7 ms·

After finishing DuoLingo’s Latin course, I wanted to read some Latin texts, but I didn't find easy enough texts and going back and forth to a dictionary was cum

After finishing DuoLingo’s Latin course, I wanted to read some Latin texts, but I didn't find easy enough texts and going back and forth to a dictionary was cumbersome. So I created this app for reading basic Latin texts. The idea of the app is to have a Latin text with translation of each word under the paragraph line, which makes it easy to grasp the meaning but also focuses on reading the original Latin text. It only has one book, if I finish it I might add others. I used OpenAI to do the translation which looks pretty good for me, with the caveat that... well... I do not know Latin. This approach will not probably work for more complex texts. This mode of reading works for me, not sure if is of interest to anyone else. The app source is on GitHub if you are interested: https://github.com/aleris/duplex-lectio https://github.com/aleris/duplex-lectio A couple of details about the dev process: https://adi.earth/posts/duplex-lectio-read-latin-bilingual/ https://adi.earth/posts/duplex-lectio-read-latin-bilingual/

mentalgear 2y ago
Thank you for the details: I really like the minimalist UI, well done! And I like to see you sharing the tools that you build to help others learning. That being said, I'm surprised though, that the translations are GPT generated, so not sure how trustworthy this actually is. Domain foreign users have to be able to trust that learning resource are proof-read / accurate. Not to say it's worthless, but you may want to note that the translations were done automatically and may contain errors.
jrflowers 2y ago
>I used OpenAI to do the translation which looks pretty good for me, with the caveat that... well... I do not know Latin OpenAI also does not know Latin. This is either a tool to troll people that can read Latin or a tool to help people “learn” a made up vaguely Latin-shaped set of gibberish that ChatGPT nondeterministically generated. This only works for a definition of “Latin” that is a sort of vibe wholly detached from structure, syntax, or vocabulary.
- rahimnathwani 2y ago
  vaguely Latin-shaped set of gibberish that ChatGPT nondeterministically generated OP hadn't used AI to generate Latin, but to generate English.
  - jrflowers 2y ago
    It doesn’t matter which set of text you generate with ChatGPT in this case. Using it in either makes the output useless as a tool to learn anything about both sets of text. This issue is compounded further when, as the OP readily admits, there is no quality check involved (OP does not know Latin)
    - rahimnathwani 2y ago
      Using it in either makes the output useless as a tool to learn anything about both sets of text OP is using it as a tool to improve their comprehension of Latin. The tool shows a word by word translation, which is jarring to read linearly, but works well for filling in gaps. It's far from useless as an aid to comprehend the Latin text. It's been over thirty years since I last studied Latin, but I still remember enough to be able to tell that this tool would be useful for a learner, even without perfect accuracy.
- empath75 2y ago
  You're severely overstating the case here, as if it just produces random text like a markov generator or something.
- dmd 2y ago
  I do know Latin (quite well), and gpt-4o does an extremely good job at translating from English to Latin and vice versa.
  - nescioquid 2y ago
    Note how one opinion provides the warrant for the other.
satvikpendem 2y ago
Duolingo really isn't good for learning languages. When I learned Latin, I used the immersive method of just starting to read simple texts to more complex ones. The principal book of this form is called Lingua Latina Per Se Illustrata, I have the color PDF if anyone so needs it. It's a book that starts off with very simple sentences and gradually introduces more complex topics like tenses and declensions as the story goes along.
- thaumasiotes 2y ago
  > It's a book that starts off with very simple sentences and gradually introduces more complex topics like tenses and declensions as the story goes along. I'm not sure this is actually an approach I'd recommend. I was recently asked to give some supplemental English tutoring to a Chinese brother and sister, 9 and 5 years old. The 5-year-old could already use and understand 'simple' sentences such as "what do you see?" and "where is your brother?", though I'll note that the subject-auxiliary inversion required by a question of that form isn't exactly a simple concept. I got them a copy of The Cat in the Hat, and their mother objected that it was too advanced for either of them, because most of the verbs in The Cat in the Hat are in the past tense, which apparently isn't covered within the first four years of Chinese English instruction. You can't learn what you're not exposed to, but you can learn a lot of what you are exposed to in a language.
  - satvikpendem 2y ago
    Lingua Latina is for adults who already have some base level knowledge of tenses in their own language, preferably a Romance or Germanic language (as I believe some languages don't have tenses), not for children who have no concept of them. Once you start reading the book, it really does start to make sense while teaching you the various forms. It's on Internet Archive if anyone wants to read it: https://archive.org/details/lingva-latina-per-se-ilustrata-pars-i https://archive.org/details/lingva-latina-per-se-ilustrata-p...
