[Edu-sig] As We May Think: What will we automate?

[Scott David Daniels]

kirby urner wrote:
> One of our Wanderers (think tank in Portland) wrote:
> 
> """
> I expect that teaching Python/Perl/Ruby/Java in the 2000s will be
> viewed with the same scorn in the 2030's. The problem with "flavor
> of the month" languages is that they are passe a month later, as
> better abstractions appear. Such evanescent ways of doing things
> are probably not the basis for life-long learning.
> ... So take a look at "programming in schools" from the viewpoint of
> an adult in 2030, not a 2009 viewpoint, and heaven forbid from the
> viewpoint of the ancient times when you and I were trained. What
> do you wish you had been taught 40 years ago? What was fashionable
> but dated? Extrapolate that forwards, and try to guess what they
> will want, not what you and I consider important /now/. For extra
> points, try to guess what they should be teaching *their* kids,
> for use in the year 2060, and get started on the theoretical
> underpinnings of *that*.
> """
To which I'd reply, this is like some reviewer in Chaucer's day
saying, "Chaucer's writing in Middle English, is such a passing
fancy, let's imagine how people will want to use text messages on
their cell phones."  After all, "Prediction is hard, especially
about the future."

I'll tell you this, in my technical education, I can think of very
little that I learned in any of Knuth's classes that is obsolete,
and that included working with the MIX computer's machine instruction
set.  I like the newer machine (a RISC family of instructions) code,
as it presents issues from modern architectures more clearly, but
getting down all the way to machine code makes you smarter about
what is inevitably slow.
At the other end, Python gives me a language I can talk to another
programmer in, and I can also run parts of the discussion on a machine.
There are other languages that do that, of course, but none that are
so easily communicated to a "random other" without spending more time
talking about the mechanics than about the idea.  I suspect this is why
Kirby likes APL so much, he can easily express large-swath ideas.  For
me, APL too quickly becomes terse little chunks.  But Kirby and I
program about different things.

--Scott David Daniels
Scott.Daniels at Acm.Org