"fooler" <[EMAIL PROTECTED]> writes:

> those questions are more on math-oriented problems... i would rather send
> engineering students with an average programming skills rather than a

Not all. Take a look at the ACM problems at http://acm.uva.es
(excellent resource!), and you'll find a number of them use computer
science algorithms quite heavily. Minimum spanning trees, shortest
paths, funky data structures... =)

But yes, a strong math background is always a good thing.

> for example for the first problem which is problem A... you dont need a
> sophisticated programmer.. what you need is a strong foundation of
> mathematics to find the answer....

... or the willingness to brute-force a solution, which will also get
you somewhere. ;) In fact, the judge who formulated this problem
initially thought of a brute-force solution and another judge pointed
out the nice, neat, elegant one-liner.

> what i really expect with programming contest is that.... give a problem and
> a solution and then let the programmer implement it with his or her own

Sometimes contest problems specify time constraints that make it
impossible to solve a problem using brute-force methods. =) However,
execution time is not and probably should not be the only metric of a
program's quality. What about implementation time?

For ACM-style contests, as long as your program runs within the given
time bounds and produces the correct output, it's okay. Cleanliness of code
or flexibility of system design isn't even scored. =)

> programming is all about algorithms... how fast is your algorithm... it
> doesnt matter how big your code is.. what important most in programming is

Programming is not just about picking the fastest algorithm, but about
recognizing the different subproblems and applying or adapting
algorithms you know. I've seen a lot of problems that added twists to
standard computer science algorithms. It thus becomes more than just
calling a library function or memorizing algorithms, but rather
combining bits and pieces of different algorithms. =)

(That said, many, many ACM problems can be solved with brute force. ;) )

Try some of the problems at http://acm.uva.es . You'll find some very
easy, but others will be pretty mind-boggling. <grin> They'll test
your ability to think on your feet, apply previous knowledge to new
problems, follow specifications exactly, and test your solutions.
I always hated having to think of all the cases! ;)

-- 
Sacha Chua <[EMAIL PROTECTED]> - Ateneo CS faculty geekette
interests: emacs, gnu/linux, wearables, teaching compsci
http://sacha.free.net.ph/
--
Philippine Linux Users' Group (PLUG) Mailing List
[EMAIL PROTECTED] (#PLUG @ irc.free.net.ph)
Official Website: http://plug.linux.org.ph
Searchable Archives: http://marc.free.net.ph
.
To leave, go to http://lists.q-linux.com/mailman/listinfo/plug
.
Are you a Linux newbie? To join the newbie list, go to
http://lists.q-linux.com/mailman/listinfo/ph-linux-newbie

Reply via email to