Apparently data science is not new. I admire the discipline. I can’t find anyone who has digitized his data set yet. I’m sure as soon as I write this somebody will beat me too it. I did find a study on the observations of weather at Erasmus High School from the early 1820’s. Perhaps one day I’ll get around to making this data set, or shall I say data setting good Ol’ Thomas’s hard work. I wonder which ontology would fit this best. Here’s an excerpt from the man himself courtesy of Founders Online(1)
“My method is to make two observations a day, the one as early as possible in the morning, the other from 3. to 4. aclock, because I have found 4. aclock the hottest and day light the coldest point of the 24. hours. I state them in an ivory pocket book in the following form, and copy them out once a week. The 1st. column is the day of the month. The 2d. the thermometer in the morning. The 4th do. in the evening. The 3d. the weather in the morning. The 5th do. in the afternoon. The 6th is for miscellanies, such as the appearance of birds, leafing and flowering of trees, frosts remarkeably late or early, Aurora borealis, &c. In the 3d. and 5th. columns, a. is after: c, cloudy: f, fair: h, hail: r, rain: s, snow. Thus c a r h s means, cloudy after rain, hail and snow. Whenever it has rained, hailed or snowed between two observations I note it thus, f a r (i.e. fair after rain) c a s (cloudy after snow &c.) otherwise the falling weather would escape notation. I distinguish weather into fair or cloudy, according as the sky is more or less than half covered with clouds.”
1. Jefferson to Thomas Mann Randolph, Jr., April 18, 1790, in PTJ, 16:351-52. Transcription available at Founders Online.