Slashdot Mirror
ACM Programming Contest Results Revised
Goonie writes: "After the controversy over the results of the ACM programming competition, it appears that the results are being revised. Check out the new standings. The winner hasn't changed, but some teams have moved up the rankings." Actually, from the look of the page at 4:15GMT this morning, "final rankings are under review." Let's hope the fairest decisions prevail, and that all involved are gracious.
What happened to them sounded pretty lame, I hate it when people can't just fess up and admit they made a mistake.
:)
The rules sounded somewhat weird, but it's probably no weirder than the programming contest I've been doing lately (its monthly, so hold your horses). (and MazeMan can admit he makes mistakes.
---
pb Reply or e-mail; don't vaguely moderate.
pb Reply or e-mail; don't vaguely moderate.
I was very disappointed in this years contest. The automatic turnin program written in Java was not functioning at the Scranton site.
All our submissions had to be hand submitted by the judges which greatly reduced the speed of which we could get back feedback (you could have submitted a program an hour into the contest and chances are you wouldn't have known if it worked or not).
All editorial writers ever do is come down from the hill after the battle is over and shoot the wounded.
I don't see how anyone could have made enough sense of it for there to be a controversy.
Damn good programmers, them.
Got Rhinos?
Does anyone know which teams used what languages? I wonder if anyone hacked something together in perl....
Got Rhinos?
how can one rate a programmer?
i program in many a language
i program so i can use my computer
in different.ways
programming is an art
http://siokaos.org/
This article references another Slashdot article that was posted almost a month ago. Could someone remind me what the issue is, and why this proved to be controversial? I see something about a question F. Why was the question flawed?
Thanks!
when this story first came out i was so happy for those guys. i went to school with both the Johns, and used to code a bit with them.
Ive taken the year off uni (so im not eligable to join in the acm), and working as a coder. My last job was to write an LDAP enabled pop-server, over the weekend and during that time my girlfried (working as a waitress) earned more than me. god the IT industry is in a crap state. or maybe i just struck out with a bad job.
heh. i dont know why im posting here. but on what turns out to be like $8 per hour, i cant afford professional help.
:wq ~ ~ ~ ~ ~
Let's hope the fairest decisions prevail, and that all involved are gracious.
No! I hope that the new results are terrible, and that everyone is pissed of. and that the resulting chaos destroys the ACM and everything it stands for!
ReadThe ReflectionEngine, a cyberpunk style n
simple (lower is better)
- does it get the job done
- does it get the job done every time
- does it get the job done every time, and quickly
- does it get the job done every time, quickly, and doesnt look like something larry wall would write
so yeah, thats one way. they might have more stringent rules then me, but hey
shaolin punk, activist post-industrial
I must have missed the controversy, or perhaps I'm looking at the wrong problem set, (gee, the pdf says 2000 ACM Programming Contest World Finals), but this doesn't seem to be at all vague. It is an application of graph theory. The classic question is 'what is the shortest route from node A to node B.' This just asks you to average all the answers to all the possible pairs of A and B.
Now, this does not mean that it is easy to do, or that I remember the algorithm I once learned for finding the shortest path from a to b on a graph, but I don't see why anyone with a bit of computer science knowledge wouldn't be able to understand the problem.
Of course, as I said, I missed the controversy, so that may not have been what the hubub was about.
-Matt
-Cheetah
I was on the U of Washington team.
I had prob F coded in 30min or so and got it wrong. We spent the rest of the time trying to figure out why!
This *might* have us solving one more problem, but it prolly won't change the relative standings.
Even if we would have gotten that one solved in 30min, I don't think we would have solved another one.
Isn't this the first time the standing have ever been changed???
The teams recently received letters from the contest directors, who say that they have finally decided to "personally re-examine all submissions of Problem F from the archives" and that "the results of the re-examination will be posted in the standings by April 24th."
So I'm not expecting to see any changes until April 24th, which is next Monday.
If they make any changes, teams will be bumped up to a higher position, but no team will have its position lowered. Best of all, they have changed their official policy so that in the future, teams will have access to their programs after the contest is over, and also there will be a standard procedure for regrading after the contest is over, in case this happens again.
I'm particularly happy since there's a reasonable chance that my team's score will improve, but I think everyone should be glad that they're making an effort to keep this a fair contest. I highly recommend this contest to all interested college students. If you haven't already, check out the problem set from this year.
