Now we’ve been going with the Pama-Nyungan comparative database for about a year, it’s time to take stock with what we have, what’s worked, and what hasn’t.
- Searching: it’s easy to search for items but all manner of different things.
- … which means it’s reasonably easy to find and tag cognates.
- Data entry is also pretty straightforward. It may be a bit too straightforward given some of the consistency issues (see below).
- Doing reconstruction work is also quite straightforward. There are some unresolved issues, like how to code reconstructions to different levels, but I suspect there is no optimal solution.
- Importing data has also been pretty easy, when the data has been in any sort of vaguely marked up file. Even the completely flat-form Word dictionaries have been possible, although not particularly easy.
What hasn’t worked?
- Database sharing. We haven’t had to share it much, but it’s really a pain. We’re working on some alternative options for the New Year.
- Exporting the data out of Filemaker is also a bit of an issue at the moment. I was exporting it to excel, until the database got too big. MySQL may be our solution to this too.
- Consistency in data entry and importing has been a bit of a problem too. Most of it is ok, but there’s some messy stuff too, like the part of speech field. Keeping up with what lists have been phonemicised and things like that have also been a bit of a problem.