Should we be worried about slipping US performance in the ACM ICPC?

May 13th, 2005

by RMG

My favorite undergrad extracurricular activity made the New York Times today, where Thomas Friedman uses it as a warning about slipping US technical competitiveness in the global market:

On April 7, CNET News.com reported the following: “The University of Illinois tied for 17th place in the world finals of the Association for Computing Machinery International Collegiate Programming Contest. …

“That’s the lowest ranking for the top-performing U.S. school in the 29-year history of the competition. Shanghai Jiao Tong University of China took top honors this year, followed by Moscow State University and the St. Petersburg Institute of Fine Mechanics and Optics. Those results continued a gradual ascendance of Asian and East European schools during the past decade or so. A U.S. school hasn’t won the world championship since 1997, when students at Harvey Mudd College achieved the honor. ‘The U.S. used to dominate these kinds of programming Olympics,’ said David Patterson, president of the Association for Computing Machinery and a computer science professor at the University of California at Berkeley. ‘Now we’re sort of falling behind.’ ”

Now, I don’t know how general US technical education compares to other countries’, but this is a pretty weak piece of negative evidence. I was in the ACM ICPC World Finals in 2003, and my (American) team had a mediocre performance, placing in the middle of the pack. Our coach knew the winning (or second place, I don’t remember) team from Eastern Europe; IIRC they spent about an order of magnitude more time training and preparing for the competition than we did. It seems to me quite possible that, given the well-known educational pressures in some other countries, the difference in ACM ICPC performance says less about the quality of technical education (anybody at the World Finals thoroughly understood all the relevant algorithmic principles involved in the problems) in the US vs. elsewhere as the differing levels of willingness to spend large amounts of time preparing for the competition, far in excess of the level where it has spillover benefits for your general ability as a computer scientist. To some extent this plays into Friedman’s thesis that Americans are lazy and complacent, but it’s possible that in some instances this is a `rational laziness’. I know I, at least, spent a lot of time that year pursuing my own interests that I could have spent doing every problem in the history of the ACM competition or something equally pointless, and I think I’m a better scientist for it.

Update: From the NYT article on the 2003 finals I referred to above:

Many of the teams had been preparing for the contest with the dedication of triathletes, practicing several hours a day. Some schools even devote entire courses to duplicating the conditions, time constraints and types of problems found at the contest….

“We wouldn’t be here unless we planned to win,” said Jan Madey, coach of the Warsaw University team, with the intensity one would sooner expect in a trainer of Olympic gymnasts. He brought his team over a week before the contest, he said, so they would be able to shake off any lingering jet lag and relax.

Were one to judge the state of computer science education solely by this year’s contestants, the best students would appear to be in Eastern Europe and China, not Cambridge, Mass., home to M.I.T., or California, where the California Institute of Technology and Stanford University have traditionally dominated computer education, and definitely not Japan, which had only two teams in the finals.

Some attribute that geographical distribution to the seriousness with which students overseas approach programming contests. The Russian, Polish, Chinese and Canadian teams have been practicing at least 10 hours a week for more than six months, with particular emphasis on the time pressure. Weekly or twice-weekly mock contests gauge their progress.

“You need a lot of practice, like playing the piano,” said Mr. Yang, who has been competing in programming contests since high school. “You have to face the computer every day and solve a lot of problems.”

April 6th, 2005

by RMG

From the US Chess Federation’s online shop’s page for a board/set/bag combo deal:

Please note that although we invite you to indicate your preference for color for the board and tote, colors will be selected randomly.

So, just to taunt you, they will ask you for your color preference, and then ignore it.

Arenas and highways

April 6th, 2005

by RMG

From a recent editorial in Louisville’s Courier—Journal arguing for building a new downtown arena:

You’ll hear a lot about economic benefits of a downtown arena, but study after study also indicates that’s a lot of malarkey; these palaces do not pay for themselves.

Neither do our highways or bridges—but what would life be without them?

