I competed in the ACM programming contest today for the first time (and the last time, as I am no longer eligible after this year). Traditionally, ISU has solved 1 or 2 problems (sometimes none). This year we had a new plan. This plan included lots of training, and some really talented coders. The results can be seen here. We competed at the University of Illinois UC site. I was on the team "ISU Null Pointer Exception."

It was an interesting time to be had by all. My team managed to place second at our site and 18th overall in the region. We solved 6 of the 7 problems correctly. I would also like to note that I submitted my first solution in 6 minutes flat (this is the time from when we see the packet of problems for the first time until I turned in a working solution). I tied for the fastest first solution in the region.

This would have been awesome, had I not then spent the next 45 minutes debugging my correct solution.

Basically this is what happened:

I turned in my solution in record time and got the report back a few minutes later informing me I had a wrong solution (actually, "time limit exceeded", but effectively a wrong solution). I couldn't figure out why my solution wasn't working, so I made a change (that should not have been necessary but fixed potentially bad data on the judges part) and re-submitted. It came back a few minutes later. "Time limit exceeded."

at this point it's 25 minutes into the contest and I've already racked up 40 minutes of penalties for us, so I was starting to feel a bit down about the whole thing. So I started sending in clarification requests to see if I was missing something. Everything came back re-affirming what I had previously thought, except I still didn't have a correct solution. We made one last modification and resubmitted a 3rd time about an hour into the contest. A couple minutes later a judge comes up to us and informs me that due to a "judging error" my first solution had been incorrectly labeled, it had, in fact, been correct.

I wasn't so much upset that I had just wasted 45 minutes of my 5 hours debugging a correct program as much as I was down-hearted about the wrong solutions. Even though we were back in the running (more or less) time-wise, I wasn't as motivated as before, and I didn't feel like I trusted the judging at all. Also, I didn't get working on the second problem and instead of turning it in in the first 20 or 30 minutes (as I should have) I turned it in over an hour after the start of the contest.

Our team had a really nice competitive advantage going in, and that was that I can solve easy problems as fast or faster than anyone in our region. I am good at that sort of thing. But the judging screw-up took that advantage away almost form the get-go. We solved the same number of problems as the team that got 2nd place in the region, we just took longer doing it. Traditionally, in practices, we have always scored amongst the top of the group that solves the same number as we do. I was just really disappointed in that today.

The overall running of the contest was seriously botched. That sort of thing is okay, except when it affects the contest results. And I think this time it did. I think we would have probably had at least 30 fewer tie-breaker points had they not messed this up, and possibly as many as 50. That would have knocked us up several spots in the regional rankings.

I can't blame it all on the judges. We spent wayyy too much time debugging a problem we should have just tossed and re-evaluated. Hopefully my teammates will take care of that problem next year.

Overall, we did a fantastic job. We showed that ISU can produce some very competitive programmers. Next year I hope to see ISU amongst the top 10 in the region.