Re: [LogoForum] Re: LOGO AND GEOMETRY

"J.W.Procopio" <[EMAIL PROTECTED]> writes:
>Mr. Harvey further suggests that the time-honored LOGO method of
>constructing a circle (REPEAT 360 [FD 1 RT 1] is preferable to the
>use of the MSWLOGO CIRCLE primitive as it applies to the objective
>of teaching a student how to construct a circle.

I think you misunderstood my point.

I didn't really mean to argue that the CIRCLE primitive is no good.  I added
ARC (from MSWLogo, a generalization of CIRCLE) to UCBLogo because I think it's
really convenient!

But your original article argued that instead of inventing improvements to
Logo we should return to its simple roots, that the very basic Logo lessons
are the best ones.  I was pointing out that *your reasoning* should, if
followed consistently, lead you to reject the use of a CIRCLE primitive.

You accept CIRCLE because, by pushing the way circles are drawn under the rug,
it frees the user's attention to focus on more complicated questions, such as
inscribed and circumscribed circles.  This is the power of abstraction, the
key idea in computer science.  And I think that's fine.

But if you're going to use an abstraction that wasn't in the traditional MIT
Logo, you can't consistently object to other people's abstractions, at least
not on a general principle.  You could, of course, argue that some particular
abstraction is harmful rather than helpful -- this is the argument we're
having in another thread about different abstractions for assigning values to
variables, for example.

Another Logo tradition is the slogan "no threshold, no ceiling."  You have
reminded us about the first half; I'm trying to remind you about the second
half!