A month ago I posted about Liver Pudding, a Carolina delicacy whose very name makes me shudder. Last week, Ann Althouse reminded us that her readers once paid her $200 to eat something she has always despised, an egg salad sandwich. Putting these two facts together, it occurred to me that I might be able to raise some much-needed funds, find something to write about, and gratify my more sadistic readers, all at the same time. I therefore undertake the following promises:

1. For a PayPal contribution of only $20, I will buy and eat an entire one-pound package of Neese’s Liver Pudding, finishing it off in no more than a week, will post pictures of the stuff in the package, in the frying pan, and on the plate, and will write about its taste, texture, and any other characteristics worth noting on both of my weblogs.
2. For another $20 each, I will do the same for Neese’s C Loaf (main ingredient: pork stomachs) and Scrapple. (I haven’t seen Liver Mush or any of the other more exotic Neese’s products in the grocery stores I frequent, but am willing to look for them, if anyone wants me to try them.) For no further charge — because I’m sure I’ll like it — I will also buy a package of Neese’s Country Sausage, to see how it compares, if readers answer the first two challenges.
3. For $100, I will make and review a haggis. My local Asian grocery store had two of the three main ingredients last time I was there (lamb stomach and liver) and I should be able to get a lamb heart there or somewhere in town — perhaps at a Halal butcher. (The recipes call for sheep, not lamb, but surely either will do?)
4. Finally, for $200 I will purchase a package of ‘pork uteri’ at the Asian grocery, cook them using an authentic ancient Roman recipe from Apicius, eat them, no matter how revolting they turn out, and provide pictures and a review, as before. Sows’ wombs were a Roman delicacy, as admired as lobster or Porterhouse steak today, but I’ve never tried them and am torn between intellectual curiosity and visceral disgust.

Now I need to look through Apicius and see if there are any more Roman delicacies to add to my fund-raising challenge. The Asian grocery carries goat penises and ‘intestinal bung’, among other things, but I don’t recall anything like either of those in Apicius. And I doubt that I can get hold of flamingo or dormouse in Raleigh.

Although I haven’t posted much lately, I have been hard at work ‘behind the scenes’ on several projects. Here are the first two:

I. I continue to add to my collection of critical texts on the web, and have just uploaded Ovid, Heroides 1 (Penelope Vlixi), with a brief apparatus and two of my own conjectures, one of them rather dubious. The rest of the letters will follow, not necessarily in numerical order, though I do plan to finish the single Heroides, and Martial’s epigrams, before turning to the doubles.

II. More important, I have begun the process of transforming my web-texts, including many more not yet ready for publication, from hard-coded HTML files into a MySQL database with a PHP interface. Overall, they will look exactly the same: here is the HTML text of Propertius 2.29, and here is the MySQL version. However, the new format offers several advantages:

