Apart from the UI around the word parsing pop-up being kind of buggy (it doesn't close readily, the scroll position seems to get saved differently depending on the word), I'm pretty sure there's an analytical error in that sentence.
The word "nostra" is glossed as Nom Plur Neut; that is, the possessive adjective "noster", "ours", in its nominative plural neuter form. "Nostra", with a short final -a, is indeed the nominative plural neuter form of that possessive adjective. However in that sentence, the word is being used as an ablative singular feminine, with a long final -â (which is unmarked because the text as they present it doesn't seem to mark vowel length at all).
archive.org has this (https://archive.org/details/caesarscommentar07caes/page/n11/...) 1918 out-of-copyright literal word-by-word translation of the text, which does mark vowel length, and you can see that the translator words the last clause as "the third, (those) who in (the) language of themselves are called Celtae, in ours, Gauls". nostrâ is modifying an implied linguâ there, which is a singular feminine noun in the ablative case, so nostrâ has to inflect to agree with it - nostrâ [linguâ] translates to "in ours [our language]".
My Latin is self-taught and mediocre at best, but even I was able to spot that error. And The Gallic War is a very famous text that is routinely taught fairly early on in traditional Latin education - a lot of students of Latin over the centuries have read that exact line of Caesar's. The fact that the LatinCy neural language model for Latin, which the site claims performed the analysis, screwed this up doesn't leave me very confident about the quality of the grammatical analysis in more difficult corners of Latin grammar.
I think most Latin text out there won't use diacritics at all. I only studied it for 3 years back in high school leading to the AP exam and I don't think I was ever taught about them or saw them once. This video https://www.youtube.com/watch?v=D3bmLi1bKI0 and wiki page https://en.wikipedia.org/wiki/Apex_%28diacritic%29 were interesting. I'd bet part of the reason is just a side-effect of teaching Latin as a dead language where you never speak it / pronunciation doesn't matter. I remember very little of the language, I believe also in part because of that.
I built something similar to this by cloning the Diogenes repo and getting Claude to re-implement it in Python (it’s a very old battle tested Perl code base, so a great reference implementation) and using the TLG database for the Greek and Latin texts. You can take this even further by integrating it with the Barrington Atlas (there are scans on Anna’s Archive) for looking up ancient place names, so you have dictionary + map lookups. If you’ve read any of the Landmark series books you’d known what I mean.
Also better if you can generate chapter by chapter critical apparatus on difficult grammar and Anki decks. It’s an annoying part of learning these languages to have to stop and look up words, I like spending a few days learning vocab before reading and it makes it so much more pleasant than having to stop and look things up the whole time.
Obviously all this stuff is copyright so it can never be shared, but I don’t care it’s for my own personal use. I also bought like 6 hours worth of Ionnis Strattakis’ recordings where he reads a bunch of Ancient Greek in reconstructed Attic pronunciation and fine-tuned a text to speech model (styletts2) with full accent markings and breathings. It’s extremely natural sounding. My long term goal is to have a personal tutor that I can speak Attic to and basically have lessons with everyday (speech to text -> llm-> text to speech). All the pieces are there to actually do this.
LLMs have been an absolute game changer for me who is a hobby classicist but also knows how to build software. They are so amazingly good at Attic Greek and Latin, it’s like having the best teachers in the word at your fingertips for some really niche topics that it wouldn’t be possible with otherwise. Also extremely good at managing, building, cleaning up, deduplicating Anki decks.
Definitely! I myself have been having a lot of fun[0] recently using LLMs to enrich one of my favorite beginner Latin readers[1] with audio forced-alignment, POS tagging, morphology, definitions, and other niceties that learners might appreciate.
[0]: https://hercules.hookbangsplat.com
[1]: https://archive.org/details/p1fablesoforbili00godl/mode/2up
Also most of the word popups use different spellings, like "iam" becomes "jam".
For example:
https://en.wikipedia.org/wiki/File:Trajan_inscription_duoton... (Latin as typically carved in stone)
https://commons.wikimedia.org/wiki/File:Herculanean_Rolls_-_... (Greek as typically written on papyrus)
(Source: not an expert in any way)
My undergrad was in classics and viticulture, then did grad school in atmospheric science, which led me to tech and hackernews.
How did you get here?
At one point I was seriously considering attempting an academic linguist career. Fortunately, I was also interested in computers; and as I often tell people who ask me about it, computer programming pays much better (and I'm glad I didn't pursue a humanities academic path for other reasons). Still, it's an interest I try to keep up with when I can.
I was learning to program at the same time, following along A K Dewdney’s Computer Recreations column. I went into college to learn pre-med but always found myself more interested in the humanities, like my survey of Greek literature, which was supposed to accelerate my prerequisites. Also hung out with friends in the engineering labs on Sun Sparcstations playing netrek and MUDs, which did for me.
Anyway, I pivoted and started taking computer courses, like artificial intelligence, reading from Russell and Norvig, Clocksin and Mellisa, and Paul Graham.
I learned Greek while consulting for companies across the country, programming network performance tests, to have something to do in the hotels.
Fast forward to 2008, 2009, and I see that the Lispy Hackers and Painters guy had a new website. I started an account but lost it. Or did they reset it after disuse? Anyway, that’s me.
Oh, I did take Latin in Catholic high school, but did not really use any of it for reading, just vocabulary, but it did help with the old-style grammar-first Greek texts I learned from.