You would think that the hosts of the world's foremost programming competition would take the time to write its own database software with a CGI interface, or at least have the sense to use a more stable, open source package. But NO!! Here's what you get by going to the ACM finals page and clicking "Teams":
Microsoft OLE DB Provider for ODBC Drivers error '80004005'
[Microsoft][ODBC Driver Manager] Data source name not found and no default driver specified
/past/icpc2000/finals/RosterPublicFull.asp, line 11
You should never take life too seriously - You'll never get out of it alive.
So, the administration may have botched a little this year, but it was still a very interesting experience, being in the company of 240+ other intense geeks for four days. Plus you can't really beat a free trip to Florida. To repeat what some other guys have said, you should all go participate in programming contests; they're a lot of fun, and you can often win cool stuff! Oh, and props to IBM for the nicest promotional pens I have ever received. :)
Okay, I know this is much lower scale than the championchip competition, but still. At our regional ACM contest, we had an painfully easy problem involving the angle between the minute and hour hand on an analog clock. The only difficult thing about this problem was the formating. In the ACM contest, it is very explicit that file output is the only thing judged, screen output is not to be even looked at. So when I jumped down at the keyboard to code this, I dumped some very wrong formating to the screen, but the file was fine. The solution got rejected. I commented out the code that printed to the screen, resubmitted, and bingo, answer accepted. Consider me (and my team) extreamly pissed. I was so tempted to throw the judges a really nasty communications form... Because of this screwup we were penalized and we would have gotten 8th place instead of 9th.
I know, minor little rant, but it just goes to show that the ACM *never* is wrong. This still ticks me off.
I was one of the University of Melbourne team (thats Melbourne, Australia).
;), others struggled with it for the entire five hours, trying to find bugs in their correct code.
;), but in this case (having spoken to several of the teams) I think most of those teams just came up against problem F and got stuck.
If it hasn't already been explained, problem F went as follows: find the average shortest path length between all pairs of nodes in a strongly connected directed graph. The judges appear to have put a non-strongly connected graph in the testing - one for which not all pairs of nodes have a connecting path. It was then pure luck whether you had chanced upon the same method for computing the average as the judges, so while some teams solved the problem in the first few minutes of the contest (notably St. Petersburg state U.
Because it was by far the easiest of the problems, many teams began with this question. By the end of the five hours, many perfectly clever teams had not solved anything, due to the stuff-ups on the part of the judges.
Now, making errors is one thing, and it certainly screwed up the contest for everyone, but in some sense errors are forgivable - what was really appalling was the way in which the situation was handled.
A protest regarding the question was submitted (by my team) within a minute or so of the end of the contest - there was no official channel for appeals, so we were just told to write it down on a piece of paper and give it one of the judges. We never received a response of any sort.
After the contest most of the teams were aware of the problem - one team had submitted a program designed to go into an infinite loop if the input graph was not strongly connected, thus working out what error the judges had made. In any case, we took our printouts to verify our solution, and later presented our arguments to the tournament director, Bill Poucher.
I have never met someone so unprofessional, with no concept of accountability or responsibility, and little even of courtesy. I am not sure what his background is, but if Mr. Poucher were the director any public company, he would have been on the street - or in court - in seconds. He refused to listen to any of the requests from the teams or coaches, telling them instead their memories of the contest perhaps weren't clear. Nothing was done and the prizes were awarded, with Bill Poucher announcing at the ceremony that "every problem was solved by at least one team" - so that there weren't any problems.
After the contest, in the face of protests from regional directors and coaches from around the world, with proofs that their solutions had been correct, the contest organisers continued to deny there had been any problem. It took quite a while for them to agree to remark the solutions.
Now, I have no especial gripe at the judges that an error was made, because people make mistakes - it's just an indication that they need better validation on the testing programs. I'm also not especially fussed that my team didn't come first - because in some sense, we were one of the LEAST affected teams. Just look at the results of the American universities. I mean, it's nice to have a laugh at America getting its ass kicked once in a while
What I do have a problem with is the contempt the organisation showed to the tens of thousands of man-hours of work that go into the contest on the part of the contestants. No channel for appeals, no official responses to protests, no means to elect new directors, no accountability, not even a report saying "we made a mistake; we plan to do this to fix it". There was a similar problem in our regional contest this year, and the regional director (Raewyn Boersen) fessed up, and sent around an email describing the problems, the changes she would put in place to the validation of the competition, and the transparency with which future contests would be conducted.
