The assorted finds of Artefact Publishing
As part of a demonstration I am building as part of a proposal for an interesting text project, I am revisiting Boethius’s De institutione musica as presented in Friedlein’s critical edition. When I worked on this text, I looked closely at the Alexander Turnbull Library manuscript, and quickly decided that I wouldn’t touch the Greek inscription there. The reason: despite having the reconstructed original at hand, I could not get more than a line through the manuscript text without getting hopelessly confused. Believe me, I tried multiple times, but to no avail: the scribe seemed to have no idea about Greek, was probably copying from a manuscript whose scribe had as little knowledge, and seemed to use a variant script of his own devising. Regardless, I was baffled, unable to match up one letter with another.
Now, years later, I am reading again the apparatus criticus in Friedlein for the inscription, and he has this to say:
Verba graeca pessime scripta leguntur in f, h, k, l, o; g tantum quattuor prima vocabula graeca habet, deinde interpretationem dat latinam, quae in f, h, k, l intra lineas repperitur. Ex ineptissimis scripturae mendis, quae maxime veri similia erant, restitui.
You’ll note the
pessime scripta and
ineptissimis scripturae mendis, and with no knowledge of Latin you can guess what he’s saying. And indeed he uses four other manuscripts (than the f, h, k, l, o and g mentioned above; the i manuscript is conveniently missing a folio at this point) to derive the lemma. So the scribe of the Alexander Turnbull Library manuscript is not alone in his failure (and you can see why by reading the result of some of my earlier work).
Posted by jamie at 17:50+12:00 | Comments (0) | Permalink
I have just released version 1.5.1 of IPA Zounds. This release fixes a couple of issues with the i18n support, by providing fallback translations for languages other than English and Dutch, and allowing for the GUI to find the translation files when called from outside its own directory.
Posted by jamie at 10:38+12:00 | Comments (0) | Permalink
It appears that Googlebot makes an unwarranted assumption when it determines what links it will crawl. The Dreaming Web site was, until recently, configured so that a request for / would result in the file start.html being served. In that file was a link to /index.html (associated with the file index.html), and Googlebot has never requested that file (other search bots have). I guess that Googlebot figures that index.html is simply the same resource as /, because the index.html is a common directory index page.
This assumption is probably handy for a lot of sites which use / and /index.html interchangeably, but it is of dubious merit: not only is it occasionally an incorrect assumption that the two URLs point to the same resource, but it hides the mistake of the website in using two different URLs for the same resource, breaking caching and the like. So, Google, fix your bot.
Posted by jamie at 16:01+12:00 | Comments (0) | Permalink
Three adolescents, sitting at the back of the bus, started examining the graffiti written on one of the seats. “What is this?” “It could be a backwards ‘r’ with a random line.” “Yeah.” “Or an ‘a’ with a random line.” “Yeah.” “Or a ‘p’ with a random line.” “*makes scoffing noise*” “Heck, it could be anything; he’s just made stuff up!”
Posted by jamie at 15:49+12:00 | Comments (2) | Permalink
I have just released IPA Zounds version 1.5. This release adds i18n support, and an incomplete Dutch translation, as well as a few fixes and improvements. This release is source only; no Windows binary has been provided (and probably won’t ever be).
The i18n support was really due to me wanting to play with this aspect of coding, rather than from any belief that it will prove useful to anyone (given that I have no evidence that the program itself has yet proved useful to anyone, such a belief would be misguided). I think I am doing it correctly, more or less, but it would be nice to know for sure. I had great difficulty finding any actual Python code using gettext, however, so I could be off base in some particulars (primarily around determining where to find the translation files; there isn’t much else the programmer has to do).
Posted by jamie at 19:50+12:00 | Comments (0) | Permalink