Google Code Jam Registration Opens Today

Bamfarooni writes "Registration for Google Code Jam 2006 opens today. The event consists of 3 online rounds (September 5-6, September 14 and September 19,) in which participants compete to solve three coding problems faster and more accurately than their competitors. If you've got what it takes, and more of it than anyone else, then you can take home up to $10k for your code-fu."

Comments

What is this...

[__aaclcg7560]

A contest to write bad code that totally jams up the system? Or is it a contest to turn could into edible jam?

Recruiting

[Heikell]

This is Google's way of recruiting the debuggers to sort out the problems with their programs.

Re:Recruiting

[jboker]

i wonder if being near the top of the winners would better your chances of getting a job with google.

i know i'd put that on my resume if i had the chance to interview with them.

Re:Recruiting

[Cybert4]

Probably would. Try the summer of code and puzzle championship as well. All recruiting tools for google. And don't get suckered by people trying to discourage competition (lots of those around). Give them all a try!

Re:Recruiting

[TheRaven64]

Are the problems going to be more interesting this year? I looked at last year's questions, but they seemed very dull; the kind of things any first year computer science student ought to be able to find an optimal solution to.

I don't think I am one of the top 100 coders in the world, and so if I entered I probably wouldn't win any money. This being the case, the only reason to enter would be for the intellectual challenge; something which seems to be missing. And Google wonder why they have a problem recruiting in the UK...

A *whole* $10k?

[Colin Smith]

Hmm, the difference between cost and value again. Beads for the natives mate...

Re:A *whole* $10k?

[infaustus]

Unless you include the value of being able to put this on your resume and generally brag about it...

Re:A *whole* $10k?

[Rakishi]

College students have time to burn and the people who enter probably consider this to be fun (personally I find interesting problems to be fun). Some people choose to value their life by how much they enjoy it and not by how much money they can make. Others may view this as a good way of making connections, a name for oneself or to impress potential future employers.

If I was a true coder and didn't already have plans for the dates in question I would strongly consider it. Seems like a fun and interesting thing to do.

You are right in some ways, note that many of the winners were from the former USSR where the standard of living is lower, unemployment higher (especially among the youth) but education is still strong. To them this is probably both much more fun and provides a better payoff (including a US work visa potentially) than the alternatives.

Re:A *whole* $10k?

[Bamfarooni]

$10k for 3 hours of work? If you can beat that, I sure would like to know about it.

Re:A *whole* $10k?

[Rakishi]

The potential for $10k given 3 hours of work, just like the lottery is the potential for $x million given $1 and 30 seconds of work. Also without preparing (a lot possibly) before hand the chances of getting that are fairly slim.

Re:A *whole* $10k?

[RShizzle]

As a college student, it definitely seems like a cool competition. It isn't neccessarily for the money as much as it for the possible prestige. Perhaps this matters less to professionals with "better things to do with their time".

Also, if you register, there's a place to upload your resume. This seems like a good way to possible create an opening for a future job...What better way to recruit than a programming competition?

Ownership

[Ellidi T]

Does Google claim ownership over the code you submit?

Re:Ownership

[SilentStrike]

It's not like these problems are designed so that you are solving things that Google needs solved. Trust me, you won't be able to put a dent in any sizable problem in an hour. Instead, the problems are designed to be fun and interesting, and to select coders for great problem solving ability.

Perl?

[XanC]

The Google Code Jam 2006 allows coders to utilize Java, C++, C#, VB.NET or Python*

Hey c'mon! VB.NET and no Perl?

Re:Perl?

[Anonymous Coward]

They did it that way to make it fair for the other languages. Otherwise, us Perl guys would dominate the contest.

Re:Perl?

[Anonymous Crowhead]

-1 Troll

They picked object-oriented languages.

Re:Perl?

[Anonymous Coward]

It's about coding, not self-flagellation.

Re:Perl?

[Anonymous Coward]

I'm wondering why they don't allow Lisp. :-(

Re:Perl?

[Anonymous Coward]

Because they want it to take three hours, not half an hour, and they want a result that their Java/C++ monkeys can understand.

Re:Perl?

[Millenniumman]

Maybe they want to be able to read the code you've written, or perhaps they don't want their entire search engine rewritten on one line.

Re:Perl?

