Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Google Programming

Google Introduces Programming Challenge In Advance Of GoogleIO 114

First time accepted submitter X10 writes "Google announced some time ago that they want only developers to attend their Google IO conference. They hinted at developing a 'programming test' that you have to pass before you can register. Now, they have introduced the Input Output machine at the same time they announced that Google IO registration will open on March 27. I take it that registrations will be ordered according to the quality of one's IO machine. Cute idea ..."
This discussion has been archived. No new comments can be posted.

Google Introduces Programming Challenge In Advance Of GoogleIO

Comments Filter:
  • by Anonymous Coward on Thursday March 15, 2012 @07:51AM (#39362759)

    n/t

    • by characterZer0 ( 138196 ) on Thursday March 15, 2012 @08:20AM (#39362959)

      Even if you have an account it asks you "Google Developers is requesting permission to: Know how you are on Google, Perform these operations when I'm not using the application". If you click "No thanks" instead of "Allow access", you get "Access Denied".

      • Company that makes money from harvesting your personal details to send you advertising demands more access to your personal details in exchange for features you may like.

        I don't think we need a film for this one, guys. Animated short at 11.
    • Tried it in Firefox, didn't seem to work; must need Chrome. Tried it in Chrome, didn't seem to work; remembered WebGL is disabled because the Linux ATI driver is blacklisted. Restart Chrome with google-chrome --ignore-gpu-blacklist but it still doesn't seem to work. Maybe it's a puzzle and I just don't get it; I'll check the comments on slashdot.

      This is what the web felt like 10 years ago. Maybe it's my fault for not realizing I had to sign in, but there isn't anything on the page to indicate that, and I do

    • by skids ( 119237 )

      Yep, they must not want any machine I might make more than they want to target me for ads. They get neither.

      • by s73v3r ( 963317 )

        And hundreds of people are willing to fill that void, and come up with machines just as great or greater.

    • by tlhIngan ( 30335 )

      Irony is, most people probably already have a G+ account, even if they never signed up for one.

      Of course, if you're wanting to go to Google I/O you probably are an Android user, which means you have a Google account and by definition, it's a G+ account even if you did nothing...

      Hell, I bet if all you did was sign up for YouTube you have a G+ account.

      • Irony is, most people probably already have a G+ account, even if they never signed up for one.

        A Google+ account is not the same thing as a Google account. You have to sign up for G+, it's not automatic. For example, Google Apps administrators have to specifically enable Google+ access for their users, after which the users then have to use their Apps account to sign up for G+.

    • Yes, wanting to go to Google I/O but don't want to participate in Google's main ID hub... Isn't that a bit like wanting to drive without a license?

  • by perbu ( 624267 ) on Thursday March 15, 2012 @07:54AM (#39362783)
    It fails to load in Chrome but works flawlessly in Firefox. Is somebody not eating their dogfood?
    • by Anonymous Coward on Thursday March 15, 2012 @07:58AM (#39362803)

      Works fine in Chrome. As I tell my users on a regular basis, there's a significant difference between "it doesn't work" and "it didn't work one time".

      • I run dev channel Chrome and it doesn't work for me. (tm)

        Firefox works... for a little while. It froze on me.

        Maybe my IO machine is just too good. I don't get how this is related to development though, it's a cute little game.

        • by s73v3r ( 963317 )

          I'm sure there's something to do with developing algorithms, or visual programming using steps and tweaking parameters in there somewhere.

          I don't know, I'm just having fun.

      • by TheLink ( 130905 )
        Works on both for me, but I find it interesting that the Google IO "canvas" can end up a different size. This means the "end-point" of the Google IO machine can be in a different position on different browsers and computers. And so an endpoint that is reachable for you might not be reachable for others.

        Not sure if this is a bug or a feature ;).
    • by Hentes ( 2461350 )

      It might be just overcrowded now.

  • I take it that registrations will be ordered according to the quality of one's IO machine. Cute idea ...

    So I guess my "blank canvas art" submission won't get be an invite?

