores in mail address; they are there to fight spam)

Matthias,

   Where to post??

Vasili

Sent via Deja.com http://www.deja.com/
Before you buy.



Sat, 16 Mar 2002 03:00:00 GMT  
 SICP & the Dumbing Down of American Comp Sci

Quote:


> >If you take what they will dismiss as a "too
> >academic" approach, you will lose the best students to a neighbouring
> >university which makes its selling point "we teach the language they use in
> >industry".
> Oh.  So you teach mostly COBOL, then, right?  :-)

No, but we teach mostly Java. This despite the fact that there's a lot wrong
with Java and the *ideal* language for teaching might well be something
else. At the moment, there's enough Java hype around for using it for teaching
to be a selling point. But if Java never really takes off with industry, I
can see it getting categorised as an "academic language" and a reluctant
return to teaching in (the even worse) C++.

Matthew Huntbach



Sat, 16 Mar 2002 03:00:00 GMT  
 
 [ 176 post ]  Go to page: [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]

 Relevant Pages 

1. SICP & the Dumbing Down of American Comp Sci

2. SICP & the Dumbing Down of American Comp Sci

3. SICP & the Dumbing Down of American Comp Sci

4. SICP & the Dumbing Down of American Comp Sci

5. SICP & the Dumbing Down of American Comp Sci

6. SICP and the dumbing...

7. SICP & Dumbing Down Part II (was God CS Books)

8. Dumbing down...

9. Dumbing down....

10. Dumbing Down Scripts to Remove TclX

11. ANN: Book: Great Ideas in Comp Sci with Java

 

 
Powered by phpBB® Forum Software