Bloody computers

It's been eight weeks since I lost my job. Eight weeks is also the minimum interval between blood donations. I last gave blood the Saturday after I lost my job. So the clockwork of the calendar dictated that this weekend I was due for another donation.

Everything went fine, up to a point. This donation caught me slightly by surprise and I didn't spend the past week cramming for the iron test. Usually I'll have a steak or two in the days leading up to the donation, and maybe some spinach or raisins. I normally get my protein from fish, poultry, and eggs, not red meat, so sometimes my iron measures below the Red Cross minimum requirement. But even that went OK on the second-chance hematocrit test.

I'm not even sure what the problem was. Something required a routine override, an override that I was apparently given every time I donated blood in the past. Only this time the computer system decided not to offer that option, and instead automatically deferred me. Correcting this error in the system so I could give blood this weekend - the Red Cross needs every donation it can get - required a manual override which required the approval of the supervisor's supervisor, who needed to be contacted by phone. In the end the registration process which normally takes five minutes wound up taking nearly an hour.

I understand an increased reliance on computers for safety and double-checks. But there's a problem when the computer system overrides any reasonable human intervention. Back when I was doing Statistical Process Control I always focused on the human element in maintaining processes - "Every control chart asks a question," I would say, and it was up to a human to determine the question, ask it, interpret the answer, and then determine the next steps - or whether the right question had been asked in the first place. Today there is an increasing and regrettable reliance on the automation of these systems, where the user is essentially saying "computer, watch this for me, and let me know if anything goes wrong."

Stepping back a level, this reminds me of one of the principles of the Biointensive organic gardening method set forth by John Jeavons: double-digging the soil by hand. This is a huge task, but well worth the effort. Not only does digging the soil by hand cause a lot less damage to the soil structure than using a motorized tiller, but it also lets you see directly what's going on with the soil: how many worms there are, how much organic material is present in the soil, whether there is a dry "hardpan" layer, and so on.

Stepping back another level, this reminds me of a deconstructive analysis I did of the movie 2001: A Space Odyssey back in college...

But that's going back too far. My point is, because of a recalcitrant computer at the Red Cross Donor Center I spent an hour and a half on my blood donation today, and that threw my whole schedule off, and I never did get over to the house to scrape, sand, and paint today.

Maybe tomorrow.