[Anonymous Coward]

While I can't imagine why the hell you'd want to use perl, it is interesting that they limit you only to shitty languages that make developing painful.

Re:Perl?

[jt2377]

Troof! however, where's Ruby?

Re:Perl?

[kemo_by_the_kilo]

Hey its a challenge isnt it?

Re:Perl?

[Anonymous Coward]

ruby should be there as well

Re:Perl?

[TheRaven64]

Part of being a good programmer means picking the right tool for the job. I looked at some of the questions last year, and found I could answer them in about half as much Smalltalk as Java. The Smalltalk was also more readable, and more flexible. Programming in any of the languages they list is like typing with one hand; you might want to try it for a bit, but it's not something you would pick when you needed to be productive.

Re:Perl?

[Geoffreyerffoeg]

Yes, Perl code would have a better chance, because there's a part of the competition where you can challenge other people's solutions - and everyone's gonna challenge the other languages before they try to read Perl.

Perl is job security through obscurity.

Re:Perl?

[Fahrenheit 450]

Really? So where's OCaml? And Smalltalk? And Scala? And Lisp + CLOS? And many, many more?
The problem us that they only want to offer up some small subset of languages instead of opening it up to any language you wish to bring to the party, which is how most every programming contest should be run. THe ICFP contest allows all languages, you'd think mighty Google could figure out how to open up the door.

Can we enter a bot?

[Cybert4]

That does the programming? In a few years, a bot will clean up this type of thing much like a forklift cleans up at a weightlifting competition.

Re:Can we enter a bot?

[coshx]

While it's true that certain types of questions get repeated in different forms (e.g. the language of the problem might clue you in to it having a dynamic programming solution) I highly doubt that in the next few years (or the next 5-20 years for that matter) a bot will be able to do any more than identify the reuse of a previous question with different wording and solve this.

This is easy to circumvent, however, as more complicated/obscure natural language constructs can be used to confuse the bot (botfuscation). What you're suggesting is that in a few years we'll have bots that are perfectly capable of parsing natural language, understanding the problem asked, and solving it. This means they'll have to perform image recognition as well for the problems that have pictures. I believe this technology is still in its infancy and is nowhere near 100% accurate. But maybe you're more up on the research than I am?

Re:Can we enter a bot?

[monoqlith]

Not only that, he's suggesting that the bot will be able to solve it the best way possible, to win the competition. But the best way possible is not algorithmically determinable, as of yet. If there was a way to solve an abitrary problem through computers - well, I guess that means that computers have achieved something like human intelligence, human beings are Turing machines, computers can program themselves, and they don't need us anymore. Welcome to the Matrix.

This will not happen in the near future if ever.

Re:Can we enter a bot?

[monoqlith]

*corrections: by "through computers" I mean "without solving the problem yourself, just by inputting the problem into the computer"

and the Matrix still needs us to survive for our electrochemical enegy, so I guess we're still useful for something.

Re:Can we enter a bot?

[Rakishi]

And 40 years ago they said it'd be only a few years to a decade before a computer is the world champion in chess. AI is crappy right now, and it can solve certain problems but fails at many others.

Re:Can we enter a bot?

[mindsuck]

I'd argue that in the case of a computer playing chess, it was instructed how to play by humans.

A better analogy would be a computer that is able to take the rules of chess as it's only input and figure out how to play and beat humans on it's own.

Re:Can we enter a bot?

[Millenniumman]

If you can write a bot that automatically writes all the code they want, you should win all the money.

Re:Can we enter a bot?

[Rakishi]

I'd argue that in the case of a computer playing chess, it was instructed how to play by humans.

I agree, my point was simply that AI research has a long history of extravagant claims of things happening "soon" which never pan out.

A better analogy would be a computer that is able to take the rules of chess as it's only input and figure out how to play and beat humans on it's own.

Even with human coding the field isn't doing well, as the best methods can only work on a subset of problems. Deep Blue had its hardware optimized to play chess, without that it would not do well at all. It didn't even play chess as most humans would but rather brute forced the problem. In more complex games AIs fare even worse with 12 years old (very good 12 year olds but 12 year old nonetheless) kicking their digital backsides in Go (too many possible moves for the algorithms used by Deep Blue, for example, to work). The board game Settler of Catan had a PhD thesis AI made for it, and the AI can be easily beat by decent players from what I hear.

