This week found me in Manila, teaching a class for IBM business partners and customers. Though I am no longer officially connected with the company, the opportunity for some odd jobs sometimes presents itself and who am I to turn away such a generous gift?
The class was a combination of lectures and laboratory exercises. Having taught the class before, I was fairly well-versed with the procedures. Indeed, things went well until the last day. Then we hit a snag that had me scratching my head.
I won't bore you with the technical details. Suffice to say, I had two computers hooked up the way the laboratory manuals said they should be hooked up, the way I had done countless other times. It just so happened that the connection didn't work the way I thought it should.
After several tries, I took the only reasonable course of action. I disconnected the servers from each other, replaced the cables, and reset the machines. With great deliberation, I reconnected the computers, testing them at every step. Ultimately, I arrived at the same set of connections that I started out with. Only this time, it worked.
I did not bother to hide my puzzlement from my students. There was nothing substantially different in how we set up the machines. We guessed at several possibilities: that the cables might have been loose, that the software didn't start up correctly, that there were conflicting addresses, and so on and so forth. Being seasoned professionals themselves, they did not come down too hard on me.
In fact, the occasion prompted us to share war stories with each other during the afternoon break. One fellow recalled that they had a problem with one of their computers. All the engineers in their company attempted a fix, and all invariably failed. Just when they were about to give up in frustration, one fellow volunteered one more try after the weekend. "Make sure you go to Mass," they jokingly told him; he did, and come Monday, after he gave it another go, the machine just worked.
My own story was a little extreme. Some years back, we had a very problematic machine with a customer. It would go down every other day, and this went on for a couple of weeks. All our attempts led to dead ends. Finally, at the end of my rope, I decided to light a candle at Church after Mass. And somehow, the situation seemed to resolve itself.
When I started my career in IT, I assumed that it would be a logical and scientific work environment that I would be entering. Cause led to effect: you flicked a switch and something came on; you changed a line of code and a program behaved differently; and all this was supposed to work consistently. That was the theory anyway. But in practice, IT systems nowadays are so complex that you never really know that one change you make could lead to an invisible chain reaction.
So really, you have no choice: you plug it, and then you pray it works.