Is this guy seriously suggesting that the US highway system does not pay for itself in terms of economic benefits? Does he have any idea how many people use these highways to get to work? More importantly, has he ever seen the bazillion semi-trucks carrying various goods that run continually along I-65?

April 3rd, 2005

by RMG

Kentucky’s basketball obsession makes Language Log.

DARPA shifts funding away from universities

April 2nd, 2005

by RMG

A big topic of discussion within the NLP community (at least at Penn) in recent weeks has been a shift in DARPA’s funding philosophy which drastically reduces money available to universities for NLP research in favor of highly restricted research by private defense contractors. The NY Times has a story on this.

Unnatural Language Processing

April 1st, 2005

by RMG

The funniest AFD thing I’ve seen: 1st Workshop on Unnatural Language Processing (via trochee). Some highlights:


Shared task—————-

Zero-Sum Corpora: Destructive Mining of the Web

Twenty teams. One Web. Three days.
Are you computational linguist enough?

Government panel————————Is Document Classification Easier on Classified Documents?
Information Extraction: A Government and Binding Approach
...
English Unzipfed: No Unigram Left Behind

Suggested paper topics———————————...

  • Scaling Down: From Universal Grammar to Galactic Grammar
  • Corpse Linguistics (transducer decomposition, final states,
    the ultimate epsilon transition …)
    ...
  • To Ken is at ion correct ion
  • Self-Reference and its Implications for This Workshop
    ...
    Machine miseducation track:
    + Overbearingly Supervised Techniques for Very Small Corpora
    ...
    + Aping Syntax: Monkey c-command, Monkey do command

March 30th, 2005

by RMG

“Parental leave programs generally are not very expensive, amounting to at most one or two percent of GDP“ ()

Self-referential sentences

March 29th, 2005

by RMG

“Inexact dates frequently need many of the date/time fields filled in. These kinds of dates cna be confusing to interpret, for human readers as well as computers. For example, if today is Monday, and the document mentions next Thursday, is the Thursday in question the day 3 days from the reference date or 10 days from that date? We are making the assumption that “next” anything is within the week that starts from the upcoming Sunday and ends the following Saturday.” ~ Identifying and Understanding Dates and Times in Email, Mia Stern

Mark: “Wait! When is the upcoming Sunday?”

March 23rd, 2005

by Ryan Gabbard

On Superpolylogarithmic Subexponential Functions (to the tune of Supercalifragilistic Expialidocious). An excerpt from the song (there’s an entire silly paper to go with it):

Chorus:
Superpolylogarithmic subexponential functions!
Faster than a polylog but slower than exponential.
Even through they’re hard to say, they’re truly quinessential.
Superpolylogarithmic subexponential functions!

Now Bennet said in ‘72 to run a program P,
you simulate the program P, but do so reversibly,
The problem with this method is that space is exponential,
so trade off time to save on space—- this really is essential! Oh!

[Chorus]

ECUSA House of Bishops Responds to Primates

March 15th, 2005

by Ryan Gabbard

A while back I wrote about the statement of a meeting of the Primates of the Anglican Communion concerning recent actions on the issue of homosexuality by the Episcopal Church of the United States and the Anglican Church of Canada. Today ECUSA’s House of Bishops released its own somewhat interesting statement in response.

The highlights:

  • They offer a sort of half-apology, expressing ” deep regret for the pain that others have experienced with respect to our actions at the General Convention of 2003 and we offer our sincerest apology and repentance for having breached our bonds of affection by any failure to consult adequately with our Anglican partners before taking those actions”
  • They [the bishops] pledge to neither authorize the public blessing of same-sex relationships nor to perform such blessings themselves between now and the next General Convention in 2006. It doesn’t appear, however, that this restricts priests within ECUSA from performing such blessings more quietly.
  • Most interestingly, in response to a request that ECUSA consecrate no more bishops in homosexual relationships, they decided they will withold their consent on the consecration of all bishops until the next General Convention.