Monday, September 25, 2006

In a comment, TChem commented on the pi sweater and suggested an accessory instead. Good idea. Or pi socks, TChem. That would be another way to take it...

I'm almost done with the sweater. I got detoured because I started a hat (just the simple little mostly-knit-round hat from Sally Melville's Knit Stitch book) and I also went out and spent a couple hours de-junglifying my side yard.

I'm also thinking about two current news stories:

first, the spinach thing. If some enterprising biochemist could come up with a quick, accurate test for the hemolytic E. coli., one that could be applied at the processor or the farm, and then have spinach be certified "clean" from the field, that might put a lot of people's fears to rest. I know it would mine.

I bought some loose green leaf lettuce this weekend but it's not the same. I miss spinach. I wonder if we'll ever be able to eat raw spinach again or if it's gone the way of real Caesar dressing and mousse-with-raw-eggs-in-it. Or tasting cookie dough from the bowl as you're making it, for that matter.

Second, the suggesting that everyone 16 to 64 be AIDS tested: has anyone Bayes Theorem-ed this? What's the false positive rate? What's the false negative rate? One of the things I teach my biostats students is the surprising nature of Bayes Theorem. The example I used is for a test that has a 95% "capture" rate (i.e., a 5% false-negative rate) and a 3% false-positive rate. It turns out that with those numbers, only about one-quarter of the people actually infected who get the test will be successfully detected, if it is a condition with a low incidence in the general population (The example I used was a 1% incidence. I think HIV is considerably higher than that, but still: not all infected people will be detected by this test). Although I understand the social-engineering desire behind it (to remove the stigma from HIV tests), I'm concerned that it may give people a false sense of security - if there are infected people the test fails to identify - or it may lead to people's lives being disrupted by a false positive. (I could even see in extreme cases, someone committing suicide after getting a positive result - only to have it turn out to have been false).

And when they say "everyone," do they mean everyone? Cloistered nuns? People who have been married for 30 years and have never cheated and never had any partner other than their spouse? People who, for reasons other than religion, live somewhat like nuns?

And I don't trust the claim of "confidentiality." I've seen how the rumor mill works in small towns and even when you can't trace things to a certain nurse or doctor, you have your suspicions...

1 comment:

TChem said...

I've already charted for a hat and pulled the yarns from the stash. When I finish what of J's sweater I can do on the bus (today or tommorrow), I'm casting on.

Dorky hat here I come!