Re:Can we enter a bot?

[dosun88888]

I think you'd have more profitable ways to use that bot than winning $10k in a code-off.

Anyhow, I know a bot.

Re:Can we enter a bot?

[Pulzar]

This will not happen in the near future if ever.

I think he was joking, aluding to the poker bots which are often thought, by laymen, to be able to clean out the online players. It should be clear to any programmer that programming bots for this kind of a contest are unlikely to be even attempted in the near future.

Re:Can we enter a bot?

[x2A]

Um, well actually, with the right starting conditions set and a decent genetic programming algorithm, it's already possible for a computer to produce something better than its creators could do manually. So, whilst not being able to parse the question in natural english, once translated by a human, this really isn't so far fetched at all.

Code Jam

[Anonymous Coward]

Is that a sequal to Space Jam?

As with most slashdot questions...

[LordEd]

perhaps you should ask google first? Terms and conditions here

Re:As with most slashdot questions...

[Anonymous Coward]

Short answer... No, you own it, but you must grant Google & TopCoder a license to it in order to win

(f) licenses to TopCoder and Google rights to all information submitted during the tournament (including rights to source code and other executables),

Re:As with most slashdot questions...

[Morkano]

This would seem to be the relevant passage:

(f) licenses to TopCoder and Google rights to all information submitted during the tournament (including rights to source code and other executables)

i want

[Anonymous Coward]

i want to pump my data into googles vast backend

oblig macslash-style article

[Anonymous Coward]

"Registration for Apple Butt Jam 2006 opens today. The event consists of 3 ass-packing rounds (September 5-6, September 14 and September 19,) in which participants compete to solve three aids-ass-gay problems faster and more accurately than their competitors. If you've got what it takes, and more of it than anyone else (GOATSE), then you can take home up to $10k for your buttgay-fu."

False Advertisment?

[Awod]

Or at least poorly written. Want a chance at $165,000? Here's how: Google Code Jam 2006. This competition is a chance for you to use your skills and compete against top programmers from all over the world. Thats on the registration page. On the main page it's slightly different, first prize 10k, 2-10 5k and so on until it all adds up to the suggested 165k. True, it all adds up, but if you wanted to enter and were given a link only to the registration page it can be quite misleading.

Re:False Advertisment?

[Anonymous Crowhead]

Uh, this isn't exactly the special olympics we're talking about here....

Re:False Advertisment?

[Millenniumman]

Maybe if you were really good you could win everything.

Have to use their IDE!

[DeionXxX]

The worst part of this is you have to use their IDE. Which means no auto-completion, no color coding, no help, etc. As someone who relies on that kind of lookup stuff daily for their work, I can't see how any normal professional will do well in that environment. However, I can see how a college student who has to memorize all of the code/functions because they are tested on it and required to write code in class will have an easier time with the contest.

I wonder if they consider it cheating to write your code in another window and paste it into their window to test.

Re:Have to use their IDE!

[coshx]

All they say is that you cannot use TopCoder plugins or modify the applet, but do not consider it cheating to copy the problem statement into another IDE, do your development there, and copy the code back. In fact, a good handwritten (e.g. you wrote it yourself and did not get it from a TC plugin) regular expression or two should suffice to parse the problem statement into class/method stubs and test cases. You tie running that regular expression to a macro and you're all set.

Re:Have to use their IDE!

[Anonymous Coward]

Most people do write code in their IDE of choice, (be it VIM, Visual Studio, etc.)

The rules even allow for using of previously written code, as long as you wrote it.

Re:Have to use their IDE!

[epiphani]

Not at all. Use your own IDE, then cut and paste into theirs when you're happy.

Anyone have any idea on the difficulty of the problems? The language choices suck, but I'm still curious as to what kind of problems they throw at you.

Re:Have to use their IDE!

[SilentStrike]

Here is a list of 10 problems used in the google code jam india. This was found with with google query google code jam 250 (250 stands for the point value of the easy problem).

You can practice on similiar problems in the same environment in the topcoder testing arena. Register for topcoder here. Referrer rrenaud.

Would be nice...

[julesh]

... if they included some previous questions with their previous results. That way, you could know what these people had done in order to win the prizes they won. Would be much more interesting than a table with apparently meaningless numbers in it.

