Hacker News new | comments | show | ask | jobs | submit login

This reminds me of the old story...

A factory is completely shut down, costing its owner $100,000 per hour. An expert is brought in to get it up and running again. He walks around examining many machines, then pulls a screw driver out of his pocket, and adjusts one screw on one machine. The entire factory instantly starts running again.

"Thank you! Thank you! Thank you!" the plant manager exclaims, "Give me your bill."

The expert presents him with a bill for $10,000.

The plant manager, surprised by such a large bill for so little effort, asked for an itemized bill.

The itemized bill:

Turning a screw: $1

Knowing which screw to turn: $9,999

1) The factory owner should have been happy with the bill, given that it was what six minutes of downtime cost him.

2) "I have found the problem, the fix will cost $10,000. Should I proceed?"

Number 2 is a sticky wicket, of course, because what if turning the one screw doesn't fix the problem?

It's extremely likely in the computer/PC world that your first inclination doesn't fix the problem. Sometimes rebooting DOES fix the problem, and sometimes it doesn't. What then? Another $10,000 for the next attempt?

Clearly, if he KNEW that turning the one screw would fix the problem, it's a different case altogether, but I'm suggesting that it's rarely that cut and dry.

Then the customer doesn't have to pay. If you don't know what the problem is and aren't sure how much it's going to cost, you bill by the hour.

I've never in all my years had this problem. There never was something I couldn't fix, but there were plenty of things that took more time than I initially thought. I usually tell the client "it seems that I won't be able to complete the task in the time of the original estimation, do you want me to stop now and you don't have to pay, or do you want me to try for an extra 1-2 hours and get it fixed?"

People always seem to pick the latter, and I haven't had any unresolved problems or unsatisfied customers yet....

then pulls a screw driver out of his pocket, and adjusts one screw on one machine

There is also a Russian version of this story, but it is about an expert sysadmin who knew exactly which point on the server box had to be hit with a hammer to fix an issue.

Definitely. That story (or rather, the concept - as there are seemingly many variations on the story) was the first thing that came to mind when I read the article. Google "knowing where to tap" for (some) variations.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact