Slashdot Mirror


Google US Puzzle Championship

friedegg writes "Google has announced their sponsorship of the US Puzzle Championship, which they describe as a "a national online competition to identify America's most logical minds." Two winners will join the US Puzzle Team, and head to the Netherlands for the World Puzzle Championship in October. The US Puzzle Championship will be held Sunday, May 31 at 1pm EDT, but registration closes tomorrow, May 29 at 9pm EDT! Make sure you read the rules of the challenge if you plan to participate. The rules note that "Members of the Canadian puzzle team may also be selected using this test. Unofficial participation is open to all puzzlers world-wide.""

8 of 141 comments (clear)

  1. Heard of the Eternity Puzzle? by Michael's+a+Jerk! · · Score: 5, Interesting

    I want to meet the man who solves it.

    The Eternity Puzzle is a new type of jigsaw. Unlike normal jigsaw puzzles, there is no picture - every piece is the same shade of green on both sides. All we know about the finished result is that it forms a regular dodecagon (12 sided polygon). The pieces don't have bumps or indentations, either, and all the edges are straight lines. This means that, when anyone looks at the puzzle, they can see that there many different ways to put two pieces together. An additional problem is that almost any two pieces can be placed together while leaving space for other pieces to go around them.

    You can tell that this particular jigsaw has been designed to be extremely hard to solve. So hard, in fact, that the inventor has offered one million pounds to the first person who solves it, as long as they do it within the next five years. That is an awful lot of money just for completing a jigsaw, and you might think that it wouldn't take all that long.

    However, when you first start trying to solve it, you'll soon see that there are far too many ways to start which go wrong. Firstly, though, a couple of things to point out about how you can improve your chances of going right.

    With the puzzle you get a backing sheet of paper with some grid lines on it, as well as the exact location of one of the 209 numbered pieces. All pieces differ in shape, so being able to put a unique piece in position will help at least a little. There are also three much smaller puzzles available to buy, similar in idea but with far fewer pieces (less than 30 pieces each). If you solve those then you are told the locations of additional pieces in the Eternity puzzle, so you can fix 4 or 5 of the piece positions immediately.

    The grid lines on the backing paper are also very useful. The backing paper is drawn up into equilateral triangles, just like isometric paper with three sets of parallel lines drawn on it. Each of the pieces can be placed on this grid so that the edges either go along the grid lines, or cut the equilateral triangles exactly in half. So every piece can be oriented in 12 different ways, only one of which will be right.

    The number of ways to orient these pieces, even if you get all the clues available, is 12204. That's just trying to get all the pieces placed at the correct angles, not even trying to put them together on the board! When we start trying to put pieces together, the number of different ways to try becomes truly staggering!

    It is extremely hard to come up with an exact number of ways of putting the pieces together "wrongly". To count them we would need to go through exhaustively checking each case, adding pieces until we couldn't add any more correctly, then taking out one of the pieces and trying again. The estimate I came up with for the total number of ways to attempt to solve it was 10500. So if you tried, just once, to solve the Eternity puzzle, then your odds of getting the million pounds would be about 1 in 10500. Compare this to the odds of the National Lottery - 1 in 14 million. The odds of getting this puzzle right, first time, are about the same as the odds of the same set of 6 numbers coming up as the National Lottery numbers every Saturday for a year and a half.

    Those are just the odds if you try it once. So you might think you could just get a computer to try all the options, and it won't take very long to find the right one. It's a nice idea, and in many problems it's the right way to go. However, the number of different ways to attempt Eternity is so large that even having hundreds of thousands of computers helping out won't really do you much good. If you had one million computers, each testing out 50 million possible ways to solve the puzzle every second, then every day you would be testing less than 1019 possibilities. At that rate, it would take the computers longer than the age of the universe to sort through all the possible solutions.

    As far as I can tell, the million pounds looks safe.

    --

    I'm not Seth.

  2. Re:Slashcode update? by Jugalator · · Score: 3, Interesting

    I think that would be a pretty good idea, since apparently the subscribers aren't (always?) catching those even if they get the chance to preview it. Would be nice to have it like this:

    1. Subscribers first get the chance to preview a story.

    2. If everything is fine, the story is posted, *but* during the first few hours, anyone get the chance to report dupes/fakes or other glaring errors.

