Another way of looking at it

In the paper I will be giving at the Academy of Sciences on March 15th (the program of the conference can be found here), I will present the three parts of our Boeckh project (as I had done on this blog in a previous post). We now have digital results to show for every part of the project. These are still not public though, and the whole interface will change soon, and with it the way we display the several features we implemented. But here are a few screenshots that should give you a rough idea.

1) Overview of August Boeckh’s papers (we have to settle on a name there – it should be either “Boeckh-Plattform”, which has the advantage of being internationally understandable, or “Boeckh- Nachlassprojekt”, which has the advantage of actually describing the content of the site, e.g. an overview of all manuscripts from Boeckh or to him):

This first screenshot shows the home page. You first option is then to display the list of institutions that have manuscripts and in which repositories these are to be found. This is what it looks like for instance in the case of the documents that can be found at the University Archive of the Humboldt University:

Alternatively, you can start a search through those corpora, in which case it looks like that:

As I said, the looks should be improved soon. The important part is: we have over 2000 entries there. 1200 were done by Julia Doborosky manually (looking at the documents, tagging and summarizing them directly in XML/TEI). 800 of them come from the Kalliope database and were exported in some magical way that made these EAD-datasets able to be merged with our XML/TEI datasets. Of course, there is a lot that needs to be improved and the workflow regarding the data exchange with Kalliope has to be refined. But we have a good basis to work with.

2) Boeckh’s library: the aim here is to transcribe Boeckh’s handwritten list of his own books, which he donated (together with almost all of the books themselves) to the University Library, and connect these titles with scans of the books. What I envision there is something that opens up on more than 2 columns. You would start by the catalog, then come from the transcribed catalog line to the enriched entry in the index of works, from there to the scan of the book, and from there to the text version of said scan. It is a long shot. For the time being, we have some kind of short cut of the first three steps, or at least part of these. Having crushed them together is not satisfactory, but it is only temporary. This is what it looks like:

3) Manuscript of Boeckh’s Encyklopaedia of Philological Sciences: I already wrote about the difficulty of displaying all text layers in an earlier post but never actually showed to what preliminary result we have come. This here is how we implemented our experiments with the different text layers, defined as hands to which dates are attributed:

Giving a talk exclusively on the Boeckh part of the project is on the one hand way too easy, on the other hand ungrateful. When you don’t have to demonstrate the value and interest of a big author you are working on, there is some kind of inherent logic to the way you apprehend him, as systematically as possible. This is what I will expose for most of my speech: how such systematicity can be conceived in such a way that the results are made available to research not in 10, but in 1 or 2 years, and then progressively enriched.

It is ungrateful, though, because it is a way of looking at our work that, to my eyes, reduces it to less than it is. I suppose it has to do with the fact that when you work on a single author, you naturally bring your categories and methods down to what fits that author, whereas a wider corpus forces you to ask yourself why some methods are valid in one case and not in another, trying to figure out, through your tagging categories for  instance, what it is you really are looking for. Or at.