A Programming Horror Story

Staring at a screen of seemingly intractable merge conflicts, I wondered how we could possibly be ready in time.

It was the night before the official launch of our much anticipated app, and I was feeling tired, nervous and a little superstitious. And I was not alone, going by the anxious emails and instant messages from my colleagues, and the faxes from my boss.

A gale was blowing outside, and I imagined my apartment’s fake window shutters flapping violently in the wind. The grandfather clock in the hallway suddenly struck midnight—that “witching time of night, when church yards yawn, and hell itself breathes out contagion to this world”, as Shakespeare so overdramatically described it.

For a moment, I felt an overwhelming sense of foreboding—but then realized how silly that was, and slapped myself across the face. Then I realized how silly that was, too (though I didn’t slap myself again, to avoid an infinite loop).

It was the text message from our sysadmin—arriving a few minutes later—that began the disastrous series of events of that night. It read, simply—“Houston, we have a problem.” (Houston is my name.)

“Problem”, it turns out, was an understatement. This was truly a catastrophe …


Would you like to know what happened next? Great! Just tell us yourself, as the story hasn’t been written yet.

Each new post should add the next twist to the story (try to keep it to a sentence or paragraph at most). Make sure to read through the existing posts first, so that you understand where the story is up to.

Have fun!

2 Likes

We had been programming this robot for months on end, with little sleep and lots of stress. The app we wrote was to give it commands on a very high level, like “clean the kitchen”, and it would go there, figure out what needed cleaning, and get to it. It was working really well and responding to our commands the way it should. The one thing that got in the way of releasing the robot to the public was that the AI was a bit too good sometimes. Like it was that night. Our prototype robot had decided it had enough of doing chores, so it had disabled its GPS module, and ran away …

2 Likes

First, the team thought that they should call the police to find it… but the bot was a beta version, unpredictable and unrealiable. The police would not be happy that an experimental robot was around, somewhere.

I had a friend in the FBI who in his free time acted as consultant for successful TV series like Criminal Minds.

I didn’t tell anyone, I didn’t want to raise expectations but I called my friend. The phone ringed and I was feeling all the weight over my shoulders. My friend was not answering. The answering machined answered instead.

I left a message “Call me. ASAP. Urgently. Like right now?”

I could only wait while the team got ready to go outside in some kind of rescue mission.

3 Likes

So how do you locate a Rogue Robot?
Do you run through the streets, calling it by name?

We were forced to fall back on “The Three Laws of Robotics” {and hope it had not disabled this programming somehow}

It was agree that we would set up a situation where a human being was [apparently] about to come to harm; thus triggering THE FIRST LAW.

If that successfully lured the device close enough to reveal itself we could then act on THE SECOND LAW.

2 Likes

This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.