OK, since no one’s jumping in, and I shouldn’t ask of anyone else what I would be unwilling to do, here goes…
The project that I’m working on is called the Magazine of Early American Datasets (MEAD). A big challenge in the digital humanities is preservation. In the study of early American history over the past four or five decades, many historians have used quantitive methods, and in doing so had to compile datasets. These are from a variety of sources, depending upon the historian and the project: census, probate records, tax lists, militia muster rolls, church journals, business ledgers, city directories, ship manifests, and so on. But in the humanities, unlike the sciences, the culture of scholarship is to publish the analysis but to keep one’s own notes and data. So many datasets are retiring or will die with their original compilers–and almost none are accessible to other scholars or the public.
Because the project hasn’t been officially announced, I don’t want to speak for anyone, but I’ll be working with a prominent historian of early American whose done a great deal of quantitative analysis, along with a center that sponsors early American research, to construct an online repository for these datasets and future datasets where they can be preserved, and publicly accessible. So far, signs are encouraging: the center’s host institution library is amenable to hosting the project on its electronic institutional repository. More anon.
What are other folks’ ideas?