Re:Would be nice...

[Doggan]

Here is a link to a previous year's problem with a C++ solution. An interesting problem.

Re:Would be nice...

[SimplyI]

Heh, I had the same problem in my ap computer science ab class last year. Though, the teacher told everyone how to do it after a day or so. Of course, by then, I'd done it, become bored, and coded Knight's Tour in C++. (ap compsci classes are in java =/)

Re:Would be nice...

[SilentStrike]

If you want to practice for the Google Code Jam, you can do ordinary topcoder single round matches. Topcoder runs the contest and uses the same technology to host the Google Code Jam as their ordinary matches. You can register for topcoder here. Referrer rrenaud ;). The next two topcoder matches are next Saturday at noon EST, and Thursday, August 24th at 11 am EST.

Re:Would be nice...

[Eli Gottlieb]

Heh. After looking at the problem I just went and coded up a rough version in maybe half an hour or 40 minutes. It would need some correction to handle each cell of the array containing a string that held multiple grid cells (grid is typed as an 2D array of char), but it basically works.

Re:Would be nice...

[Geoffreyerffoeg]

That doesn't look all that hard (famous last words, but still) if you use recursion. Although my solution wins no points for efficiency, here's the pseudocode.

int solutions = 0
array target = ...
matrix maze = ...

procedure step (int distance, coord position)
    if target.length = distance
        solutions++
        return
    next_letter = target[distance + 1]
    foreach direction in {N, NE, E, SE, S, SW, W, NW}
        if maze[position + direction] = next_letter
            step(distance + 1, position + direction)
end procedure

foreach letter in maze
    if target[0] = letter
        step(0, location of letter)

step() will never test the same path twice, and recursion will take care of the nastiness of backtracking. As I said, it wins no points for efficiency. There's a lot of space and time overhead.

Re:Would be nice...

[Anonymous Coward]

The solution fails the 2 second rule on the exhausive search example (quote from page):

hareesh@ncdm153:~$ time ./cjam-tirupati
-1
real 0m2.494s
user 0m2.491s
sys 0m0.001s

Any thoughts or clever ideas ?

Re:Would be nice...

[tmoertel]

Google is being tricksy with their examples for this problem in order to make a simple search appear viable. They do, however, provide a hint of the impending combinatorial explosion in the 2x2 example of all "A" letters, which despite being tiny has 108 solutions. Scaling this example up to the maximum grid size of 50 × 50 and the maximum word length of 50 characters reveals the sinister nature of the problem: there are about 50 × 50 × 8^49 (= 4.5e47) legal paths, and a straightforward search must trace all of them, which isn't computationally feasible. (See my other post in this thread for a link to more information on the problem.)

Re:Would be nice...

[mm05]

Thanks. I am curious if anyone noticed a relationship between last year problems and this year Google "secondary services" from an implementation point of view? It is at least difficult to find last year problems on their site and it seems to be something to make available Easily to new players?

Hopefully somebody can code up a way...

[wwiiol_toofless]

to delete the record of every google search ever submitted. evar. especially the ones about "how to heal bacne".

Re:Hopefully somebody can code up a way...

[pembo13]

why?

Re:Hopefully somebody can code up a way...

[wwiiol_toofless]

because I'm cultivating anthrax in my back fat, and I don't want teh Dept. of Domestic Surveillance on my ass.

Fraud..

[destiney]

I declare fraud!

Languages
The competition is available in five programming languages - Java, C++, C#, VB.NET and Python.

What abour Ruby? You include Python, but not Ruby? Fraud!

Re:Fraud..

[Ougarou]

Hey, if you include Ruby with it's blocks, you need to include Smalltalk to. Yes. Fraud! We still need: Javascript, Ruby, PHP, AWK, Fortran, Perl, Lisp, Prolog, Bash and Smalltalk. (I'm still busy learning the last)

Re:Fraud..

[dmh20002]

the thing is, the topcoder infrastructure supports certain languages and it costs them $$$ to implement additional ones. When there is a payoff to adding Ruby, I'm certain they will add it.

Using Python is problematic anyway as they state that due to slow runtime of python, some problems won't be solvable. They will time out. I expect this would apply to Ruby also. And I would hate to invest the time to solve something in Python/Ruby only to find out that it won't run, then have to port it.

