Zatera Ul

…skimming through the HARRY_READ_ME.txt; highly entertaining reading. It’s not a regular here’s-what-you-need-to-know-before-you-use-this-program kind of readme, it is more like one guy’s research notes covering a period of months. Really, it is a long saga of a guy trying to make something besides “garbage out” come out of “garbage in”.

To me, it was very reminiscent of my old research notes. If you’ve ever wondered what science graduate students do all day, often it goes like this: “Ok, I’m going to do X” and “What?! Why didn’t that work?” and “Oh, I see–fixed it!” and “Oh wait, what’s this?”; repeat a few thousand times and you get the picture.

But poor Harry had a far harder time of it than I did. Loads of incredibly crappy, inconsistent weather and climate data from all over the world. New crappy, inconsistent, and incomplete data coming in that he was supposed to incorporate into the dataset. Numerous crappy, inconsistent, buggy, and barely documented programs to process the data. Odd factors of 10 and 100 that kept coming in and going out. A scaling factor sometimes set at 0.5, and sometimes set at 2.5, for no discernible reason. A program that claimed it calculated one thing, but actually calculated another. New bugs from all the code he wrote trying to sort out the mess. Worse yet, he was expected to recreate some earlier analyses from several years back (impossible, under the circumstances). And all this under heavy time pressure. The text file ends abruptly and makes one hope that Harry finally gave up and started running out of the dungeon toward the light.

One reason that I read the whole thing is that I wanted to form my own opinion of Climategate, aside from the “Look: global warming science with its pants down!” and “No, no, it’s ok! Good science! Science is shiny!” memes. Overall, my impression of the research from this file was more of unsoundness than of real fraud. But still, very serious unsoundness. They’ve-been-publishing-papers-for-years-but-they-don’t-even-have-a-good-system-for-whipping-that-crappy-data-into-shape-yet-so-how-could-they-even-get-anything-that-was-worth-publishing! unsoundness. Like I said before, this is definitely a case of “garbage in”. Harry discovered that they never did anything consistently. My conclusion: this research definitely should not be used as part of the argument that we need to pledge several gazillions of our tax dollars, and about half of our remaining freedoms, to fight global warming.