  • Not So Correct (Score:2, Informative)

    by Anonymous Coward

    Google has commented on various other comments in the google+ posts for IO that it's a first come first served process on the 27th. There have been some comments about an easter egg in the game which gives a link for preregistration but google has not confirmed. Known IOn members (Those who have attended 3 IO's) have gotten preregistered this week, and the rumor is that Past attendees will get the nod next week. But as of now google has been saying it's a free for all on the 27th. I'm betting they want

  • How do you use this... the ball always seems to bounce violently off of it.
    • It's more of a repeller then a track the ball follows. When it gets close it will bounce back at the complementary angle to its approach to the track for a normal ball. I haven't tried it yet with an Anti grav or Android ball.

      • by Nursie ( 632944 )

        Right, as a novice javascript hacker who's had a couple of beers and is not going to IO anyway - how do you activate the three other ball types?

        I've tried looking at the source, I didn't get anywhere. Give me some good old-fashioned C any day...

        • by Nursie ( 632944 )

          Hmm, turns out I may have just over-thought this as now I've published the empty machine it's unlocked another ball type.

          What exactly is the point of this nonsense? I can't exactly see how it's a developer challenge either.

    • by psergiu ( 67614 )

      Like this:

      One piece I/O machine [google.com]
      (you might have to resize the browser width a bit)

  • Way to go google!

  • by Anonymous Coward

    .. this brings to mind the experience of a friend who applied for a job as technical writer. He was given a "test" to write a fully-specked chapter of documentation for a supplied program undergoing beta testing. On showing up for the interview carrying his "test' assignment, he ran into another interviewee who carried a draft of a different "test" chapter.

  • by Anonymous Coward

    "Sign in with Google+"

    They just want to track people who can program.

  • still looking at the 'spinning record player of Death' icon...

  • The challenge is to port a stable copy of ICS to the Nexus S for OTA delivery?

    I can only dream of the day when the challenge is to write Android in such a way that it runs any hardware that meets basic standards, a la Windows. As opposed to the current model, which seems to be lengthy OS re-writes for each individual handset.

    • by robmv ( 855035 )

      For that to happen, you first need a standard. Ask that to the countless ARM licensors, I am sure they don't want to do that

    • by s73v3r ( 963317 )

      Tell ya what: You go make a standard SoC that ALL of the Android handsets will use, and I'm sure we can make the Android thing work.

      Hell, we're comparing it to PCs. Make two. Shouldn't be that hard, right?

  • by ledow ( 319597 ) on Thursday March 15, 2012 @08:30AM (#39363061) Homepage

    Just built a machine that, depending on some hidden factor, either catches the ball and transports it, throws the ball away off the bottom of the screen, or just misses the catch entirely and lets it smash into the end wall. But the actual result seems random because I don't change anything in-between, just press the spacebar to "launch" a ball.

    Sorry, but I played The Incredible Machine when I was a child, thanks, and it was frustrating enough even when it WAS consistent. I don't program in languages that like to change the parameters at random.

    • state machines are for cowards :p

    • by tom17 ( 659054 )

      I even found that the tutorial is hit & miss. First in FF, the initially placed pendulum NEVER caught the ball. Being an older PC, I figured I should throw chrome at the problem in-case it was some timing problem. Now the same pendulum catches the ball about 50% of the time.

      It's too frustrating to even do the tutorial when you don't even know if the adjustment you have made has helped or not. Bah, looked like it would be fun too!

      • by tom17 ( 659054 )

        Unless, if it is designed to be inconsistent rather than it just being buggy, then I may be inclined to have more of a play with it...

        • by s73v3r ( 963317 )

          I don't think so. On my machine at work, it's pretty consistent. It just really sucks that you have to be so precise with your placements, and you can't even see the last path the ball took, or see the expected path.

          • by tom17 ( 659054 )

            I think that there was more to it than I thought... I think placing new objects may have a gravitational effect on the ball...

