berggeo@nucsrl.UUCP (06/04/87)
As there is some sentiment for a new topic - let me suggest something...
In the latest issue of the Communications of the ACM there was an article
on revising the computer science curriculum. Among the things that struck me
were that the author(s) strongly advocated making the CS curriculum much more
theoretical (more emphasis on data structures and algorithms).
In the proposed core curriculum subjects such as Artificial Intelligence and
Computer Graphics were distributed amongs other classes as opposed to taught
as separate classes (although I think that there were to be advanced classes
taught in these subjects).
Another things that I found noteworthy about this curriculum
proposal was that it took a different approach to books. The curriculum was
viewed as levels. Each level consisted of 3-4 classes. After taking those
classes, the student moved on to the next level. Here is where the difference
comes in - students aren't to by a separate book for each class, but rather a
few books which will be used, to a greater or lesser extent, by all of the
classes at that level. This way the students will (it is claimed) view books
not as use-once-and-sell items, but rather as a professional library - a
resource to use throughout one's professional lifetime (or until the next
paradigm change :-) ).
These ideas appeal to me. I think that a good theoretical background
is of amazing benefit. If one has a good theoretical background as well as
sufficient grounding in actual, hands-on practice, I believe that it prepares
the student for professional challenges much more than a heavily practical
curriculum.
I'm not really sure that distributing the graphics and AI really makes much
difference, so long as one can take senior-level courses which concentrate in
the subject.
The way the texts are used sounds interesting. Is there anyone out there
with a curriculum organized in that fashion? What are the pros and cons once
you actually put this idea into practice?
George Berg
EECS Dept.
Northwestern University
Evanston, IL
berggeo@nucsrl.UUCP
P.S. If I have made any mistakes in remembering the CACM article, I apologize
heartily, and in advance. I welcome corrections.