Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] iPod music compression



Edmiston, Mike wrote:

Even if we can get students to review class examples, textbook examples,
and problem sets... there is a great tendency for them to try memorizing
the detailed steps of each example problem rather than noting the
overall pattern. Their mode of operation might work if their memory is
good and they run into a problem just like one they already memorized.
But they are in trouble if their memory is not good or if they run into
a problem that has a twist on what they have already seen.

That is a big, important issue.

The iPod analogy might help a few students undertand the need to spend
time on pattern recognition rather than exact details, even if the
analogy is not totally correct. To this end I think it is worth a try
to see if it helps.

Alas, IMHO that's where the ipod analogy is weakest.

You can download all 40 of the "top 40" songs into your ipod, and
then you've got a complete grasp of this week's top 40. It's finite,
and quite manageable. More generally, you can load up an ipod with
a set of songs that more-or-less exhaustively covers all your needs.

In analogy, one can imagine memorizing all the examples in the text,
and all the examples discussed in class. Again, that's a finite
list. However, a grasp of those examples does not provide what is
needed. The textbook examples do *not* provide exhaustive coverage
... not anywhere near exhaustive.


If we're gonna make ipod analogies, I would call attention to what
the ipod *cannot* do: It cannot download today the songs that will
be in the top 40 a year from today.

In contrast, the miraculous thing about a good education is that
you can learn principles that will solve not only the problems you
encounter today, but next year and even ten years down the road.

If you learn 10 principles, you can apply them in all permutations and
combinations, which covers a lot of cases. Maybe not all 10 factorial
cases are interesting, but still 10 principles covers a whooole lot
more than 10 cases ... whereas if you simply learn 10 examples by rote,
that only covers those 10 examples.

It never ceases to amaze me how strongly some students are attached
to the idea of just learning all the cases one by one.