We’ve been asked to consider if we would use text mining as we went over it today. I could see creating a corpus of documents written by the subject of my research (Augustin Smith Clayton 1783-1839) and then attempting variations of distant reading to see what topics might pop up that I hadn’t thought to include. I possess pretty much everything the guy ever wrote, so it is possible.There is the problem of preparing the hundreds of pages of his awful novel, The Mysterious Stranger (even people at the time thought it boring), as well as his many political essays, reports and speeches, not to mention all the notices of him in the newspapers, and the ghostwriting he did for Davy Crockett. All of that is done in the 1820s and 1830s. None has been well OCR’ed, which means I have to do it and correct it, just to see if text mining produces anything useful. My take is that while possibly interesting, the return on the investment of time and effort, for this project, would not be sufficient to justify all the work involved.