    - thaumasiotes 2y ago
      > preferably a Romance or Germanic language (as I believe some languages don't have tenses) A couple of points: - If you natively speak a Romance language, learning Latin by example is going to be really easy for you. This doesn't belong in a comparison with anything else. - Germanic speakers have no special advantage over any other Indo-European speakers. - You might be interested to know that while Mandarin verbs don't inflect for tense, the negative particle does, so you have to observe a distinction between past and present tense whenever you're negating a verb. 我不理解 I don't understand 我没有理解 I didn't understand
Timwi 2y ago
I love this mode of reading very much. I think this could be great for any language, not just Latin, and any target language, not just English. The only difficulty is sourcing good texts that are both fun to read and also suitable for language learning at various levels of skill.
- thaumasiotes 2y ago
  I've been working on a tool to display the structure (and glosses) of a text. All the structural and definitional annotation is provided by me based on personal knowledge. So far I have most of a single short fable in Mandarin Chinese done. ;D Does this sound like something you'd be interested in checking out?
eyko 2y ago
I do not speak latin, although I studied it for two years in high school and I'm a native speaker of a romance language, so my understanding of latin is pretty much basic to guesswork. This is a really cool tool -- I often read latin texts with the original on one page and the translation on the other, just because I think it's interesting to see how they wrote/spoke at the time, but for the most part certain words or declinations throw me off guard. Inline literal translations really help there. That being said, I noticed whilst reading some of the texts that the inline literal translations are still in latin, e.g. in "Part IV. I Some Barbarous Customs", most of the translated text is just latin. I guess OpenAI won't take all our jobs just yet! I do have one suggestion for improvement though. Many of these texts have translations that are already in the public domain (older translations). It would be helpful to display the original Latin and a fluent English translation side by side, whilst still being able to toggle the literal translation on or off. This setup would make it easier to compare the original text with a fluent English translation, similar to the format used in some bilingual books.
philsnow 2y ago
“Translation of each word” is also called a “gloss”, and I think it’s absolutely vital for trying to read works in translation. Lots of words and phrases have multiple meanings and connotations in their origin language and it’s not usually possible for a translation to bring the richness into the target language. (I’m going to butcher this because I don’t have the text in front of me, but) Thomas Aquinas composed several hymns for the feast of Corpus Christi, one of which is “O salutaris hostia”, which contains a reference to “fer auxilium” which is often translated to “bring help”. The choice of the word “fer” isn’t the most obvious choice for “bring”, though. Some translators have speculated that Aquinas chose “fer” because of how close it is to “ferculum”, which is a litter or wooden frame upon which spoils are carried, which refers to the crucifixion. .... I think that’s right. Anyway if you have a gloss along with a translation, it’s easier to include context like that as footnotes on individual words/phrases.
- thaumasiotes 2y ago
  > The choice of the word “fer” isn’t the most obvious choice for “bring”, though. Well, if you asked me how to say "bring" in Latin, that would be my first choice, and the irregularity of the verb tells us that it's very common in general, though not necessarily for this. Lewis and Short has "In general, to bear, carry, bring"; "In particular, to move, bring, lead, conduct, drive, raise", which seems to hit the concept of "bringing help" squarely in the center. https://www.perseus.tufts.edu/hopper/morph?l=fero&la=la#lexicon https://www.perseus.tufts.edu/hopper/morph?l=fero&la=la#lexi...
kamma4434 2y ago
I think it’s pretty cool, but with Latin a simple word by word translation will quickly become unmanageable because you need to re-order and break up sentences to make them understandable in other languages. If you take Cicero for example, it is common to have periods that last one full page of text.
Alex-Programs 2y ago
If you'd like to try to use Wiktionary instead of OpenAI, you may find this useful: https://dictionary.nuenki.app/get_definition?language=LatinClassical&word=quo https://dictionary.nuenki.app/get_definition?language=LatinC... The code is here: https://github.com/Alex-Programs/nuenki-dictionary https://github.com/Alex-Programs/nuenki-dictionary However, feel free to use the API for small scale usage; the API can handle ~5 orders of magnitude more requests than it currently receives. You can see that each word has many different definitions. It's very difficult to do a word-by-word translation that takes context into account, though I'm going to attempt it at some point using a small LLM that merely picks from Wiktionary data for a https://nuenki.app https://nuenki.app hover mode.
sydbarrett74 2y ago
Thank you for creating this. I took Latin in HS (over 30 years ago!) and would like to brush up.

