Since my last post, I have been continuing my explorations in digital techniques for scholarship. First of all, thanks to everyone who commented on that post, or the related question on academia.edu. Clearly there are some issues worth thinking about. Some comments have been very positive about the possibilities of annotating PDFs, others prefer ‘traditional’ transcription. This in itself is not surprising; as usual, a combination of techniques is probably going to be most appropriate
For myself, I still find the idea of annotating PDFs interesting. I acknowledge that it cannot be the only method: there are times when you just have to take notes. I still like to read books the oldfashioned way when I can, and the British Library – among other archives – does not allow photography. Even so, I have begun to try it, and though it doesn’t change my working method that much, and I am not entirely convinced it is that time-saving (especially with the added processing, about which more below), there is one main reason it appeals, which I only mentioned briefly before: it keeps you close to the sources.
As one friend described it in a recent conversation, notes add ‘layers’ between us and our evidence. Sometimes we even go through several layers between research and writing. For example, I often take details from my thesis for papers and so on, which more often means plucking from the thesis text rather than the original notes. I don’t agree with everything the late great Geoffrey Elton said in his book Return to Essentials – a transcript of lectures he gave in America, effectively an off-the-cuff rant about theory ruining history – but his evocation of the Renaissance humanist maxim Ad Fontes, ‘to the sources’, struck a chord which still resonates. Though I’m not sure what he would have thought of this computer jiggery-pokery (there are some interesting thoughts on Elton and his work in this blog post).
The problem I have faced is producing PDFs from my archive photos of a manageable size, because the freeware converter I am using, PDFill, doesn’t compress when it converts. My much more tech-savvie brother suggested I reduce the size of the original photos. After more online rummaging, I found Xnview, whose Xnconvert can turn pretty much any image file type into any other type, and batch-process. Using this, I have cut my photos down to the smallest readable pixel size and made them all greyscale. This took about 1-2 hours for a folder of 600-700 pictures, and then converting them to PDF is quite quick. The result is files somewhere between 60MB and 300MB depending on the number of photos. Still much bigger than a text doc, but much better than 2GB.
For a while I also toyed with the idea of not only annotating, but ‘coding’ (I didn’t really know what this meant until a few days ago). This requires CADQAS: Computer-Aided Qualitative Data Analysis Software – for a list see the wikipedia page, and the University of Surrey provides a good guide to different packages. Put simply, you load documents into this software and can then drop in ‘codes’ attached to a particular theme, name, place, or whatever, so that you can then call of these up at once. Many of the programmes allow you to do more complicated things, such as statistical or mapping analysis. I would use it to link all the different references across multiple documents relating to a single court case, person, ship, and so on.
The problem with these, from my point of view, is that they are mostly text-based, making them incompatible with my PDFs full of secretary hand. I tried Nvivo, downloading it through my university (which took a long time). You can ‘code’ to coordinates on a PDF, not just words, but you have to load documents into the software, resulting (for me) in unmanageably large files, and I found the interface counter-intuitive. An alternative is ATLAS.ti, where you don’t have to load documents into one file, but you can only code to pages, not to a point on the page, at least in the free trial version. If anybody feels like designing a programme combining Nvivo’s coding with ATLAS’s external use of documents, I would be very grateful…
So coding hasn’t worked for me, not yet, but I don’t think it was a complete waste of time. I have found out a lot that I didn’t know about software available for research. I have also started to think about how we might introduce these resources to students (or how they, if they are well-informed, could help me find out about these things: it is a bit ironic that I learned a lot of this by word-of-mouth). So it was worth taking note after all.