    3. When the correction period is out, the post turns Slashdot Green(TM) and it's set in stone.

    --
    Beware: In C++, your friends can see your privates!
  3. How will Google Enforce 'No Cheating'? by Red+Pointy+Tail · · Score: 5, Interesting
    From the 'rulez':

    Outside help of any kind is not permitted. This means: no assistance of any kind from any other person; also no books, calculators, computers, or tools other than items explicitly permitted. You are allowed to use writing implements, erasers, paper, and any items explicitly required to solve a specific problem. (All such items are listed on the Hints and Tips page.)


    How is this enforceable if it's free-for-all over the web?

    Also, from the sample questions from the Dutch version of it, many of the questions seems to yield to a brute-force computational approach.
    1. Re:How will Google Enforce 'No Cheating'? by untaken_name · · Score: 5, Interesting

      Outside help of any kind is not permitted. This means: no assistance of any kind from any other person; also no books, calculators, computers, or tools other than items explicitly permitted. You are allowed to use writing implements, erasers, paper, and any items explicitly required to solve a specific problem. (All such items are listed on the Hints and Tips page.)
      How is this enforceable if it's free-for-all over the web?


      Well, it might not be enforceable for this preliminary test, but if you cheat your way onto the team, you have only huge embarassment waiting for you. I don't think it would be worth it to cheat in this instance. It would come out when you got to the contest and couldn't perform.

    2. Re:How will Google Enforce 'No Cheating'? by Debian+Troll · · Score: 3, Interesting
      Also, from the sample questions from the Dutch version of it, many of the questions seems to yield to a brute-force computational approach.

      Yep, this is a pretty obvious way to approach it. Indeed, a team from the Information Technology and Systems faculty at the Delft Technical University in Holland have published a report on how they did exactly that.

      My Dutch isn't all that great (I spent 12 months as a postdoctoral researcher in their operating systems software distribution group working on a Beowulf-aware version of apt-get, but most of the staff spoke perfect English), but it looks like they deployed a 256-node cluster running a highly tweaked version of Debian GNU/Hurd, and were able to solve the problem in 14 days, 4 hours and 17 minutes of computation time.

      While I'm sure that's not impressive for the folks at Google, what with their massive GNU/Linux clusters for web searching and keeping their apt.sources files up to date, it does go to show how even a smallish CS faculty can crack these tough sorts of problems using the power of clustering, Debian, and apt-get.

  4. Rubiks' cube speed contest? by 192939495969798999 · · Score: 3, Interesting

    I hope they still include a Rubik's cube speed contest. I'll never forget seeing someone solve one on TV in about 3 seconds... amazing!

    --
    stuff |
  5. Re:Here is a link by qortra · · Score: 3, Interesting

    Thanks for that link.

    From what I saw there, it seems to me that describing the competition as a search for the "most logical minds" is probably inaccurate. Although many solutions can probably divined by logic, it seems to me that most of the solutions require a more creative mind to solve within the time limit. Any thoughts?

  6. Most Logical Mind? by istartedi · · Score: 2, Interesting

    the US Puzzle Championship, which they describe as a "a national online competition to identify America's most logical minds."

    Ummm... are the "most logical minds" going to be drawn to a contest where, given that your skill is an unknown, your odds of winning are 1 over all the participants?

    You can expend considerably less labor at many other endevours, and expect a much greater return. You can put $10,000 in corporate bonds these days, and still get $500 or more at the end of the year. Not too shabby. If you don't have the $10,000, just google for cheaper rent, get a 2nd job, or whatever. The really logical (though not particularly scrupulous) minds are fleecing marks in Vegas and scheming on Wall Street all the time.

    With venues like that favoring the success of a "logical mind", why fuss with some silly puzzle contest?

    Prediction: they will attract a lot of people who love puzzles, and the most logical mind within that subset will have a good chance of winning, but they will most certainly not attract the most logical minds of all, unless... there is a mind out there that's so uber that it knows it can solve any puzzle they throw at it with minimal effort. I suspect there are enough egoists who think they are that mind, but probably very few who are that good. So, unless you are the ubermind, why bother?

    --
    For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?