Knower of things

Last year a team closely allied to my team developed a product using our software on a different hardware variant. It was a great success. they got it done really quickly. They ran into some issues along the way but kicked them out of the way pretty quickly.

There was one problem that I got called in on. I was able to help a bit with the diagnosis but I didn’t help fix it. They discovered a workaround almost immediately which they could use until the hardware fix was done a week later. The product was launched a few months later.

Time rolls forward until today when a fault report comes in. It sounds like a software thing so one of my guys goes off to help out. He comes back 30 minutes later with the offending hardware and a story that sounds familiar. It is the same problem I was asked about last year. I remembered the workaround. And guess what? It fixes it. That means it’s probably the same problem. A quick check-in with the workaround-finder of 12 months ago confirms my suspicions. All prototype hardware needs the workaround; we didn’t retrofit the hardware fix to the prototypes we had lying around. And that is how we got all confused today.

And I, in my keeper of knowledge role, remembered what happened last year.

If my synapses hadn’t fired just right confusion would have lasted until news got around and others synapses started triggering. That may have been a few days, or until we worked it all out again just like we did 12 months ago.

We have a habit of continuing to use prototype hardware. We tell ourselves we’ll remember this one is just a prototype and needs to be handled just a bit differently. We have a lot of hardware running automated and manual tests. With a 25 person team we have at least twice that number of units in our testing systems. Most of them are younger than myself. They are convinced they will remember what is different.


Instead of wasting mental energy remembering what is, and is not, different (or writing it down), we could keep everything the same; the same as our customers that is.

We waste time remembering ours isn’t quite the same as a customer’s because of reason X so we can’t reproduce fault Y that is getting reported by a customer.

You need to eat your own dog food, not its precursor.

Still inexperience thinks it will remember. It won’t forget.

Then 12 months rolls by. And the remembering falls to someone with a set of synapses that don’t always trigger right but do trigger on a hint of a vaguely remembered something.

Like what you read? Give Gordon J Milne a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.