As some of you know, I’ve been working (with Paul’s invaluable help) on putting up the chunk of the Esposizioni that we’ve more or less completely encoded. This amounts to the Accessus, Canti I-III and most of Canto IV. Though this might not seem like much, let me assure you, it’s quite a bit of text, and the rest of the text is well on its way to being encoded as well. We probably won’t complete the encoding process until next fall, when Roberto may be able to join us again.

So. In order to put up the text in a useful way, that is, with at least some bells and whistles instead of what would look like plain text, we need to index it. Among other things, this means regularizing proper names. Those of you who did this with the Decameron will be feeling a slight chill and numbness as you read this; I’m thinking we need a better way of doing it.

Right now, I’m working on the index of people. This amounts to over 500 entries at last count. In order to make the index we need regularized names, a task that is complicated by the fact that sometimes Boccaccio refers to the wrong person, or refers to the right person but gives the wrong details about them, which makes their identity rather difficult to verify.

I’m far from an expert in historical work, but I’m working on coming up with regularized names for every entry, even if they are wrong. I’m doing this for several reasons. One is practical: I’ll soon be going away for a month and would like to put the thing up so that people can play with it. Another reason, though, is that this indexing business seems like a great opportunity to test out this online-collaboration thing we’ve been talking about. Here’s a detailed description of how this could work; it may sound definite, but I’m looking to y’all for opinions, suggestions and warnings of possible crash-and-burn.

We’ll have an index of names, coded in some way to distinguish between the names we’re pretty sure about (probably easiest would be with color – regular black for the ones we don’t need checked out, red for the ones we’re seeking help on). Each name will be a link to a page that will show you the context(s) for that name. From what I’ve seen so far, it would probably be enough to have the paragraph in which the name appears as well as the two paragraphs immediately preceding and following it.

I’ll set up a discussion forum, to which anyone will be able to post. We will advertise this everywhere we can: Kalamazoo, medievalist and Italianist and perhaps even history and comp. lit. mailing lists, individual emails to your friends who you think might be interested. Hopefully, there will be enough interest that people will post information about specific names on the forum, telling us the regularizations of these names, which of a set of people Boccaccio is likely to be talking about, and the source for this information.

The VHL team members will be the only ones who will change any of the information in the index itself. We’ll set up some sort of a checks-and-balances process, as appropriate.

The index will have only the names themselves (and any variants). Any additional information about the names will be put in as annotations to the text itself. If the annotation content has been research by someone who doesn’t have a participant-user account, we either make them a user (perhaps after a certain amount of input on their part, and with context-appropriate privileges) or one of us will post the annotation crediting the original author of the research and linking to their forum post.

How does this sound? What am I missing? Would you use such a system?