Investigations of a Dog

I haven’t mentioned Zotero for a long time. I was really excited when I first heard about it, and tentatively started using it last year, but then I accidentally wiped my Firefox profile and lost all the stuff I’d put in. It wasn’t much - mainly books from EEBO and notes for my posts on cavalry charges - but after that I got out of the habit. Now I need to manage the bibliographies for some articles I’m working so I’ve decided to start using Zotero properly. That involves importing over 1,000 records from my old database (which I wrote about here). I decided to use MODS XML as an intermediate format as Zotero can import and export MODS, and it’s also used as an intermediate format by bibutils. So far I’ve written a PHP script to pull records out of the MySQL database and display them as MODS XML. This bit went smoothly but while I was testing it I found what I think is a bug in the Zotero MODS translator (read all about it on the Zotero forum). Until that’s sorted out I can’t do the import unless I want to spend a lot of extra time manually changing creator types from author to editor.

I also need to think about adding new records. I’m trying to get on top of the debates about the causes and outbreak of the English Civil War, something which I’ve previously tried to avoid. Some of the literature is already in my old database from my PhD research but I need to find more. The most obvious place to look is the RHS Bibliography of British and Irish History as this is a more or less complete database of academic works with good search facilities (including subject headings and dates covered). Another potential advantage is the option to select records from the search results and display them as XML. The problem here is that they’ve chosen Adlib XML, which doesn’t seem to be very well supported outside the proprietary Adlib software. There isn’t a Zotero translator for it yet and I’m not really capable of writing one myself - if I couldn’t fix the bug in the MODS then it’s unlikely that I’d be able to adapt it to handle Adlib instead. What I might be able to do is write some XSLT to transform Adlib XML into MODS XML, which I can then import into Zotero. I’m not sure if it’s worth doing this. In practice most records in the RHS database are only a couple of clicks away from a record which Zotero can scrape. All records have a link to COPAC, which is fine for scraping books. Journal articles have a link to GetCopy, which usually leads to a record that can be scraped. Essay collections are a potential problem because the RHS has a separate record for each essay but there are no links to any other pages with these details as COPAC and other sites only list the volumes as a whole. So it’s a choice between entering these manually or getting to grips with XSLT (without the benefit of oXygen).

However I do it, I should have time to write about something interesting once I get it out of the way…