start > Programming Fun Challenge

Programming Fun Challenge

Created by tmoertel. Last edited by tmoertel 1220 days ago. Viewed 11879 times. #22

[diff] [history] [edit] [rdf]

labels

Category:

attachments

The Programming Fun Challenge is a series of small yet tricky programming challenges that are occasionally offered up to interested programmers in the form of a competition.

PFC Revival

I'm keeping information on the coming great "PFC Revival" on a separate page: PFC Revival. Contribute!

The idea

Here's how it works:

The task organizer creates a challenge, posts it, and sends me (via tom-pfc@moertel.com) a link to the task. I'll then forward the link to the members of the PFC-announce mailing list.
The challenge offers a challenge task, a submission policy, a deadline, and judging notes.
Programmers who accept the challenge write software to solve the challenge task. They submit their solutions (source code) according to the challenge's submission policy.
The judge(s) review the submissions and post the results.
As tradition holds, the winner(s) of the challenge task are offered the opportunity to design the next challenge.

The Challenge

This is the Programming Fun Challenge: The primary objective of is to have fun. That means that challenge tasks must have a high fun-to-yawn ratio: more problem solving, less boilerplate coding. Small yet difficult problems are preferred.

Further, most participants have busy schedules. Good challenge tasks, therefore, can be solved in a few hours by a skilled programmer. (Back in November komet ran a poll in a PFC-related diary entry. Over three-quarters of programming respondents preferred a task that could be solved in an hour or less.)

Summarizing, good PFC challenge tasks have the following characteristics:

Heavy on problem solving
Light on boilerplate coding
Can be solved by busy people, with little time

Notes on solutions

Any programming language is acceptable, provided that open-source compilers or interpretors for that language are available, and provided that the judge can execute programs written in that language on test computers.

Judging

[The following are jacob's opinions, not necessarily held by the rest of the PFC community.]

Criteria for judging should be clear but general — saying "write a good program to do task X" is too vague, and saying "write a program that executes in the smallest amount of time X memory given the full text of the King James Bible" opens the door for overspecialized optimization that many people don't enjoy.
Remember, the point is to have fun, not write mission-critical software; if somebody has a stupid trivial bug in their output format it's good form not to worry about it. Also I think it's a good idea to favor ties over complicated decisions as to whose code is really better, particularly when it comes down to a matter of size or microefficiency.