Better late than never some impressions from the HDDA 2013 workshop. This workshop took place at Ghent University begin September and did not only include fantastic speakers (hdda_leaflet), but was also very well organised, including fantastic sandwiches for lunch and splendid sunshine (though the organisers might not be held responsible for the latter). It was only a bit unfortunate for a DH workshop that Wifi was not working in the lecture room, but again I am sure that the organisers didn’t have a hand in this and I can only speculate that the planners of the UFO (the modern housing of the history department and parts of UGent administration) meant it to be like that to prevent people from checking their mails in the lecture rooms.
Unfortunately I missed the first sessions due to some private coincidences, the rest of the morning lectures of these three days spanned from Bert van Raemdonck (Ghent) who lectured on editing letters in TEI to Caroline Macé (Leuven) who lectured on how to use digital tools to analyse and visualise the history of texts (stemmata).
A very relevant point was that TEI is only one of the available codes, but also the most widely used code, so if it fits your needs USE it. This will make your results shareable (and please do share your code !), easily mineable, and it offers also the advantage that more TEI-advanced scholars are mostly very willing to lend newbies a hand (for this one can for example join the TEI-list).
Another important point was that the scholars who use digital documents and tools have to be aware what they are doing and which implications it brings for your scientific work. Worst case scenarios mentioned here included scholars who measured properties of medieval manuscripts using digital facsimiles and not taking into account that measuring a picture will maybe not lead you to correct measurements. Very fascinating were also the lectures on computational topic recognition and computational authorship attribution. I have to admit that this sounded in the beginning like magic to me, but after a while hearing and reading more about the methods and tools I start understanding the underlying logic.
The afternoons were reserved to a hands-on training in TEI-conformant XML with David Birnbaum (Pittsburgh) using Oxygen. He focussed in contrast to earlier trainings in which I have participated more on actually encoding the body of the text than the TEI-header so that in the end all participants had some ideas how to actually practically encode a historical document in TEI, including abbreviations and variants in the transmission history. And yes, we sighed…because if you want encode all this, your TEI-code starts looking very unattractive, that will say, so chaotic that you almost don’t see anymore what you are doing. That is why the advice to first think very good about what you actually need, who your audience is, what kinds of uses you want to enable, is a very good advice indeed.
To sum up, to DO Digital Humanities means in many cases learning to handle code and tools that are rather unfamiliar to the traditional humanities scholar. It is a barrier one has to take and not being afraid to ask and make mistakes probably is an essential part of the process. To remind us in future of the fact that editing with TEI is a lot of work (so rather start sooner than later) David Birnbaum gave everybody a “Shut up and Code” button. I don’t regret having spent my birthday coding…it was fun! Thanks to the organisers (especially Tjamke Snijders and Els De Paermentier, UGent) and sponsors and I really hope that next year a follow up will take place. Why not have then hands-on experience with XSLT or Mallet?