mentalgear 2y ago

Thank you for the details: I really like the minimalist UI, well done! And I like to see you sharing the tools that you build to help others learning. That being said, I'm surprised though, that the translations are GPT generated, so not sure how trustworthy this actually is. Domain foreign users have to be able to trust that learning resource are proof-read / accurate. Not to say it's worthless, but you may want to note that the translations were done automatically and may contain errors.

jrflowers 2y ago

>I used OpenAI to do the translation which looks pretty good for me, with the caveat that... well... I do not know Latin OpenAI also does not know Latin. This is either a tool to troll people that can read Latin or a tool to help people “learn” a made up vaguely Latin-shaped set of gibberish that ChatGPT nondeterministically generated. This only works for a definition of “Latin” that is a sort of vibe wholly detached from structure, syntax, or vocabulary.

rahimnathwani 2y ago
vaguely Latin-shaped set of gibberish that ChatGPT nondeterministically generated OP hadn't used AI to generate Latin, but to generate English.
- jrflowers 2y ago
  It doesn’t matter which set of text you generate with ChatGPT in this case. Using it in either makes the output useless as a tool to learn anything about both sets of text. This issue is compounded further when, as the OP readily admits, there is no quality check involved (OP does not know Latin)
  - rahimnathwani 2y ago
    Using it in either makes the output useless as a tool to learn anything about both sets of text OP is using it as a tool to improve their comprehension of Latin. The tool shows a word by word translation, which is jarring to read linearly, but works well for filling in gaps. It's far from useless as an aid to comprehend the Latin text. It's been over thirty years since I last studied Latin, but I still remember enough to be able to tell that this tool would be useful for a learner, even without perfect accuracy.
empath75 2y ago
You're severely overstating the case here, as if it just produces random text like a markov generator or something.
dmd 2y ago
I do know Latin (quite well), and gpt-4o does an extremely good job at translating from English to Latin and vice versa.
- nescioquid 2y ago
  Note how one opinion provides the warrant for the other.

satvikpendem 2y ago

Duolingo really isn't good for learning languages. When I learned Latin, I used the immersive method of just starting to read simple texts to more complex ones. The principal book of this form is called Lingua Latina Per Se Illustrata, I have the color PDF if anyone so needs it. It's a book that starts off with very simple sentences and gradually introduces more complex topics like tenses and declensions as the story goes along.

thaumasiotes 2y ago
> It's a book that starts off with very simple sentences and gradually introduces more complex topics like tenses and declensions as the story goes along. I'm not sure this is actually an approach I'd recommend. I was recently asked to give some supplemental English tutoring to a Chinese brother and sister, 9 and 5 years old. The 5-year-old could already use and understand 'simple' sentences such as "what do you see?" and "where is your brother?", though I'll note that the subject-auxiliary inversion required by a question of that form isn't exactly a simple concept. I got them a copy of The Cat in the Hat, and their mother objected that it was too advanced for either of them, because most of the verbs in The Cat in the Hat are in the past tense, which apparently isn't covered within the first four years of Chinese English instruction. You can't learn what you're not exposed to, but you can learn a lot of what you are exposed to in a language.
- satvikpendem 2y ago
  Lingua Latina is for adults who already have some base level knowledge of tenses in their own language, preferably a Romance or Germanic language (as I believe some languages don't have tenses), not for children who have no concept of them. Once you start reading the book, it really does start to make sense while teaching you the various forms. It's on Internet Archive if anyone wants to read it: https://archive.org/details/lingva-latina-per-se-ilustrata-pars-i https://archive.org/details/lingva-latina-per-se-ilustrata-p...
  - thaumasiotes 2y ago
    > preferably a Romance or Germanic language (as I believe some languages don't have tenses) A couple of points: - If you natively speak a Romance language, learning Latin by example is going to be really easy for you. This doesn't belong in a comparison with anything else. - Germanic speakers have no special advantage over any other Indo-European speakers. - You might be interested to know that while Mandarin verbs don't inflect for tense, the negative particle does, so you have to observe a distinction between past and present tense whenever you're negating a verb. 我不理解 I don't understand 我没有理解 I didn't understand

