Gabriel Sechan wrote:
While I firmly believe in seeing an applicant program, the problem
with doing it ahead of time is that they'll find answers on the web,
or get help from others. Otherwise this isn't too different from
things I've asked people to whiteboard program.
I don't consider this a problem. I'm going to make sure that they can
think on their feet, too.
You adjust something simple in the problem statement, or, even better,
just give them an odd case and see how they work it out.
My experience has been that so few people test the edge cases that
everybody misses one that I can hit them with.
One of the best things along this line was actually not a problem
statement, it was simply this:
"Send us 1000 lines of your best code."
That hits so many points it's really quite amazing. What do you
consider good code? Do you have 1000 lines of code? Do you have 1000
lines of code you can *give out* (aka do you program on your own)? What
kind of problems do you solve? What techniques do you tend to use?
What language do you use? What tools do you use? etc.
It's a nice filter. I'm still going to follow it up with questions in
the interview, but it works quickly.
-a
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
