[Tfug] Re: CIS faculty at PCC west disavow existence of Linux classes

[Angus Scott-Fleming]

On 30 Jan 2003 at 8:28, Sam Hart  wrote:

> My main thrust was not to mention how bad or good a particular CS dept
> was, just that, if you want to learn /real/ programming, then I could
> think of no other way than getting involved in some
> Free-Software/Open-Source project. It also teaches teamwork, and
> (possibly) even leadership skils (if the project manager suddenly
> becomes sick of a project and hands it over to you). 

Here are some relevant possibly interesting comments from a former ECE 
instructor who taught programming and other computer-related classes at 
the UofA and is now teaching at Texas Tech in Lubbock.  I forwarded him 
one of the messages in this thread to see if he had any comments ....

> I believe he is right in some ways: that the ECE dept. has several
> more practical courses than C.S.  However, I think he has a rather
> narrow view of what C.S. is...sounds like he expects the focus of a CS
> program to be to teach good coding skills.  From my perspective, at the
> Uof A, the study of computers has evolved in two distinct paths.  The
> practical side, studying hardware, problem solving approaches, and the
> engineering application of computers is in the realm of the Computer
> Engineering wing of ECE.  The theoretical, studying issues like the
> mathematical rigor of fault tolerence, development of new paradigms for
> computer languages, etc., is the realm of the C.S. folks, who, by the
> way, are in the college or arts and sciences at the UofA, not the
> college of engineering.  Creating good programmers is generally not a
> high priority of computer science these days...that has become a
> practical problem, just as practical application of mathematics is done
> primarily in the engineering school, not the math dept.  When I did my
> masters there, I don't believe the C.S. dept. even had an undergraduate
> program, although they did have some undergraduate courses. 

> (Personally, I suspect that the best place for a straight-forward
> this-is-how-to-get-it-done course for developing coding skills these
> days is a place like Pima C.C., or even a technical school like Devrie
> (sp?) up in Phoenix area.  Those places focus almost exclusively on
> "how", less on "why"). 

> Even here at Texas Tech, where the CS dept. is in the college of
> engineering, our focus is not on creating good programmers.  We do put
> our undergraduates through 5 or 6 courses where coding is a large part
> of the class, but these are all lower level courses.  The upper level
> and graduate courses, where the (supposed) real meat is, are courses on
> things like languages/compilers; computer vision;  neural networks; 
> artificial intelligence;  parallel computing; networking systems,
> networking security;  data compression;  etc. Obviously being able to
> write programs is important for many (but not all) of these, but not
> the central focus. 

> Here is an interesting perspective on the issue:  I took a graduate
> neural networks course here in the CS dept a couple of years ago.  We
> looked fairly deeply into the math underlying the different neural
> networks that we looked at (and coded up).  The coding problems were
> all very theoretical...fun but generally not practical -- solve the
> travelling salesman problem;  create a network to solve the XOR
> problem, etc.  Kudos were awarded for trying out variations, exploring
> where they worked well and where they didn't, etc..  I am currently
> taking a pattern recognition course over in the E.C.E. dept., and in it
> we are looking at implementing many of the same algorithms for
> real-world problems, like recognizing defects in cotton, separating
> fish by type in a processing plant, etc..  However, the algorithms are
> presented largely as: here's the algorithm, code it up, show that it
> works, and, "we won't go into details, but here are some references if
> you really want to understand the underlying concepts."  The bottom
> line: the C.S. course deals with the theoretical, the ECE course deals
> with the practical.  If I were a manager who needed an employee to
> solve a pattern recognition problem in the short term, I'd send him to
> the ECE course. If I were hiring a new graduate who might work on that
> or any number of other problems, I might want a more theoretical
> background. 

> Finally, I think it's highly unrealistic to expect an undergraduate in
> any discipline to be particularly skilled yet.  That has to be learned
> on the job.  All four years of education does is give you enough tools
> to get started.  Remember, a full-time student is only part time on
> anything.  It takes several years of making mistakes full-time before
> one learns how to either avoid them or recover from them gracefully. 

> Feel free to share these thoughts with that discussion group. 

--
Angus Scott-Fleming
GeoApps, Tucson, Arizona
http://www.geoapps.com/   
1-520-290-5038 / fax 1-208-248-3124
+----------------------------------------------
----------+