On the other hand, if the productivity claims of Python/Ruby are true, then a more fair contest might weight time to complete the problem vs. runtime. E.g. if you implement a correct solution quickly, then you get more runtime to test. Something like that.

Re:Fraud..

[Anonymous Coward]

From the instructions page:

*NOTE: All submissions have a maximum of 2 seconds of runtime per test case. This limit is used in harder problems to force submissions to be of a certain complexity. Because of the inherent speed differences between Python and the other offered languages is large, some problems may require extra optimization or not be solvable using the Python language.

Considering Ruby (in my experience, at least) is considerable slower than Python, I can see why they left out Ruby.

Re:Fraud..

[SporkLand]

Malbolge would clean up at a competition like this. But if they required it they would have to call it the "Google Brain Jam".

Re:Fraud..

[5plicer]

No C? WTF?

Re:Fraud..

[Millenniumman]

Why not assembly code? Binary?

Re:Fraud..

[Bionic_Baboon]

Erm. . . C++ is a superset of C. Thus C programs are also valid C++ programs.

Re:Fraud..

[jrockway]

Yup, this C (header) works great when compiled by a C++ compiler :)

typedef int protected;
protected private(int public, char *protected);
void class(void);

and so on :)

Re:Fraud..

[Anonymous Coward]

No Morse? WTF?

Re:Fraud..

[achacha]

Where is LOGO?! That turtle is a mean problem solver.

TURTLE GOTO ANSWER

And that's it!

Re:Fraud..

[xynopsis]

You're a fool. C++ can can compile C code!

Re:Fraud..

[xynopsis]

Then why the hell are you using those reserved C++ keywords in your headers in the first place?

Re:Fraud..

[5plicer]

C++, unlike Objective-C, is not a true superset of C.

Re:Fraud..

[Anonymous Coward]

No it can't. C++ is not a superset of C.

Try running:

void main() {
        int class;
}

through a C++ compiler.

Re:Fraud..

[Anonymous Coward]

Using Python is problematic anyway as they state that due to slow runtime of python, some problems won't be solvable.

Yet they offer Java as an alternative? I know, I know. Java can outperform C on certain microbenchmarks. But we're talking about more real-world applications here. And frankly, from everything I've seen, Java just doesn't cut it.

At the labs I work IT at, we use Python, Java, and Fortran for various tasks. We recently transitioned some of our groupware frontends from Java-based solutions to one written in Python. While we did it initially for maintainability, what we found was that the Python software offered the same functionality, but was far quicker and far more memory-efficient than the Java-based software we were using before. We were actually able to decrease the number of servers we were using, switching some of our production servers to backup servers.

I have also talked in the past to some of the scientists who have started using Python for some of their numerical work. While they admit it can't get anywhere near Fortran when it comes to raw performance, they do admit that it makes their code far quicker to write, and much easier to debug. One particularly interesting piece of simulation software I was told about was originally written in Fortran, but transitioned to Java in order to take advantage OO techniques. The performance was extremely terrible, and a rewrite was thus done in Python. I don't recall the exact number, but I remember them saying that the Python implementation was 15 times faster than the Java one, but itself only three times slower than the simulator written in Fortran.

So from my experiences, a Python program is often far quicker (and less memory-intensive) than a Java equivalent. For them to suggest for this contest taht Python is unusable for some tests because of its speed, well, I think they are mistaken. My point is further backed up when they offer Java as an alternative.

Re:Fraud..

[Anonymous Coward]

He's just pointing out that C++ is not a true superset of C, even though it is for most practical purposes.

Re:Fraud..

[bdcrazy]

I would imagine google is looking for ways that are faster than their current methods (hence the runtime limits) and not necessarily interested in how long it took to code it. This way they get to see many different insights into how the problem can be looked at.

vb.net?

[marko_ramius]

The Google Code Jam 2006 allows coders to utilize Java, C++, C#, VB.NET or Python* as a competition programming language.

Can vb.net even be considered a language?

Re:vb.net?

[Anonymous Coward]

Yes. It can.

Re:vb.net?

[Blakey Rat]

Ignorant or elitist? Taking all bets!

Re:vb.net?

