There was a fun one on OSNews:
https://www.osnews.com/story/142469/that-time-ai-translation...
What Google Translate can tell us about vibecoding
111–120 of 175 posts
Re: What Google Translate can tell us about vibecoding
#112> ... a translators’ and interpreters’ work is mostly about ensuring context, navigating ambiguity, and handling cultural sensitivity. This is what Google Translate cannot currently do. Google Translate can't, but LLMs given enough context can. I've been testing and experimenting with LLMs extensively for translation between Japanese and English for more than two years, and, when properly prompted, they are really go…
I've ended up doing a lot of research into LLM translation, because my language learning tool ( https://nuenki.app ) uses it a lot. I built something kinda similar, and made it open source. It picks the top x models based on my research, translates with them, then has a final judge model critique, compare, and synthesise a combined best translation. You can try it at https://nuenki.app/translator if you're interested…
Very nice! Thanks for the links.
Re: What Google Translate can tell us about vibecoding
#113Earlier quoted context omitted.
Americans are extremely polite, and American English is replete with niceties in everyday speech. It's funny, having been a tourist abroad, it's only living in a significantly non American country for a few years that lead me to realise this. I even lived in another country albeit a former US colony for ten years and didn't even notice it there given American influence. I stopped saying, stuff "I would like a latte t…
> Americans are extremely polite Having grown up in the UK and living in Australia, I do not find Americans polite. To me, politeness is "please", "thank you", "may I have", etc, whereas "I would like a latte today" reads to me as a demand. In context it's fine (it stands out a bit but not in a problematic way), it's not particularly rude, but in general just stating your desires is not considered polite in my experi…
As native-born American who has always lived in a Midwestern part of the States where visiting people often remark upon our unusual politeness:
I've never ordered a latte "today".
A simple statement of "I'd like a latte" fits fine in our regional vernacular.
I think that "I'd like a latte today" would appear a weird bit superfluous of specificity, unless it comes from a regular and familiar customer who might normally order something different and/or random.
I mean: "Today"? As opposed to later this week or something? It's implicit that the latte is wanted as soon as they can get around to it. (Unless yesterday's order was something different, and then adding "today" may make sense.)
---
But English is fucked up, and I'm getting old, so I've spent a lot of time observing (and sometimes producing) slight miscommunication.
In terms of things like ordering food or drink at a counter, it still works fine as long as nobody gets grumpy, and the desired item is produced.
I'm reminded of a local festival I went to as a little kid, with my parents, sometime. I was getting hungry and they were asking me what I wanted to eat. There were a lot of pop-up food vendor there, mostly with tables and stuff instead of the now-ubiquitous "food truck."
In the corner was a gyro stand with amazing-looking racks of spinny-meat. I wanted to try whatever that was.
The big banner said "GYROS" and we got in line.
Discussions were had between my parents about this "GYROS" concept, which they'd never seen before either. Was it a singular, or a plural? How many "GYROS" might a boy reasonable want? How was it pronounced? It looks like "gyro" as in "gyrocopter" but it seemed to them that it must be Greek, so they went through some variations on pronunciation as the line moved forward.
As we got closer, some of these questions were answered: The sign definitely referred to a plural offering, and seeing people leave it became clear that [unlike things like chorros or tacos or donuts] one was plenty.
And when we got to the front, the conversation went like this:
Parents: "Our son wants one of these... but we're not sure how to say it. Jye-roe? Hee-roe?"
"They're just gyros," he replied to them to them dismissively in a thick Greek accent, and motioned for his staff to produce 1 gyro.
And then the man looked at me, with his skin sweaty on that hot sunny day and a thick gold chain around his neck, and said to me in his very best and most careful English something to me that I can never forget. "I call them gyros. But listen to me, you can call them whatever you want. Jye-roe? Hee-roe? Yee-roe? Sandwich? Whatever you say, and however you say it: If you get what you expect, then you said it right. Alright?"
My trepidatious nods made sure that he was understood, and the awesome fucking sandwich-taco I had that day changed my entire perspective on food -- and ordering food -- forever.
So, sure: Ordering "one latte today"?
It sounds weird to me, but if it is understood and you get what you want, then it works. Correctness? Politeness? Whatever. Despite seeming weird: It works.
(Up next: There's a lot of ways to mispronounce approximately everything related to ordering pho, and none of them are wrong.)
Re: What Google Translate can tell us about vibecoding
#114Earlier quoted context omitted.
