DailyDirt: Help Me, Software, You're Our Only Hope...

from the urls-we-dig-up dept

Everyone relies on software nowadays -- sometimes without even realizing it. But when an entire airline shuts down due to a computer outage, our dependence on technology becomes obvious. (And Skynet is simply reminding us who is really in charge.) Here are a few links on software projects that humans might want to keep an eye on. By the way, StumbleUpon can recommend some good Techdirt articles, too.


Reader Comments (rss)

(Flattened / Threaded)

  •  
    identicon
    Rekrul, Jun 29th, 2011 @ 7:07pm

    The world's best chess software has been caught cheating (because its human author copied from open source chess programs). Sorry, Rybka, the International Computer Games Association (ICGA) says you can't play anymore.

    The headline makes it sound as if the chess program cheated. It's claimed that the author of the program cheated by using code from other chess programs. The program itself doesn't cheat.

     

    reply to this | link to this | view in chronology ]

  •  
    icon
    Greevar (profile), Jun 29th, 2011 @ 7:13pm

    Sore losers. He built a better chess program.

    Just because they didn't get to see the source code and it performs similarly to Fruit and Crafty, they automatically assume that it was taken from those two programs? Could Rajlick not have developed his similar algorithms independently by studying the competition and find his own way to incorporate them? Besides that, it seems Rybka was doing it better since it beat everyone else. I think these people are stuck on the illusion that two people can't come up with the same idea and just assumed he "stole" it. If he's truly genuine, then this is just cruel and damaging to an innovative programmer.

    If he did just simply use their code, then he should give credit, lose his current winnings and be disqualified, but since nobody is allowed to look at it, we can't know for sure. That's certainly not a justification to call him a cheat and take away his winnings retroactively. That's just being petty. "We think you're plagiarizing, but we can't prove it; so we'll just judge you with ambiguous evidence instead." People can be so possessive of ideas and, when threatened, will act indignantly towards those that come to similar discoveries independently.

     

    reply to this | link to this | view in chronology ]

    •  
      icon
      Michael Ho (profile), Jun 30th, 2011 @ 12:39am

      Re: Sore losers. He built a better chess program.

      Greevar,

      It seems like the author has the option of revealing his source code to the contest judges.. but isn't doing so. Maybe Rajlick didn't plagiarize -- but if he didn't, then revealing his source code would certainly put these accusations to rest, no?

       

      reply to this | link to this | view in chronology ]

      •  
        icon
        Greevar (profile), Jul 1st, 2011 @ 9:06am

        Re: Re: Sore losers. He built a better chess program.

        That's like letting the cops search your house to prove you're not a criminal. It's the old assumption that if you've done nothing wrong, then you have no reason to bar them from looking. Yet, he does. He worked on that software and probably doesn't want other people snooping around his code as it would expose his work to people he probably competes with. I have the right to not have others poking their nose in my personal privacy based solely on the accusation of my wrongdoing and demand a search for evidence to the effect. So does this programmer.

        They merely suspect he was cheating and want to breach his privacy to find the evidence that they obviously don't have to support that accusation. They're assuming guilt and demanding he prove his innocence. Since he didn't comply, they automatically assume he cheated and responded as such. We may be a country that believes in "innocent until proven guilty" when it applies to criminal law, but we are very much still of the mentality of "guilty until proven innocent" outside of a courtroom. That's just sad.

         

        reply to this | link to this | view in chronology ]

        •  
          icon
          Michael Ho (profile), Jul 1st, 2011 @ 9:22am

          Re: Re: Re: Sore losers. He built a better chess program.

          Greevar,

          He entered a contest with a prize... I'd expect the judges to be able to view his source code as part of the rules, but maybe that wasn't explicitly stated. In any case, a non-disclosure agreement would solve this problem by allowing only a limited number of people to see his code to verify the absence of plagiarism.

          On the other hand, I do agree that it seems wrong for the judges to award a prize and then try to take it back without solid justification.

           

          reply to this | link to this | view in chronology ]

  •  
    identicon
    A Guy, Jun 30th, 2011 @ 2:57am

    Critical systems shouldn't rely on an operating system to run. Software may be helpful for creating logs, monitoring, and communications with offsite locations but an airplane or should NEVER be designed so that it will stop functioning if a program throws an unhandled exception.

    I find xkcd is often right about these things:
    http://xkcd.com/463/

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      A Guy, Jun 30th, 2011 @ 3:02am

      Re:

      That should read airplanes or insert your --critical system here--.

      Apparently techdirt doesn't like some characters.

       

      reply to this | link to this | view in chronology ]

  •  
    icon
    Marcel Popescu (profile), Jul 1st, 2011 @ 6:24am

    My problem with proof of correctness...

    ... is that you can easily prove that a program calculating the average of two numbers, a and b, by doing

    result = (a + b) / 2

    is correct. And then you run the program and it works, until a year later someone sets a and b to very large values, and the program crashes (or returns a negative value). That's why I prefer testing to proofs, even though I agree that proofs are theoretically better.

     

    reply to this | link to this | view in chronology ]

    •  
      icon
      Michael Ho (profile), Jul 1st, 2011 @ 9:14am

      Re: My problem with proof of correctness...

      Presumably, this crash-proof software tries to test various inputs that might cause invalid output...? So it's not just a "theoretical test" but automated testing.

       

      reply to this | link to this | view in chronology ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Save me a cookie
  • Note: A CRLF will be replaced by a break tag (<br>), all other allowable HTML will remain intact
  • Allowed HTML Tags: <b> <i> <a> <em> <br> <strong> <blockquote> <hr> <tt>
Follow Techdirt
A word from our sponsors...
Essential Reading
Techdirt Reading List
Techdirt Insider Chat
A word from our sponsors...
Recent Stories
A word from our sponsors...

Close

Email This