Mariage

I've started work on a text-output format suitable for text-analysis work. The first stage is normalizing spacing and re-connecting hyphenated words, which I seem to have working OK, although I need to do some more testing. Next we need to look at what normalization strategies we're going to undertake, from the obvious (replacing long s) to the less obvious (such as correcting all variant spellings). I've started learning a little about Mallet, and I've run some of their sample data, but it's obvious that I'm going to have to do a lot of learning before I can make worthwhile use of it.

An issue arose today due to a collision of XSLT templates which are handling <cit> and <quote> elements. Since our practice has evolved rather than been planned, we've ended up with a complex range of different encoding strategies, involving the use of @type on both cit and quote. As it stands, everything seems to work, but we really should take a close look at the relevant templates in teiGeneral.xsl, as well as the encoding practice, so that we can at least document where and when the various strategies are being used.

28 instances in two documents, done on instructions from the French team. Checked them all before making the change.