I am writing a small python script which looks for the occurrences of an expression in a text file and then prints a list of the sentences it occurs in. The purpose is to have access to a wide variety of Latin sample sentences.
To that end, however, I need some input files. From the outset I ruled out the texts available at The Latin Library because these have too many typos for my liking.
Instead I decided to use the Perseus-editions. However, after having a closer look at some of these, for example their Titus Livius-edition based on the text edited by Weissenborn/Müller, I am not so sure anymore. Somehow it looks strange, and I have the feeling that the conversion of the scanned text has not been conscientious enough. Not too mention their edition of the Satyricon which has some weird stuff in it.
So I am now thinking about using the IntraText versions instead. They look clean. However, I am well aware, that just looking clean is not necessarily a hallmark of quality. Therefore I would like to hear what you think about these different e-text versions: The Latin Library, Perseus, Intratext.
Forget about Perseus, it has tons of typos too. And IntraText is just recycled editions of the Latin Library, I believe.
I suggest http://latin.packhum.org. These are the texts of the Packard Humanities Institute. I have yet to find a single typo! You’ll need to figure out a way to download the entire text, because the site only displays one page at a time.
The texts of the Packard Humanities Institute really look interesting. I can’t say, however, that I am overly fond of those using u instead of v. That may be more true to ancient usage, but it does not exactly improve legibility. It is not that easy to download entire texts as there is a javascript doing some weird stuff to the hyperlinks. On the other hand, the site-creators went out of their way to make it easy to manually copy a page. I wonder why. Hm…
What do you think about the Latin texts available at Gutenberg.org? These ought to have been properly proofread. There are only a few Latin texts, but some of them, for example their edition of Sallust’s De Bello Catilinario et Jugurthino , look fine.
The spelling (u/v) will vary from text to text, since the PHI texts are transcriptions of critical editions from several different editors. The same goes for other sources like Perseus and the Latin Library. Even for the same editor, I sometimes find some texts that use only u, others that use v.
As for copying the site, there’s no need if you get your hands on the database which I understand PHI will send you on a CD-ROM if you ask. There are also less legitimate ways of getting it.
As for Gutenberg, their English texts are usually very good quality, so I expect the same for their Latin texts, but I really don’t know.