Timwi 2y ago

I love this mode of reading very much. I think this could be great for any language, not just Latin, and any target language, not just English. The only difficulty is sourcing good texts that are both fun to read and also suitable for language learning at various levels of skill.

thaumasiotes 2y ago
I've been working on a tool to display the structure (and glosses) of a text. All the structural and definitional annotation is provided by me based on personal knowledge. So far I have most of a single short fable in Mandarin Chinese done. ;D Does this sound like something you'd be interested in checking out?

eyko 2y ago

I do not speak latin, although I studied it for two years in high school and I'm a native speaker of a romance language, so my understanding of latin is pretty much basic to guesswork. This is a really cool tool -- I often read latin texts with the original on one page and the translation on the other, just because I think it's interesting to see how they wrote/spoke at the time, but for the most part certain words or declinations throw me off guard. Inline literal translations really help there. That being said, I noticed whilst reading some of the texts that the inline literal translations are still in latin, e.g. in "Part IV. I Some Barbarous Customs", most of the translated text is just latin. I guess OpenAI won't take all our jobs just yet! I do have one suggestion for improvement though. Many of these texts have translations that are already in the public domain (older translations). It would be helpful to display the original Latin and a fluent English translation side by side, whilst still being able to toggle the literal translation on or off. This setup would make it easier to compare the original text with a fluent English translation, similar to the format used in some bilingual books.

philsnow 2y ago

“Translation of each word” is also called a “gloss”, and I think it’s absolutely vital for trying to read works in translation. Lots of words and phrases have multiple meanings and connotations in their origin language and it’s not usually possible for a translation to bring the richness into the target language. (I’m going to butcher this because I don’t have the text in front of me, but) Thomas Aquinas composed several hymns for the feast of Corpus Christi, one of which is “O salutaris hostia”, which contains a reference to “fer auxilium” which is often translated to “bring help”. The choice of the word “fer” isn’t the most obvious choice for “bring”, though. Some translators have speculated that Aquinas chose “fer” because of how close it is to “ferculum”, which is a litter or wooden frame upon which spoils are carried, which refers to the crucifixion. .... I think that’s right. Anyway if you have a gloss along with a translation, it’s easier to include context like that as footnotes on individual words/phrases.

thaumasiotes 2y ago
> The choice of the word “fer” isn’t the most obvious choice for “bring”, though. Well, if you asked me how to say "bring" in Latin, that would be my first choice, and the irregularity of the verb tells us that it's very common in general, though not necessarily for this. Lewis and Short has "In general, to bear, carry, bring"; "In particular, to move, bring, lead, conduct, drive, raise", which seems to hit the concept of "bringing help" squarely in the center. https://www.perseus.tufts.edu/hopper/morph?l=fero&la=la#lexicon https://www.perseus.tufts.edu/hopper/morph?l=fero&la=la#lexi...

kamma4434 2y ago

I think it’s pretty cool, but with Latin a simple word by word translation will quickly become unmanageable because you need to re-order and break up sentences to make them understandable in other languages. If you take Cicero for example, it is common to have periods that last one full page of text.

Alex-Programs 2y ago

If you'd like to try to use Wiktionary instead of OpenAI, you may find this useful: https://dictionary.nuenki.app/get_definition?language=LatinClassical&word=quo https://dictionary.nuenki.app/get_definition?language=LatinC... The code is here: https://github.com/Alex-Programs/nuenki-dictionary https://github.com/Alex-Programs/nuenki-dictionary However, feel free to use the API for small scale usage; the API can handle ~5 orders of magnitude more requests than it currently receives. You can see that each word has many different definitions. It's very difficult to do a word-by-word translation that takes context into account, though I'm going to attempt it at some point using a small LLM that merely picks from Wiktionary data for a https://nuenki.app https://nuenki.app hover mode.

sydbarrett74 2y ago

Thank you for creating this. I took Latin in HS (over 30 years ago!) and would like to brush up.