[R.Mo_Robert]

I know you were trying to be funny, but ... in all seriousness, VB.NET and C# are very similar in capabilities, especially with their 2005 (.NET Framework 2.0) versions, and I barely even hesitate to say that they are practically the same language with a different syntax, though I believe there still are a few things one can do that the other can't, and vice versa.

I moved over from VB.NET to C# (granted, I was already familiar with C++ and Java, the inspiration for C#'s syntax) without needing anything more than the occasional glance at Visual Studio documentation--as opposed to a book, which is what it took for most of the other languages I know (including the jump from VB 5/6 to VB.NET).

Yeah really...

[FatSean]

If I come up with a kick-ass idea for my employer, I'll stand to benefit much more than simply $10,000.

I guess if you are job-less, this is a cool way to get some credentials of a sort.

Otherwise, it is pretty lame. BOFH never felt the need to prove his skills :D

Re:Yeah really...

[baadger]

> If I come up with a kick-ass idea for my employer, I'll stand to benefit much more than simply $10,000.

Thats right, you get to keep your job

Ruby?

[Cybert4]

Both elegant AND practical. Would love to see that here.

How is VB.NET in there?

[outZider]

I don't get the addition of VB.NET, except if it was a matter of "Well, TopCoder had it available". Even Microsoft doesn't want to believe it exists.

Maybe I'm just irritated that Perl isn't an option. Then again, we'd know what language made it. :)

Re:How is VB.NET in there?

[Linkiroth]

For the first part, they're all object oriented, so Perl is (for the most part) out of the question as per languages to use. Also, if you look at the sheer numbers of people these days that know VB.NET vs. the number of people that know Perl, the number of people who know .NET is significantly higher. Most college students are taught VB.NET as their first language.

Re:How is VB.NET in there?

[Sartak]

For the first part, they're all object oriented, so Perl is (for the most part) out of the question as per languages to use.

I'll ignore the remark about Perl's OOP and go right to the heart of it. Object oriented programming is hardly a part of the TopCoder/Google Code Jam experience. Any language that has the capacity to understand strings and multi-dimensional lists (including C) could possibly be added to TopCoder. One of the biggest complaints against Perl's addition to TC is that it would make the challenge phase (where people look at and challenge eachother's solutions) more difficult. The four languages they use all have similar capabilities and syntax (or at least recognizably similar by a programmer of one of the other languages) -- this means you can still do well without being an expert in all the languages they currently support.

I'd definitely get back into competing in every SRM if TopCoder added Perl. But until then, well, C++ just isn't good enough for me any more.

Also, if you look at the sheer numbers of people these days that know VB.NET vs. the number of people that know Perl, the number of people who know .NET is significantly higher. Most college students are taught VB.NET as their first language.

I suspect you're grossly underestimating Perl's userbase (or perhaps I'm grossly underestimating VB.NET's userbase). VB.NET is only about four years old -- Perl is three times older, and was for a long time the de facto language of the web.

Re:How is VB.NET in there?

[Sartak]

Oops. Perl is nearly five times as old (it was first released in 1987), not three.

Re:How is VB.NET in there?

[jt2377]

VB.net and C# both compiled into MSIL since they got C# there. WHAT THE FUCK IS THE DIFFERENT TO INCLUDE VB.NET? Either STFU or enter the contest. jeebus, do you want some chesse with your whine?

Nice of them to dovetail with DragonCon

[BirdInHand]

Nice of them to schedule it after DragonCon, another geek-fest, Sept 1-4, 2006. http://www.dragoncon.org/

Python, too

[hauntedspaceship]

This will be the first year that Python will be allowed along side the traditional TopCoder languages (Java, C++, C#, and VB). I think it will be very interesting to see how Python, which more often than not get can get more done with less code and less time, will compare to the heavier languages like C#, C++, and Java. I wonder if Guido working at Google had anything to do with it?

Re:Python, too

[kripkenstein]

This will be the first year that Python will be allowed along side the traditional TopCoder languages (Java, C++, C#, and VB). I think it will be very interesting to see how Python, which more often than not get can get more done with less code and less time, will compare to the heavier languages like C#, C++, and Java.