I have many interests but botany became a big one, and the scientific naming scheme is mostly Latin and Greek, which leads to some interesting word history. "Sativa" means anything "cultivated" (example, "Avena sativa" is "cultivated oats" and "Crocus sativa" is "cultivated saffron"). "Nemo" means "no one" (so the "Search for Nemo" is the search for no one, haha), the root word for "service" means "slave". There are many more interesting associations.
I find the deepest and most awe-inspiring truths within physics (out of which biology and chemistry are arguably sort of abstractions in terms of order of complexity). Behind this is a long awkward history of curious people fumbling for answers in the dark. I am an odd nut, but it is connecting in a comforting way the way I can see others have asked the same burning questions of the "how" and the "why" of things, and the myriad subjects it has lead them to.
As John Muir once said: "When we try to pick out anything by itself, we find it hitched to everything else in the universe."
[0]: https://en.wikipedia.org/wiki/Lingua_Latina_per_se_illustrat...
A lot of role playing games, especially from Square Enix,leverage a lot of Latin and classical education.
The thing that strikes me from ancient texts is not the obvious differences from our day, but the similarities.
It also provides one-click definitions and grammar analysis. The main difference from this project is that it remembers your vocabulary across texts. You can mark words as new, learning, or known. It then recommends texts based on the vocabulary you already know.
It also includes flashcard review, sentence explanations, text import, and an Old English library.
For tbos version, I gather that a bilingual presentation would be more than necessary: keep Ancient Greek text on a side and display a scholar translation into a selection of switcheable languages
The way the Greek is displayed isn't very helpful, though. Specifically, any vowel with a grave accent displays the accent as a separate letter, which makes it very distracting to read. I don't think it's a problem with the encoding, as I can copy the inline text and it displays just fine - though with some weird spaces before commas or periods:
ΕΝ ΑΡΧΗ ἦν ὁ λόγος , καὶ ὁ λόγος ἦν πρὸς τὸν θεόν , καὶ θεὸς ἦν ὁ λόγος .
A propos of breathing marks, iota subscripts, and the three different accent marks of Classical Greek, when I learned it at school we had to remember the breathing marks and iota subscripts, and would lose marks if we omitted them, but we didn't need to learn the accents. Modern Greek now has only (acute) accents, which you need to know to stress the correct vowels, exactly where the accents were in the equivalent Classical Greek words.
What should actually be done, and what OP should do, is take the Perseus website and make it so it doesn't 503 all the time (it is incredibly unreliable).
Then, FOIA the State of California for the pay-to-play data which the UC Irvine-based TLG hoarders are withholding from the public (it is a publicly funded project...) so that the corpus can be meaningfully extended and built upon.
The 1990's tier html vibe of Perseus is to its great advantage.
Ancient Library is some vibecoded alternative that the LLM translation it uses will be incorrect. Color me unimpressed.
I don't get a definition for 'fulgere', third word first entence, just a reference to 'fulgo'. I can guess what it means though from the more common 'fulgur', maybe something adjacent to flashing or lightning, but translations seem a bit sketchy.
It seems some books are missing chapter markings. Revelation, for instance. The verses are numbered, and you can notice when chapters change, but one can be lost when searching for a specific chapter.
Dictionary entries are cool, but I would want the in context meaning at least highlighted, so I don't have to read the full entry and do that myself.
Overall, nice idea but it seems like a very barebones implementation that needs a tremendous amount of polishing to be useful.
From the point of view of a learner it's unfortunate that macrons (diacritical marks for long vowels) have not been added to the text, and that 'u's have not been altered to 'v's where appropriate.
The word-by-word treatment of translation makes this broadly a part of the category of interlinear texts. For comparison here's one of the early pages of Max Müller's 1864 Sanskrit-to-English interlinear of the https://en.wikipedia.org/wiki/Hitopadesha : https://archive.org/details/firstbookofhitop00ml/page/2/mode... . This is part of a revival of interlinear texts, as a resource for language learners, which traces back to John Locke but really caught on in the English-speaking world in the early nineteenth century thanks to James Hamilton. It's unusual for "traditional" interlinears in having one row which gives a(n explicit) morphological gloss of each word in the original (the "da, 3 sg. Pres. Par" and so on) in the same column as the original word, an English translation of the word, and (in this case) a transliteration. But morphological glosses are a standard feature of modern "interlinear morphemic glosses" https://www.christianlehmann.eu/ling/ling_meth/ling_descript... made by and for linguists who want to analyse and compare languages, rather than learn them. Linguists' adoption of modern IMGs seems to have taken off in the 1960s (though some use has been made of interlinears for "serious" linguistic purposes since at least the 1890s beginnings of the Linguistic Survey of India https://en.wikipedia.org/wiki/Linguistic_Survey_of_India ).
https://theamericanscholar.org/the-new-old-way-of-learning-l... is a very incomplete history of interlinears (the French are especially shortchanged) but probably still the best thing out there in English. See also https://www.reddit.com/r/interlinear . (I have to release some things about the history of interlinears myself but I'm years late at this point.)
https://shop.hyplern.com (might as well add an affiliate link! https://invi.tt/N5VZXP7g ) and https://interlinearbooks.com/ are two publishers offering modern interlinears aimed at language learners. The polished but expensive Legentibus service https://legentibus.com/ offers some interlinears too.
Guess who Vincent F. Hopper https://archive.org/details/chaucerscanterbu0000chau_k0e4 was for a time married to!