1. When the user interface is ready, readers will be allowed to select whichever section of the text they wish to read, instead of being given a whole book and then scrolling down. Here is the MySQL version of lines 11-20 of the same poem. Until the interface is ready, and I add more poems, readers can alter the line numbers in the URL and see what happens. The only other poem available now is Propertius 4.10 (here), which illustrates how I handle transpositions (not that I’m entirely convinced by that one).
2. Once the user interface is written, readers will be able to compare two or more passages from the same or different authors, with or without apparatus.
3. At the moment, the apparatus criticus for a given line consists of a single character string. These will be broken up into their constituent parts and stored as variants, conjectures, lacunae, transpositions, repunctuations, and so on. Once that is done, and once the database contains a sufficient number of works, interesting searches will be possible. Readers will be able to locate all of Bentley’s conjectures on the Heroides, or see how often iam has been confused with nam in manuscripts of various authors, or (if they’re feeling malicious) see who is most futile emender of a given author, i.e. the one with the most conjectures published without any of them making it into the text.
4. A more sophisticated interface will allow interested readers to modify their own electronic copies of my Latin texts, promoting variants and conjectures from apparatus to text, adding or subtracting variants and conjectures, transposing lines to see how they look and then transposing them back, and so on: like penciled marginalia, but far neater. If I were doing this just for Propertius, I’d call it ‘Quot Lectores, Tot Propertii’, but I suppose I need a more general name for the project.
5. When the editorial interface is ready, it should speed up the process of editing and publishing texts. That will allow me to finish up the various texts I have under construction more quickly, though making up my mind what to print is already the greater part of the labor.
6. In the long run, I plan to license my texts and the editorial interface, which will allow others to edit and publish their own versions of my texts and of any others they care to upload. Of course, anyone who takes a text (mine or another’s), edits it, and publishes it on another website will need to be careful to avoid copyright infringement, making sure that the new text is not too similar to the source text or to any other text still under copyright. Comparing two web-texts will be easy enough: software can easily be written to calculate the number of significant differences between any two of them. (Determining how different is different enough to avoid copyright infringement may be tricky, though. In fact, the necessary degree of difference would differ from author to author: independently-edited texts of the Aeneid might well have whole pages identical in text and apparatus, while independently-edited texts of Propertius might well differ in every couplet.)
7. In the even longer run, text and apparatus need to be plugged into a larger system that interlinks repertories of conjectures, bibliographies with (when possible) links to the full texts of the articles listed, lexicons, commentaries, metrical analyzers, and so on. Combining multiple texts and apparatus critici into a single database is just the first step.

To sum up, this is the obvious next step in the computerization of classical texts: bare web texts, and even web-texts with apparatus criticus (not that anyone else seems to be particularly eager to edit those) are not very different from printed texts. Making the text and (even more important) the apparatus into a searchable database will provide enormous advantages to teachers, students, researchers, and even ordinary readers.

Though I’ve been thinking of something like this for years, I began putting together the PHP and MySQL the day after Christmas, and have put 45 hours into it so far. A great deal of work remains, but it is already clear that the job is doable, and worth doing. The SQL is turning out to be easier than I expected, considering it’s been nine years since I’ve done any, and twenty since I did SQL for more than a few months. On the other hand, the PHP is trickier than I thought it would be. However, I’ll be at the public library when it opens at 9a.m. tomorrow, and they have several books with both PHP and MySQL in the title, and they should help.

Update: (1/28/08, 8:45pm)

Apologies to anyone who tried the links and found that they didn’t work. I saved a copy of the program in a different directory for public use, but forgot to make a copy of the database. Once I started adding and deleting columns in the tables to fit modifications in the (non-public) editing version of the program, the public version naturally stopped working. I’ll try to fix that problem in the next day or two, but it should work until I modify the database again.

One more point: If the results of the comparison between HTML and SQL texts are unimpressive, that’s the point: SQL can reproduce the HTML texts — not that I’ve tried to make the formats precisely the same — though it can also do much more.

Over the last month or so, I have added quite a few Classics titles to the list of Books for Sale (link in left-hand column), including several that are, so far as I can determine, not available elsewhere either new or used. The most notable addition is a complete set of Momigliano’s Contributi alla Storia degli Studi Classici e del Mondo Antico I-IX, complete in six single and three double volumes. Please buy the books I don’t need, so I can buy the ones I do need.

If we’re going to give our students texts in which v is used for consonantal u, shouldn’t the enclitic conjunction be -qve, not -que? The latter confuses some students in their scansion exercises, since they try to take the u as a vowel. For that matter, shouldn’t desueta in Aeneid 2.509 be desveta? It’s scanned as a trisyllable, and ue is not a diphthong, so the u must be a consonant here. For the opposite problem, I assume that Horace’s trisyllabic forests in Epode 13.2 should be spelled siluae, not silvae, even in elementary texts, to show that the word is an anapest, not a spondee. The Latin Library does precisely that. I should check my other Horaces.