    • I wondered if I was the only one that this was happening to. I'd be half way through building something and then all of a sudden the ball starts flying past the contact point where it was hitting before. I'll adjust for that, it'll work fine for awhile more and then third step in it starts missing that contact point. I couldn't figure out if there was some weird variation happening or if my work filter was screwing with whatever physics calculations were happening or if it was supposed to be truly random

    • Yeah. This is really frustrating! I just had one working, and then it mysteriously broke. And I can't imagine that sharing them via G+ would get much more than a bunch of "uhh.. it doesn't work, idiot" responses.
  • by elucido ( 870205 ) on Thursday March 15, 2012 @08:42AM (#39363159)

    A tip for anyone who wants to go through with it. The simple design is usually the better design. The inconsistency are factors which you have to factor for in your design. Kind of like how in life nothing is truly consistent.

    • by Svartalf ( 2997 )

      KISS principle. If I had time I'd have designed it with only one part. I managed with two that's consistent. The less moving parts you can get away with, the better- whether it's a physical device or software.

    • The inconsistency are factors which you have to factor for in your design. Kind of like how in life nothing is truly consistent.

      I deal with issues such as these on a regular basis. You want your physics to run at discrete increments, so if too much time has passed between updates, you don't just process the time delta in one pass, you execute multiple physics steps eg:
      while( playingTheGame ) {
      ProcessInput();
      for ( timeAccum += getElapsedMillsec(); timeAccum >= stepSize; timeAccum -= stepSize )
      UpdatePhysics( stepSize );
      Render();
      }

      This will only update the physics in consistent discrete steps such that the simulation runs t

    • by malakai ( 136531 )

      You can also edit the source directly ( click the </> on the bottom right) and they don't do much checking. You can make a track with 50 points, 200 long. Basically a hover conveyor belt to the output.

      Also, many people don't realize you can move the input and output gates.

  • Haven't played around much with this yet, but it could be used in a similar way to how Minecraft was used to build an ALU, right? Wonder if they'll eventually allow you to chain multiple machines together with multiple outputs...then I think you'd be able to do it...
    • by s73v3r ( 963317 )

      It'd be cool, but I think from what I've seen, it's too small of a work area.

    • I don't think it's possible, since the components don't change state. I tried to use the random escapes of the ball to make it go around in my machine a number of times then on an escapement, get thrown to the output. But the escapes happen in such strange ways, they seem to be just bugs (totally defy the logic of the component), so even that is not much fun, being so random and illogical.
  • There's nothing to support the idea that machine quality is linked to registration.

    I believe the hike to $900 is their response to 'freeloaders' showing up just to get swag.

  • by swsuehr ( 612400 ) on Thursday March 15, 2012 @09:09AM (#39363463) Homepage
    Ugh. Seeing a sentence like this "...architect a machine only you could have dreamt of" makes me think that they may want to consider allowing just *one* non-developer... someone who can write.

    Steve
  • The tutorial level you have to complete in order to actually be able to play with it has an annoying non-dismissable pop up at the bottom which covers up part of the machine. Either it's buggy or they don't like my monitor size.

  • Input: Any text. Output: "Buy (any text) online in canada! Available at your local shopping mart!!!! Great deals (any text) secrets they don't want you to know!!"
  • It does not allow for predictable state changes!? Stupid machine...
  • by znrt ( 2424692 )

    i didin't even get past the fist tutorial puzzle. this is way too dumbass even for a dumbass like me.
    so, googies, here is my application:

    arbitrarily limited component size: fail.
    gui - geekish look : nice
    gui - crappy responsiveness : fail
    lousy mechanics : fail
    embarrassingly idiotic animation : fail.
    random behaviour on a logical puzzle: fail (cool for rpgs, though)
    too pissed off to continue / evaluation aborted.

    well, just the cool looks is not enough, guys.
    it wasn't so hard, you should just concentrate in inn

It was kinda like stuffing the wrong card in a computer, when you're stickin' those artificial stimulants in your arm. -- Dion, noted computer scientist

Working...