Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Programming IT Technology

Google Code Jam 2003 Announced 303

An anonymous reader says "O'Reilly Developer News is reporting details of the newest Google programming contest, Google Code Jam 2003. Prizes range from t-shirts to ten grand and you can use any programming language you want to solve the increasingly challenging problems." Update by J : ... as long as it's Java, C++, C# or VB.NET.
This discussion has been archived. No new comments can be posted.

Google Code Jam 2003 Announced

Comments Filter:
  • C++ only? (Score:5, Interesting)

    by pigscanfly.ca ( 664381 ) * on Wednesday September 17, 2003 @12:49PM (#6986713) Homepage
    If I write C code that compiles under C++ would that be considerd "C++" (for the purpose of said competition . While I can program C++ if its a timed competition C would be faster for me) .
  • by RobertB-DC ( 622190 ) * on Wednesday September 17, 2003 @12:49PM (#6986714) Homepage Journal
    This is an interesting way of looking for recruits. An individual interview is one thing -- I've seen some real losers get past that process. Google seems to be looking for a way to entice and reward developers who can use their l33t coding sk1llz in a team environment. It's one thing to be able to write good code -- the ability to work with other great coders is valuable indeed.

    Here's what Google values, from their Job Opportunities [google.com] page:
    What we look for when hiring great people:
    * People with broad knowledge and expertise in many different areas of computer science and mathematics, including distributed systems, operating systems, data mining, information retrieval, machine learning, performance optimization, algorithms, user interface design, statistical inference and information theory, and related areas.
    * People with world-class programming skills.
    * People with excellent communication and organizational skills.
    * People who are passionate about their work and are great colleagues.
    * People who enjoy working in a high-energy, unstructured environment on very small project teams to build amazing products used by millions of people every day.
    * People with diverse interests and skills.
    What intrigues me personally is that this contest takes place in an online collaborative environment. Does this mean that Google is considering opening up to remote working -- as in, I can live in Dallas and "work" in the Googleplex? As much as I'd love to work at a place like Google, there's no way I'm moving to Cali-fall-into-the-ocean-fornia [slashdot.org].

  • by Anonymous Coward on Wednesday September 17, 2003 @12:53PM (#6986762)
    But what are the problems they want solved? It sounds like they want h0xerz/crackers, not real problem solvers/programmers.
  • Re:Cheap Labor (Score:5, Interesting)

    by WIAKywbfatw ( 307557 ) on Wednesday September 17, 2003 @12:59PM (#6986836) Journal
    When your stuck on how to code something....
    run a competition and let others do the coding for you.

    oh and don't forget to through in a t-shirt or to for all the hard work


    And lather, rinse, repeat if you're after some original designs for t-shirts for your website...

    To be honest, I like the idea of this Google competition. I'm not going to be winning it (my programming skills aren't fantastic) but it's a great way of fostering relationships between you and your user base, finding good coders (and potential future employees), getting some good code and/or code ideas and, above all, having a little bit of fun with the community as a whole.

    If all companies were like Google then we'd all be happy as Jay and Silent Bob at a weedfest.
  • by jellomizer ( 103300 ) on Wednesday September 17, 2003 @01:02PM (#6986872)
    I don't know if they fixed it or not. But I hated it when I beta tested it a while ago. I found it had trouble compiling anything even a hello world application. The fact that you must get use to using TopCoder is a reason not to use it. Besides I am against programming competitions because they encourage speed in programing not not a good planned layout of thought. The primary reason why most of the code out their is crappie is because their is lack of design and everyone trying to write this super optimized code. So once their is a problem the code is so optimized that it cannot be fixed. But if you win Congrads, if not dont take it seriously you could still be the better programmer.
  • by mikeee ( 137160 ) on Wednesday September 17, 2003 @01:06PM (#6986897)
    When I was in school, one sick $#%@#$$ did a programming problem set in Postscript. You had to send his code to the printer to execute it. :)
  • by Agave ( 2539 ) on Wednesday September 17, 2003 @01:07PM (#6986909) Homepage
    the way to win a TopCoder competition is to not only be analytical enough to figure out the solution but also to able to type quickly and accurately as well as to take shortcuts like using short variable names. I certainly wouldn't want any 'TopCode' in a production system.

    I was disappointed with my experience in the TopCoder competition at Java One because even on day three, when I was comfortable with the TopCoder IDE, knew that speed was king, and I knew the solution right away I was still beaten out by over two minutes. I mean, I knew the solution, typed it in, compiled first time, ran a test case, passed, and submitted the solution. Damn near perfect in my mind but I was very low man on the totem pole time-wise (~4min vs ~2min).

    I guess learning to touch type by MUDing just doesn't cut it :)

    flee
    flee
    flee
  • by autopr0n ( 534291 ) on Wednesday September 17, 2003 @01:23PM (#6987037) Homepage Journal
    I donno, maybe the problems will be more 'google specific' but it sounds like a regular TopCoder compo. I'm glad TC is able to find some sponsors, but this isn't really all that exciting.

    Anyone remember when TopCoder gave out TONS of money? They would give out cash to 3/8th of the participants like 3x as week. It rocked. And it was distributed to each 'room' so you only had to compete against 8 other people (so in order to get cash you only needed to beat 5 people). Despite this, I only placed once :P. They accidentally sent me two checks though.

    The craziest thing was that they started this after the dot-com crash, so everyone thought they were basically insane, but we were happy to take free money :).

