… and so it is with space.

Stop Design talks about the ripple effect:

In grade school, many of us would have loved to see a “95%” scribbled across the top of our test papers or filled into a report card. A number that high could produce a euphoric sensation, confirming just how “good” we were. Better yet, it might have meant a long-awaited trip to Chuck E. Cheese, where numbers that high would produce the maximum amount of grade-based reward tokens.

But in grown-up society, 95% is not good enough. Getting grouped into the 95th percentile is not good enough to open the doors of some of the best schools, even for an ambitious student. 95% percent correct means nothing in the business world, where a 5% error in reporting can land a business in court, or send it into bankruptcy. IT professionals would scoff at a server that boasted 95% uptime. When a doctor tells us there’s a 95% success rate for this procedure, we immediately begin to worry about the remaining 5% of cases. A 5% delta in course could result in a New York-bound passenger jet landing in Miami.

So we’ve been conditioned to think 95% good is unacceptable. And we seek out the remaining 5%, drawing as much attention to it as we can.

Same with NASA, which has a 98% success rate with STS.

Of course, that 2% failure rate means that over a dozen lives have been lost on Challenger and Columbia.

The original flaws in any product are continually magnified by a small portion of its most vocal critics. The criticism ripples outward, causing more people to join the fray, until it seems the entire product is flawed. This deception might disappear if only the critics would be silent long enough to let the ripples die out. Is it really that flawed?

We live in an imperfect world, and we produce imperfect things. Every thing and every person can use improvement somehow, somewhere. But let us not forget so many of the things which are good, lest we also forget to thank those responsible for the good.

Amen.

4 comments

  1. It seems to be even worse in the world of embedded systems (what I do for a living). In my specific situation, I have a processor running at 66 MHz. That normally doesn’t seem that fast in the world of multi-GHz PCs, but just think about it for a second. 66 MILLION instrictions can be executed in that amount of time. If just one is flawed, the whole system can come tumbling down. In my field, this doesn’t cost lives, but it can take down communications a cell-phone tower, peoples DSL lines, or just basic telephone service. People don’t respond well to a loss of service like that. Oh yeah, we don’t get to run for just 1 second either. The code I write is expected to run error free for years at a time. Could you ask your windows machine to run for even months at a time?

    Just remember this the next time a piece of software crashes. We’re talking millions of lines of code out there where 1 or 2 lines can bring the whole thing crashing to the ground.

  2. I can get a few weeks out of my 2K box at work, but my laptop gets rebooted much more often. I think a good bit of that has to do with me hibernating it rather than doing more full shutdowns. Oh well.

Comments are closed.