Two insights on teaching computer programming

Kirby Urner kirby.urner at computer.org
Tue Jan 3 16:20:14 EST 2006


Kathi Fisler:
> A prerequisite for linking together tools to process data (from a
> students' area of interest) is understanding that data.  Most
> non-majors (and incoming majors, for that matter) have never been
> asked to think about data or information, that it has structure and
> form that governs how to process it.  HtDP's focus on identifying and
> articulating the structure of data and using that to drive process and
> testing makes it valuable for majors and non-majors alike.

Yes. What excites me most about Matthias' project is he's pushing it down to
K12, not just focusing on colleges.  My base position is I think most of my
goals for CS0 (overview of the territory, good grasp of generic concepts)
should be accomplished in K12.  I'd be willing to sacrifice a lot of what
passes for "math education" to make this happen.  Basic numeracy training
needs to include programming in this day and age (how else'll they learn to
write nifty apps for their Nokia cell phones? [1]).

According to this model, a freshman electing a biomed sequence might place
directly into Advanced SQL, as elementary SQL would've been covered back
when we do the Venn Diagrams, starting around Algebra 1.  Pulling from sets
using Boolean filters is a lot of what SQL is about, plus unions and
intersections.  So why no SQL in high school, except in an exclusive few?
Why no XML?  Beats me.  At least in Portland, we're gonna fix that, or risk
losing our reputation as an open source capital.[2]

I'm less familiar with HtDP than SICP or TAOCP, but assume it's everything
its evangelists say it is.  I hope your curriculum continues to make
inroads.  I realize there may have been some bad feelings before, like when
DARPA took Python under its wing around the CP4E initiative, but that's all
water under the bridge by now.  The truth of the matter is:  North America
is simply too huge to give any one process a lock.  We'll be running
numerous programs in parallel.  Upshot:  no matter how good HtDP is, I'm
confidant my own OCN/4D work will pay off as well, even though it's very
different (some call it "extremely alien" -- and I'm not offended).

Kirby
OCN/4D

[1]  http://www.forum.nokia.com/main/0,,034-821,00.html
[2]  http://www.csmonitor.com/2005/1128/p03s02-ussc.html






More information about the Math-thinking-l mailing list