From: "Bill Poucher" <Bill_Poucher@acm.baylor.edu>
To: <bill_poucher@baylor.edu>
Cc: <acm_support@baylor.edu>
Subject: Problem F - Final Results
Date: Tue, 25 Apr 2000 15:32:16 -0500
Message-ID: <002b01bfaef5$80878990$8f953e81@baylor.edu>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0)
X-MIMEOLE: Produced By Microsoft MimeOLE V5.00.2919.6700
Importance: Normal
Status: RO
Content-Length: 5654
Lines: 134

Coaches,

The finals results of the ACM Programming Contest are posted at

http://acm.baylor.edu/past/Finals/Standings.html

These results reflect changes due to restoring the contest environment and
re-judging Problem F.  For the first time in over 10 years, an error in test
data was not discovered prior to the contest, during the contest, nor during
the post-contest review.

Problem F was solved by two judges in three languages.  These solutions were
run on the contest environment and judged correctly two days before the
contest.  The World Champion team solved the problem in only 17 minutes -
following the tactic that there was only a minute's difference in solving
the problem in the most general way and solving it with the restriction that
the data set was strongly connected.  A third judge visually examined the
test data during the contest.  The flaw was again not detected.  Within one
hour of the end of the contest, with only two reports of difficulty, the
systems were archived, packed, and stored for shipment.

Several additional reports of problems with F came in over the next several
hours.  These were fairly routine, but more insistent than normal.  I
formulated a pathological data set and distributed it to those who continued
to question the results.  I was surprised that no defects were uncovered by
that data set, which was nearly identical to the data set that lead to the
most rejections.  Upon reviewing the results, I was surprised that so many
teams failed to solve problem F.

As soon as it was possible to assemble systems identical to those at the
World Finals and to restore the contest environment, the state of the
contest at contest close was examined.  It was verified that the actual test
data during the contest had a defect.  The results of Problem F were rolled
back, the test data defect was repaired, and the problem was re-judged by
Dick Rinewalt, the Director of Judging, and Jo Perry, the Finals Chief
Judge.  The standings now reflect the higher of the placement at the end of
the contest and after the contest with Problem F re-judged.

Once the defect was discovered all agreed that this was the appropriate
course to take.  I have ruled that the standings posted on April 24, 2000,
are final in accordance with authority vested in me to rule on unforeseen
circumstances.  Furthermore, we have taken measures to leave the contest
environment intact and perform a more thorough audit rather than a review at
the conclusion of the contest.

The following method was used to determine the finals standings.

1.  The top 10 teams in each standing were merged with each team receiving
the highest ranking earned.

2.  The remaining teams were sorted according to the number of problems
solved in the re-judged ranking.  Teams that solved the same number of
problems were assigned the highest ranking in either standings of any team
in the group.  This resulted in only one anomaly.  The 11th place team in
the second standings solved one more problem than the others.  Since ties in
the top ten vacated the 10th ranking, that team was promoted to that
ranking.  Teams outside of the top 10 were then listed in alphabetical
order.  No statistics are printed as the standing is a composite.

Thank you for your consideration as we carefully worked through this
difficulty.  Replacement certificates will be mailed to you for your team if
your position in the standings has increased.  I hope to see you next year
with an even better contest in Vancouver.

Sincerely,

William B. Poucher, Director
The ACM International Collegiate Programming Contest


P.S.  It is reassuring that so many teams actually did solve Problem F.
Here are the teams that solved problem F followed by the number of
submissions submitted up to and included the accepted submission.
Submissions after an accepted submission were ignored.

----------------------------------------------------------------------------
---------
Albert Einstein University Ulm	1
Amir Kabir University of Technology	2
Bangladesh University of Engineering and Technology	3
Belarusian State University	7
Bucharest University	10
California Institute of Technology	4
Carnegie Mellon University	3
Charles University Prague	10
Cornell University	5
Duke University	4
George Mason University	7
Georgia Institute of Technology	2
Harding University	4
Harvard University	1
Indian Institute Of Technology Kanpur	3
Institute of Informatics PMF Skopje	9
Iowa State University	6
Kyoto University	1
Linkvping University	5
National University of Singapore School of Computing	9
National Tsing-Hua University	14
National Taiwan Normal University	10
Moscow State University	21
Massachusetts Institute of Technology	3
Novosibirsk State University	5
Rose-Hulman Institute of Technology	9
Shanghai JiaoTong University	9
South Dakota State University	4
Southern Ural State University	3
St.Petersburg Institute of Fine Mechanics and Optics	1
St.Petersburg State University	1
Stanford University	4
The American University in Cairo	5
The Chinese University of Hong Kong	2
The Johns Hopkins University	4
The University of Melbourne	8
The University of Queensland	6
Tsinghua University	7
Universiteit Leiden	3
Universidad Simon Bolivar	1
Universidad Politicnica de Madrid	6
Universidad de las Americas-Puebla	2
Universidad de Buenos Aires	1
University of Alberta	5
University of Arkansas	1
University of Calgary	4
University of California, San Diego	9
Virginia Tech	1
University of Waterloo	4
University of Toronto	5
University of Texas at Austin	1
University of Pretoria	1
Warsaw University	6
University of Central Florida	5
ZhongShan University	2
University of Washington	4