My question for slashdot is, how do you get an organisation like this to to act like it's accountable to the people who enter?
In any case, I'd still recommend entering the contest to any CS students - its an amazing experience, however it goes. I've never learnt as much CS in such a short time as in the week's training leading up to the finals. Ask yourself - if you had to, how fast could you code up a 3D voronoi tesselation? Or a fast constraint solver? Can you find a bug in someone else's uncommented code under time pressure, or look at a problem and say how long it will take to code, and what the best order algoritm is? These are the kind of things you learn.
cheers all,
John FitzGerald, University of Melbourne,
Australia
I suppose, but you rate code based on efficiency? Different people learn different styles... And someone has the gaul to give a rating? It is a challenge, but for what merit?
http://siokaos.org/
Click on Teams in the Finals side bar
Microsoft OLE DB Provider for ODBC Drivers error '80004005'
[Microsoft][ODBC Driver Manager] Data source name not found and no default driver specified
/past/icpc2000/finals/RosterPublicFull.asp, line 11
Why are these results posted in a different form on a different less-professional looking site? What happened to the original? Does anyone still have the old url on hand, and does it show these revised results as well?
As someone who works with computer systems that have been coded in the near to remote past by other people, I can assure you that if your program displays some values, someone somewhere perhaps in the remote future will look at them and try to use them. Even if your specification gives the file output as the designated deliverable, in the real world outside of academia, I would consider any incorrectly formatted output a problem with the code.
Anomalous: inconsistent with or deviating from what is usual, normal, or expected
Anomalous: deviating from what is usual, normal, or expected
Canard: a false or unfounded repor
I don't know Mr. Poucher, but he sounds like the stereotypical American career academic. I've had the misfortune of studying under a few.
They rise to a certain level of recognition, probably higher then they feel (deep down) they deserve. They resolve this insecurity about their competency by believing that they are actually the 'right hand of God', and look down on the unwashed masses with utter contempt, from their Ivory Tower.
'How DARE this student (spoken through clenched teeth) question MY process and MY organization of MY contest!? Without ME he is nothing.'
Typical case of recto-cranial inversion, easily cured with a clue-stick beating and a few months at a REAL JOB.
-- What you do today will cost you a day of your life.
Absolutly, however... We were explicitly told that this would not be a problem.
I gave up on programming contests in grad school, after I discovered that ill-defined questions and untested submission software were the norm. Maybe if I were the sort of person who likes hacking out -- er, reverse engineering -- random code I would have been happier, but IM absolutely correct O real computer science and/or software engineering is about something other than the ability to guess what errors the judges might have made.
I participated in the ACM contests many many years ago. I was somewhat unhappy with the way the international level was run. One year, AT&T sponsored the contest, so they forced us to use SysV Unix and vi. Well, gosh, I had been raised on (gasp) PC editors, where I could use *cursor keys* to move around, and I was plonked into vi. I learned jklh as well as 'i' for insert and 'x' for delete one character. That slowed me down tremendously. Oh well. That wasn't my major complaint though. The main problem was that the contests are in an artificial setting -- your judges are sitting high above, handing down yes/no verdicts, without explaining WHY. Okay, fine. That's just how the contest is run. But what bugs me is that at the ceremony, when they're describing the contest to representatives from industry and government, they say that the contest is very much like the REAL WORLD! HELLO! In the real world your customers can tell you what they don't like. They don't come back and say, "no". It's a completely different environment when you can communicate with your customer -- to get clarification of requirements, to find out what they didn't like, to modify the requirements, and so on. In the ACM contest (back then, at least), you just can't communicate with the judges. You're totally isolated. Then they mislead the press into believing that doing well in these situations (learning unfamiliar tools in an hour, not communicating with customers, using no libraries) are somehow correlated with success in the real world. Bah.
I had a lot more fun at the IEEE contests, run by none other than Mr. Transmeta, Dave Taylor.
- Amit
I competed once in an ACM programming contest (the regional one this year), and have to agree that it's a sketchy affair run by people who really don't know what they're doing.
For example, it was fairly obvious that the problem writers had no real grasp of Java, even though its use was allowed. One problem in the contest involved taking the reciprocal of arbitrarily long integers that are a power of 2 or 5. My group was the only one to finish this problem: all the other groups thought it was one of the most difficult. Considering that Java has its java.lang.BigInteger class, solving the problem was trivial, and a no-brainer.
