begin  quoting Andrew Lentvorski as of Fri, Feb 15, 2008 at 10:00:43PM -0800:
> Mark Schoonover wrote:
> >Well, it's going slow - atleast for me. Partially because of outside
> >influences. I'm currently working through section 2.3.2, but I'll tell you,
> >Lisp is starting to drive me a bit batty. Maybe I should stop SICP, and
> >actually learn Lisp.
> 
> Then I recommend "The Little Schemer" and "The Seasoned Schemer".  They 
> look silly, but just do the exercises.  They tend to be in byte-sized 
> chunks that you can do at your own speed.

I should add those to my "want" list... you keep mentioning 'em.
 
[chop]
> >                  I'm thinking I could be fighting previously learned
> >habits from learning assembly, Pascal, C and Perl that Lisp is sufficiently
> >foreign enough that I'm not able to understand what the code is telling me.

I discovered that it took about two hours to twist my brain into a shape
where scheme-think was natural, and that my brain wouldn't stick in that
shape unless I did some scheme every day.

> It is not just you.  Many people complain about the impedance mismatch 
> between Lisp and algorithms.  Lisp maps to recursive/tree algorithsm 
> very well.  Iterative and decision algorithms, not so much.

Mergesort in Scheme is a dream.

> Of course, doing tree-based algorithms in Java and C tends to be a 
> painful experience.

Um.... why?

And do you have an example?

All I can think of is that in C (and Java), you have to define your
tree data structure, or you have to use a clunky generic predefined
one.  It's not until I have to play with XML that I start feeling pain
with tree-structures in Java or C, but I suspect that's a library
misdesign problem.

-- 
I really do like the idea of code blocks
"Here, do this to every one of those!"
Stewart Stremler

-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg

Reply via email to