Re: U.S. colleges retool programming classes - Yahoo! News

[David Farber <dave () farber net>]

Begin forwarded message:

From: Andrew Hunt <andy () pragprog com>
Date: June 3, 2007 1:41:16 PM EDT
To: John Adams <jadams01 () sprynet com>
Cc: dave () farber net
Subject: Re: [IP] Re: U.S. colleges retool programming classes -  
Yahoo! News


On Jun 3, 2007, at 12:11 PM, John Adams wrote:

> I wonder--do you feel the same way about physicists versus  
> mechanical/electrical/civil engineers? Chemists versus chemical  
> engineers?

Not my field, so I can't comment.  This is my field, and I see far  
too many folks who might score well on the theory of finite automata,  
but have never used *any* version control system, or debugged a non- 
trivial program, or interviewed a real person to elicit requirements.

> I'm not sure what disturbs me more (and I'm speaking as someone  
> wishes he'd had a graduate program in software engineering  
> available to him): The idea that computing science is a solved  
> field--"Some folks to teach, some folks for compiler design, and  
> we're done"--or the idea that what "we" (we who?) need is more  
> education tailored to the narrow interests of employers.

It's not a narrow interest by any means.  I'm talking about basic  
fundamentals common across individual language technologies and  
disparate areas of applications.

And in fact, from what I hear, many universities have done away with  
some very valuable, generic courses -- topics such as assembly  
language, or survey of programming languages that cover Lisp, Erlang,  
Smalltalk, and other mental weightlifting.  That stuff teaches you to  
*think*, and yet people tell me it's not commonly offered any more.   
Why not?  Because it's hard to grade, as the OP suggested?

> I don't think it's the job of universities to turn themselves into  
> vo-tech schools

And why is it, that whenever the specter of practical application  
enters into the discussion, academics always fall back on that "we're  
not a votech school" argument?

I'm not suggesting teaching the latest point-release of Java, or how  
to use the latest commercial tool set or popular IDE. I *am*  
suggesting teaching undergrads fundamental theories and techniques to  
construct non-toy software programs with other people.  Is that so  
much to ask?  Is that not a reasonable basis to teach *any* engineer  
who will be involved in software in some way, much as the slide rule  
used to be?

> Bypassing that burn-in period is a prime example of the sort of  
> short-term false economy that's ripping the guts out of my  
> daughter's future.

No one said anything about by-passing on-the-job burn-in.  You'll  
always need that.  But those few months aren't enough to play catch- 
up on all the topics missed in the official curricula.

I submit that the prime example of short-term false economy lies in  
churning out graduates who cannot create accurate, reliable, robust  
software programs.  (Now that's fairly wide brush, and I do know of  
some real top-notch programs at some schools who are turning out  
dynamite folks.  But there's far too many of the other kind.)

Regards,

/\ndy



-------------------------------------------
Archives: http://v2.listbox.com/member/archive/247/=now
RSS Feed: http://v2.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com