    Anyway I might play this time around. Of course, you'll need to do better then n-249, rather then 3 but it still might be fun :)

    TC used to allow C++ though, I wonder why they gave up on that.
  • by Anonymous Coward on Wednesday September 17, 2003 @01:25PM (#6987054)
    It's not about hiring people who think they are the best, it's about hiring people who are the best.

    The problems you are described occur because of management problems.
  • Re:TopCoder! (Score:4, Interesting)

    by Thuktun ( 221615 ) on Wednesday September 17, 2003 @01:31PM (#6987102) Journal
    The only problem is that TopCoder style programming contests select for speed of implementation over quality, efficiency, readability, and maintainability. Worse, since your code is challenged by other coders, the more unreadable and impenetrable your code, the less likely subtle bugs in your code will be detected in the short challenge period.

    Still, it was nice when they were awarding $300 for first place in your room in Division I for contests every week. When they dropped the prize money for weekly matches, there wasn't a compelling reason to spend my time on it.
  • by Anonymous Coward on Wednesday September 17, 2003 @01:33PM (#6987120)
    AFAIK all C++ compilers will compile C code; if they're getting picky, just change all structs to public classes.

    But that's not the real problem. The real problem is they've provided a choice of four languages which are all based in the same programming paradigm. What's missing from their list of approved languages? Well, how about any functional languages -- ML/OCAML, Haskell, heck I'd even settle for LISP! Modern functional language compilers now produce code whose runtime rivals that of C or C++ (of course when your bug rate drops an order of magnitude, and you can concentrate on algorithms instead of low-level implementation, that's the least of your concerns!)

    What about Prolog? If runtime speed isn't an issue, it's an ideal fit for many AI problems. I've even found myself using oddballs like Forth in the real world.

    A good programmer, just like a good physician or a good carpenter, becomes fluent in the tools of the trade and selects them according to the problem domain. Having a programming contest in which only OO languages are permitted is like having a cabinet-making contest in which all work must be done with screwdrivers.

  • by tempny ( 602740 ) on Wednesday September 17, 2003 @01:33PM (#6987121)
    There are many things that make a coder good. Insightful and elegant design are one. But being able to quickly bang out a solution for a self-contained problem quick and dirty style is just as sure a sign of skill, to me at least.
  • by MickLinux ( 579158 ) on Wednesday September 17, 2003 @01:34PM (#6987132) Journal
    Is it possible to have code written ahead of time? I mean, it says "ideas must be your own." But if you've put together a large library of useful functions that you've coded yourself (fourier transform; heap sort; Wheeler-Burrows transform; entropic compression; and so on), could you grab those code sections, and throw them in?
  • by K-Man ( 4117 ) on Wednesday September 17, 2003 @01:39PM (#6987177)
    I've noticed they also bought some of their own keywords pertaining to certain CS topics, eg:

    Inverted Index [google.com]
    Page Rank [google.com]
    Suffix Array [google.com]

    They've even put some individual researchers' names in for sponsored links:
    Udi Manber [google.com]
    Gene Myers [google.com]

    It's interesting, as they seem to have some things but not others. The suffix array stuff is useful for full-text indexing, which they may be interested in, but they don't flag searches on more recent topics in the field.

  • by drpickett ( 626096 ) on Wednesday September 17, 2003 @01:51PM (#6987269)
    I think that this is a neat idea - Of course, nobody here wanted to discuss the pros and cons of the contest, they just wanted to prove to the rest of the world that they could sling senseless esoterica and criticize Microsoft

    Anyhow, all bull$hit aside concerning Lisp, Python, Haskell, Modula-2, Ada, COBOL, ALGOL, FORTRAN, and whatever other nonesense I saw, I'd like to see some of the socially challenged here at /. put their money where their hacking chops are (assuming that their mom's will let them stay up past their bedtime)

    It would be amusing to see those who would engage in mindless holy wars put their programming chops to the test, and earn some legitimate bragging rights

    I agree with a previous poster that it would be useful to have more contests like this - Here is my programming assignment for you:

    Using a computer language of your choice, write a program that will factor a very large integer of arbitrary precision (1024+ digits) in as little time as possible - Points awarded on speed of execution

    8^)

  • by botzi ( 673768 ) on Wednesday September 17, 2003 @02:01PM (#6987338)
    But it can be.
    I've adopted the habit to always search for a rapid solution of a problem and then once I have a functional code I start to optimize,build, rebuild and defragment it in smaller modules.
    I Have a friend who at the same time is uncapable of coding anything at all before writing down some project design notes on paper.
    You'll be surprised how often at the end we arrive at the same solution(at least we're surprised;o))).
    So,back on your point, fast coding != robust, durable, re-usable code, but I don't really see a way to put a note on those for a competition purposes or am I wrong?????
  • by Anonymous Coward on Wednesday September 17, 2003 @02:21PM (#6987550)
    Yes.

    In fact this is how many of hte early competitions went. The upper echelon of competitors also happened to write problems. Sure they would be exempt from competing when it was their problems on the block, but they freqently cavorted their submitted problems between each other.

    What was worse is that TopCoder used to round robin the top people so that most of them competed in their own rooms. This was eventually changed.

    There were countless other "cheats" to winning with TopCoder.

FORTRAN is not a flower but a weed -- it is hardy, occasionally blooms, and grows in every computer. -- A.J. Perlis

Working...