Go Golden Knights.
:)
Booyah
Worst Sig Ever
I went to the ACM site to check out the full list of teams (looking for my alma mater, Technion) and
found that they need to revise their code as well (see below for the error message).
a) Maybe _they_ should hire a medalist...
b) Obviously, they should have used Linux.
I got:
Microsoft OLE DB Provider for ODBC Drivers error '80004005'
[Microsoft][ODBC Driver Manager] Data source name not found and no default driver specified
/past/icpc2000/Finals/RosterPublicFull.asp, line 11
It's at http://acm.baylor.edu/past/icpc2000/Finals/
and then click on teams.
I wasn't at the overall contest, and didn't see the problems given, but...
:) Look at it from their point of view: Some of the teams got the problem the same way you did...this validates (in your mind) your solution. Other teams struggled but didn't. They then bring accusations that the problem was unfair. There are official channels for complaints as per the rules as listed on the official site.
:)
I was webmaster and technical assistance for the 1999 Southeast Regional Contest. Bill Poucher came to the contest and was quite available. From my interaction with him (albeit on the side of the people administering the competition), he seemed quite reasonable and quite nice.
This is not to say that the contest ran smoothly. A more detailed account of what the tech staff went through is available.
There were some complaints, but they were handled politely, and by the end of the contest, everyone was tired but happy.
Now, back to the international contest. I don't know what happened. I can guess that the judges got hit with a salvo of complaints and joined forces to repel the assualt
I said it before, and I'll repeat it. From my experience, Mr. Poucher is a decent and courteous individual. Managing a contest like this frays your nerves, and when people start bombarding you with challenges and accusations, it is easy to lose patience with them. There are also established procedures for dealing with appeals, and circumventing them by directly approaching the judges isn't likely to influence them positively.
Who am I?
Why am here?
Where is the chocolate?
What is your Slash Rating?
That's a pretty clueless statement coming from the director of a programming competition. Borland C++ in fact ships with two command-line compilers:
First rule of programming: Choose the right tool for the job. Apparently, this fellow skipped that class.
Dr. Poucher's a great guy. You just have to remember, in any contest, that you are catching the contest director in the absolute worst possible circumstances. I hope that you get to meet him some other time. Re: the ivory tower comment -- That may be a problem of some profs, but not Dr. Poucher -- he does a lot of "real world" work on the side.
I have had two classes from Dr. Poucher (one theory and one software development) and I never knew him to be anything but fair to students.
Ask yourself this, though, what would you have done in his situation? You knew that three people, whom you trusted, had checked the problem data. You knew that the problem had been solved on the data already. You've not encountered such a problem in ten years of running such a contest. You knew that every contestant at the contest was way keyed up and running on adrenaline.
I agree about contests though. They are a lot of fun and you do learn a lot. This is true in general, IMHO. Texas has statewide academic contests in high schools -- participation in those takes away all the "test anxiety" many otherwise excellent students face. Programming contest is the same way. It gives you a great confidence, whether you win a contest or not, to just sit down and start slamming out code.
Once again, I am sorry that you had to meet Dr. Poucher under such adverse circumstances!
Zorn
/ is the root of
If "the matrix" is an adjacentcy matrix, just consider an edge between any two nodes. For a graph that would still be connected without the edge, the average shortest distance between nodes for the graph with the edge is bounded above by that for the graph without the edge, but, as the length of that edge goes to infinity, your sum of matrix elements diverges.
Is there something more about graph, such as a space in which it can be embedded?
I hope that this was not this guy's [other] issue with the competition.
--
The shareholder is always right.
They haven't had a judging problem that survived the end of the contest in 10 years.
We have first-hand experience with two judging problems on previous finals. In 1995, we were awarded a balloon only to have it taken away much later. In 1997 we had a problem judged incorrect for some reason (probably due to ambiguity in the output). We moved on to another question and an hour or two later a "correct submission" message popped up for that question.
Judging problems at the finals are rare but do occur. Judging problems at the regionals are more common, because they are run locally and those in charge may lack experience.
In any event, judging problems do occur and dealing with them is part of contest preparation. However, it is important not to overestimate the chance of error - even in the worst-run regional, a judgement of "incorrect" is much more likely an error in the submission than an error in judging.
Gordon Cormack
Coach, Waterloo ACM team