Yes, but read the fine print. Programs must run in under a certain amount of time. And they write that because of it's slower speed, Python may not be able to solve some of the problems. Python's strength is certainly in speed of coding and not execution, admittably, but it's still sad to me that a better solution wasn't found for this problem than "some problems may not be solvable".

Re:Python, too

[powermacx]

Sure, you may be able to write faster but the question is: Will it run faster?

Otherwise, this could be a problem:

*NOTE: All submissions have a maximum of 2 seconds of runtime per test case. This limit is used in harder problems to force submissions to be of a certain complexity. Because of the inherent speed differences between Python and the other offered languages is large, some problems may require extra optimization or not be solvable using the Python language.

From http://www.topcoder.com/pl/?&module=Static&d1=goog le06&d2=instructions

Bad timing for university students!

[5plicer]

The qualification round coincides with my first week of classes :(

Re:Bad timing for university students!

[Bushido Hacks]

Well, what are you waiting for? You've got a week before class starts. Start hacking! *cracks the whip*

Don't forget to include a GANTT chart. Developers love those things.

Re:Bad timing for university students!

[franksands]

Well I work from 9am to 6pm (GMT -3), how am I supposed to compete in this? They should have thought about this better...maybe in the weekends?

Re:Bad timing for university students!

[havoc]

Use your lunch break. Problem solved. BTW, any employer who wouldn't let you take an hour out of your day for something like this is an employer I wouldn't want to be working for.

Ideas for employers?

[Cybert4]

Wow, I come up with kick-ass ideas that lead to a shorter project time--and I get let go sooner! Where do you work???

MOD PARENT UP!

[5plicer]

This is exactly the info I was looking for :)

Re:MOD PARENT UP!

[Anonymous Coward]

Thanks for the informative post. Now, go back to digg.

Minimax

[Cybert4]

That analogy is already true in trivial cases like tic-tac-toe. The simple minimax algorithm (invented by Dr. Min Irving Max) will be a perfect player "on it's own".

Definitely an American code jam

[br00tus]

A look at who is not eligible to participate shows that it includes "citizens of Iran, Cuba..." etc. I have been reading the Advogato blog of roozbeh for a while, an Iranian who is pissed off he can't participate. Here is a middle class Iranian who wants to participate in something like this and he is shown the door. Of course maybe there's some legal reasons for Google to do this, one can look at their kow-towing to the Chinese government to see it wouldn't be for any ethical reason.

Roozbeh is the one who got me thinking about this, about how this fun worldwide code contest is marred by a no thank you to those countries on their hands and knees to the US elite. On the news you'd think all Iranians are screaming bloodythirsty people in the dark ages, these restrictions help people forget that there are normal hackers like anyone like Roozbeh, I guess they want to dehumanize and alienate us from Iranians so it will be easier to go to war with them. I guess with policies like this, the US government is trying to keep Iranians in the dark ages. Then there's Cuba, the only country in Latin America not on its hands and knees to the US (although Venezuela has not been so over the last few years either). I'm not religious but it reminds me of Daniel 3:4-6 "To you it is commanded, O people, nations, and languages, that at what time ye hear the sound of the cornet, flute, harp, sackbut, psaltery, dulcimer, and all kinds of music, ye fall down and worship the golden image that Nebuchadnezzar the king hath set up: and whoso falleth not down and worshippeth shall the same hour be cast into the midst of a burning fiery furnace."

Re:Definitely an American code jam

[DuChamp Fitz]

Indeed.

Re:Definitely an American code jam

[LordEd]

Of course maybe there's some legal reasons for Google to do this

If you read your own link beyond the list of countries, you'd see

with whom we are prohibited by U.S. law from engaging in commerce

I believe the listed countries are being actively boycotted by the US right now (but i can't find a link to verify it).

The obvious reason that people in these countries can't compete is because Google would be unable to legally give any prize to winners in those countries.

Re:Definitely an American code jam

[SimplyI]

"Cash and prizes, including t-shirts, will not be provided to residents of Iran, Iraq, Libya, Sudan or Syria."

It does not say that said residents are not eligible to participate.

And... just speculating, but Google might not want to send stuff to those countries in fear that it might be said to be an American company funding terrorists... Not that it would be true, but they might be trying to prevent some possible bad press. Meh...

