Borrowed Time

October 16th, 2009

Today I am still working. I have come home and I have done some things for myself but now I am back, connected by VPN to the corporate network and remotely accessing a PC hundreds of miles away. My boss has been asking and asking when I am going to get this fixed and if it works yet and if we should just scrap the whole project. I think mainly he wants to have something that works when I am there next week and we are talking to the important manager who is questioning the results the current late-shipment analysis is giving him.

Here’s how this works. First, it takes a long time to run. Yesterday when I ran it, it took over two hours. I changed some things so today it takes only about half an hour. When it does something wrong I do not usually know where along that half-hour cyle things went wrong. I have to investigate.

Investigating something requires sustaining and modifying a hypothesis. You pretty much have to do this in your head. If you could write everything down you would already have the problem fixed. But if you are formulating a possible explanation of what went wrong, you have to keep in mind why you think this might be the problem, how you are trying to test it, and, as you test ideas, what you have already shown to be false.

This requires concentration. In other words, you can’t do it when your boss is asking you to help out someone who is buried in work hundreds of miles away (in the other direction), and other people are asking you to guarantee doubtful things that it is your job to make certain (expedites), and people are calling you on the phone asking you to sound certain about doubtful things (intracompany orders). In other words, it is pretty hard to do any work at all on the late shipment analysis while I am at work.

Not that I haven’t tried; not that I haven’t made progress. But it is hard to make much progress when your re-test cycle takes that much time; it’s hard to jump right on it and try the next idea and hard to keep engaged long enough to try more than one or two ideas each day.

And on Monday I leave. So Friday, I work.