Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How Translation Works, Book Title Edition

whatever.scalzi.com

Post reply on HN
Print view View original

11–20 of 24 posts

Re: How Translation Works, Book Title Edition

#11

Post by layer8 »

lmm wrote: post #8
laurieg wrote: post #7

Earlier quoted context omitted.

I love catching these little translation flubs. Japanese movie and TV subtitles seem to have quite a few, probably because of the aforementioned time pressure. For example, I saw Oppenheimer in the theater and "crown" had been mistranslated as "clown". Perhaps my favorite subtitle mistake of all time is in the subtitles of Brooklyn 99. One of the characters talks about getting someone a "boogie board" (a small surfbo…

You might enjoy the notorious Star War the Third Gathers: Backstroke of the West. At turns ludicrous and oddly poetic.

It has a higher IMDB rating than the original: https://www.imdb.com/title/tt18183916/ ;)

Re: How Translation Works, Book Title Edition

#12

Post by barlog »

IsaacZerh wrote: post #2

"Common phrases in one language don’t exist in another; cultural references in one country mean nothing elsewhere, and so on. This is why a computerized translation is fine for a bland business email but will utterly fail for a novel." I'm sure it's going to be indistinguishable very soon, if not already.

yes, WIP :)

Re: How Translation Works, Book Title Edition

#13

Post by numpad0 »

IsaacZerh wrote: post #2

"Common phrases in one language don’t exist in another; cultural references in one country mean nothing elsewhere, and so on. This is why a computerized translation is fine for a bland business email but will utterly fail for a novel." I'm sure it's going to be indistinguishable very soon, if not already.

Sure, translations work better for bland, unpleasant, and redundant text, regardless of purposes.

So if business emails gets creative and dramatic, there will be heightened need for resources and risk for mistranslations, or if a novel would be written like law text, there will be less ambiguity or processing cost. Or if were to be written relying on English-centric LLM, it'll do worse in languages other than en_AI.

Re: How Translation Works, Book Title Edition

#14

Post by ericrallen »

tkgally wrote: post #4

This is a nice essay, with an excellent example of the challenges and art of literary translation. But the author seems to assume that "computer translation" is limited to services like Google Translate or DeepL: put in a text, get a translation, with no consideration of the context or purpose. Properly prompted LLMs yield much better results and are already helping human translators produce better translations than…

Not an up or down here, but I’d be curious to see some sources for this claim:

> Properly prompted LLMs yield much better results and are already helping human translators produce better translations than they could on their own.

Re: How Translation Works, Book Title Edition

#16

Post by jonathanlb »

tkgally wrote: post #4

This is a nice essay, with an excellent example of the challenges and art of literary translation. But the author seems to assume that "computer translation" is limited to services like Google Translate or DeepL: put in a text, get a translation, with no consideration of the context or purpose. Properly prompted LLMs yield much better results and are already helping human translators produce better translations than…

I've been out of the translation game for quite a while, but I remember there being more to translating as a profession than just communicating ideas from one language to another. Sometimes precision and consistency is needed; complex scientific manuals come to mind. I recall using translation tools (e.g., translation memory) being essential to providing quality translations. TMs would help ensure that the same terms were used consistently throughout the text, and that similar phrases were translated similarly as well. The TM was an essential part of the workflow, because the first line of translators would pass their translation and the TM off to an editor, who would make suggestions in the text and TM, and the translator would update the TM based on those suggestions. It's been a while so I might have gotten a detail or two wrong, but that's the gist of the workflow.

I'm curious what sorts of prompts can replicate this. Again, there's a lot I've missed in the decade or so that I've been out of the industry. Does translation memory now become part of the context window? Does the context window self-actualize with new translations? What sorts of prompts can help produce quality translations?

Re: How Translation Works, Book Title Edition

#17

Post by tkgally »

jonathanlb wrote: post #16
tkgally wrote: post #4

This is a nice essay, with an excellent example of the challenges and art of literary translation. But the author seems to assume that "computer translation" is limited to services like Google Translate or DeepL: put in a text, get a translation, with no consideration of the context or purpose. Properly prompted LLMs yield much better results and are already helping human translators produce better translations than…

I've been out of the translation game for quite a while, but I remember there being more to translating as a profession than just communicating ideas from one language to another. Sometimes precision and consistency is needed; complex scientific manuals come to mind. I recall using translation tools (e.g., translation memory) being essential to providing quality translations. TMs would help ensure that the same terms…

Thanks for the reply. A few days ago, I posted several comments explaining my own use of LLMs in translation and related issues in this thread:

https://news.ycombinator.com/item?id=42894215#42895610

I didn’t say anything about translation memory tools, though, nor about maintaining vocabulary consistency. Those are not big issues in the type of translation I do now, though I recognize that they are important for many types of technical translation.

I actually don’t know how good LLMs are at maintaining vocabulary consistency even when they are provided with glossaries or translation histories. That deserves some testing.

Re: How Translation Works, Book Title Edition

#18

Post by tkgally »

ericrallen wrote: post #14
tkgally wrote: post #4

This is a nice essay, with an excellent example of the challenges and art of literary translation. But the author seems to assume that "computer translation" is limited to services like Google Translate or DeepL: put in a text, get a translation, with no consideration of the context or purpose. Properly prompted LLMs yield much better results and are already helping human translators produce better translations than…

Not an up or down here, but I’d be curious to see some sources for this claim: > Properly prompted LLMs yield much better results and are already helping human translators produce better translations than they could on their own.

Thank you for the question. A few days ago, I posted several comments explaining my own use of LLMs in translation and related issues in this thread:

https://news.ycombinator.com/item?id=42894215#42895610

I didn’t mention it there, but I have had quite a few discussions with professional translators about LLMs over the last two years. Their reactions and attitudes towards AI are definitely mixed, but I do know translators other than myself who are using LLMs productively as part of their translation process.

Re: How Translation Works, Book Title Edition

#19

Post by rererereferred »

Sometimes translators make weird decisions. Ocean's eleven in Latam Spanish was "La gran estafa" (The great swindle), which to me sounded much more generic, then came Ocean's twelve and they called it "La nueva gran estafa" (The new great swindle). On the 3rd movie they had to give in and called it "Ahora son 13" (Now they are 13), losing the connection to the others completely. For the next one they didn't even bothered and called it "Ocean's 8: las estafadoras" (Ocean's 8: The swindlers).

Bonus: the original movie from the 60s was named "Once a la medianoche" (Eleven at midnight). I would have preferred they kept that name.

Post reply on HN