Re:Definitely an American code jam

[SimplyI]

Your link is to Google's Summer of Code. The Code Jam terms do not say Iranians are ineligible, just unable to recieve t-shirts, money, etc. Here's the terms for the Code Jam.

Timeline...

[Known Nutter]

seems to be some discrepancies in the timeline?!? http://code.google.com/soc/studentfaq.html#timelin e

Re:Timeline...

[MyLongNickName]

Summer of Code != Code Jam.

With instruction reading skills like that, I don't think you should enter...

Re:Timeline...

[Known Nutter]

wasn't planning on it, thank you very much. after all, i still don't know what an operand is. four hours of sleep + a 13 hour day will screw up your comprehension skills, that I do know.

Re:Timeline...

[alienmole]

after all, i still don't know what an operand is.

Operands are the things that operators operate on.

Re:Timeline...

[Breakfast Pants]

'i still don't know what an operand is. four hours of sleep +'

Funny that immediately after you say you don't know what an operand is, the next words out of your mouth are.. an operand.

Have to use their words.

[Anonymous Coward]

"I wonder if they consider it cheating to write your code in another window and paste it into their window to test."

Contestants must not cheat. All ideas for code and/or challenge submitted must be the contestant's alone. In this case, the word "cheat" means collaboration of any sort -- including, but not limited to, working with another competitor, submitting code that was not fully authored by you, and using resources, software, etc. that are not allowed in the competition. TopCoder and Google reserve the right, in their sole discretion, to remove any competitor from the tournament who they suspect has cheated.

You better ask, but remember the last round is onsite, and any "questionable" methods may disqualify you. No point in wasting all that round one and two time just to be kicked out.

Foot in mouth...

[SimplyI]

Crap. Saw the terms on another link which stated it a bit differently... And missed the other post.

Nothing to see here...

Re:Foot in mouth...

[SimplyI]

Actually... your link is to the summer of code, not code jam... So, I was tentatively right in my original post.

Developers?

[Anonymous Coward]

Does anyone else prefer http://developers.developers.developers.developers .slashdot.org/ over http://developers.slashdot.org?

Age limit

[ComradeSnarky]

It's a real pity you have to be at least 18 to enter the Championship round. I have a friend who finished in the top 10 in Qualifying last year, but he was only 16 so he wasn't allowed to take part in the Championship round.

Publicity

[SgtChaireBourne]

Nah. It's more for publicity than recruiting. Look at the venue. Those in the US that want to go to NYC are already there. Those overseas will have trouble with customs, given that they are 1) foreign, and 2) good at hacking code. It would have been smarter to choose Toronto or Vancouver, where it would be safer and easier for the contestants to travel.

I can see several reasons why it's to Google's, the contest's, and the contestants' advantage to have the final elimination round take place at the same phsycial location, but NYC is so 20th century.

Re:Perl? - or any functional languages

[flux]

I think a more glaring omission is that there is not a single language that could be considered being of the functional language family - or even if you'd count Python as one (I definitely wouldn't), no statically typed functional languages. But I guess the "you need to be able to read other people's code" explains that partially.

Previous problem set?

[Anonymous Coward]

Anybody knows where we can find the previous problem set (and preferrably *with* the test data). Before topcoder was bought by google it used to make this stuff available.

Please let it be a farting contest...

[Craig3010]

Please let it be a .bat contest...

Actually I get paid extra.

[FatSean]

You say that like it is a bad thing. I'd rather keep my well-paid job than get a measly $10,000 and 'recognition'.

I get money for rising to challenges, money for patents filed, money for new ideas. Of course it helps keep me away from the killing floor when job-cuts come arround, but I've survived 11 years...I'm doing pretty well.

This is the Republican America, where 9-5 is a dream. Just accept that you need to work a little harder than most countries...in order to benefit from a richer life-style than most countries, and you will understand.

Einstein would lose

[PietjeJantje]

Good thinking !== Fast thinking

redundant?

[Anonymous Coward]

Interesting how the AC parent was modded redundant, but the same message the next day by a registered user was modded informative.

Thanks! (And another solution)

[tmoertel]

Thanks for linking to the problem, which turned out to be very fun to solve. If you want to see my solution, which I wrote in Haskell, I posted it here: Solving the Google Code Jam "countPaths" problem in Haskell. Thanks again for the link.

Cheers,
Tom