I’ve just posted a writeup and some data on hand-corrected geolocation extraction over at the Uses of Scale site (associated with the Mellon grant Ted Underwood and I are running). The idea is to share as much as possible of the tediously achieved process stuff that’s required for computational research but that isn’t itself “achieved” results. In addition to my post on geography, these’s also information (from Ted and others) on OCR correction and on removing running headers from scanned texts. Not always sexy, but we hope it’ll help others do related work without having to start entirely from scratch. And I suppose we’re also selfishly hoping for feedback and improvements from anyone else who might have experience dealing with related issues.