There was a great thread about that here four months ago: https://news.ycombinator.com/item?id=42894215#42897856
Good, but not exactly what I expected for "easily fixed". It includes a lot of steps and constant human evaluation between them, which implies that decisions about tone are ultimately made by whoever is prompting the LLM, not the LLMs themselves. > "If they are generally in the style I want..." > "choosing the sentences and paragraphs I like most from each..." > "I also make my own adjustments to the translation as I…
> It seems like a great workflow to speed up the work of an already experienced translator, but far from being usable by a layman due to the several steps requiring specialized human supervision.
Just like programming. And anything else "AI" assisted.
Helps experts type less.
Re: What Google Translate can tell us about vibecoding
#115The distinction between what people typically imagine a translator's job is and the reality reminds me of pixar movies being "localized" instead of just translated (green beans on a plate in the Japan release instead of broccoli because that's the food that Japanese kids don't like). Lacking cultural context while reading translated texts is what made studying history finally interesting to me.
Is that beans example a real thing? If so, I would hate for my kids to be subjected to that. The best thing about watching films from another country is that you're exposed to the culture of that foreign place and learn about how it's different from yours - I don't see why we'd try to localize away the human experience as if these differences don't exist.
Re: What Google Translate can tell us about vibecoding
#116This seems like a terrible comparison since Google Translate is completely beat by DeepL, let alone LLMs. (Google Translate almost surely doesn't use an LLM, or at least not a _large_ one given its speed)
That’s absolutely not the point of this article. The point was that people said, once Google Translate was introduced, that translators would lose their jobs. Just like people say the same thing about developers with LLMs nowadays. The point is: they didn’t, and they won’t. DeepL is not part of that point. Yes, maybe eventually, developers will lose their jobs to something that is an evolution of LLMs. But that’s not…
> The point was that people said, once Google Translate was introduced, that translators would lose their jobs. Just like people say the same thing about developers with LLMs nowadays. The point is: they didn’t, and they won’t.
Translators are losing their jobs now though. Google translate wasn't very good for Japanese so a lot of people assumed that machine translation would never be a threat, but deepl was better to the point where a lot of translation moved to just cleaning up it's output and current state of the art llms as of the last six months are much better and can also be given context and other instructions to reduce the need for humans to clean up the output. When the dust settles translation as a job is probably going to be dead.
Re: What Google Translate can tell us about vibecoding
#117Earlier quoted context omitted.
> Americans are extremely polite Having grown up in the UK and living in Australia, I do not find Americans polite. To me, politeness is "please", "thank you", "may I have", etc, whereas "I would like a latte today" reads to me as a demand. In context it's fine (it stands out a bit but not in a problematic way), it's not particularly rude, but in general just stating your desires is not considered polite in my experi…
As native-born American who has always lived in a Midwestern part of the States where visiting people often remark upon our unusual politeness: I've never ordered a latte " today ". A simple statement of "I'd like a latte" fits fine in our regional vernacular. I think that "I'd like a latte today " would appear a weird bit superfluous of specificity, unless it comes from a regular and familiar customer who might norm…
As an Australian, I'd modify my order with "today" if I'd asked for a flat white or a piccolo the last time I was in the shop. It would be a way to highlight that my order has changed, and said to a barista who knows me and is likely running on autopilot.
Re: What Google Translate can tell us about vibecoding
#118The distinction between what people typically imagine a translator's job is and the reality reminds me of pixar movies being "localized" instead of just translated (green beans on a plate in the Japan release instead of broccoli because that's the food that Japanese kids don't like). Lacking cultural context while reading translated texts is what made studying history finally interesting to me.
Is that beans example a real thing? If so, I would hate for my kids to be subjected to that. The best thing about watching films from another country is that you're exposed to the culture of that foreign place and learn about how it's different from yours - I don't see why we'd try to localize away the human experience as if these differences don't exist.
Same. I don't understand why people want (them and their kids) to be isolated from other cultures. If you're watching a movie set in Japan, India or China, have it be about their culture. If there's something you or your kids don't understand, make an effort to learn about it (and the green peas thing seems trivial to understand).
Netflix also does something absurd with their subtitles. I was watching "The Empress" (which is set in the Austro-Hungarian Empire) with German language and English subtitles. I like listening to the real actors' voices, and learning the sounds and cadence of the language. So the characters speak in Italian for a while (subtitles say "[speaking in Italian]", and when they switch back to German the subtitles clarify.. "[in English]". The fuck, Netflix? Surely the viewer of this show understands they didn't speak in English in the Austro-Hungarian empire, so why write it's English? What the hell is Netflix even trying to achieve here? Seems robotic: "us = English speakers, therefore everyone's default must be English"?
Re: What Google Translate can tell us about vibecoding
#119Earlier quoted context omitted.
CAD, Matlab, and Altium made electrical and mechanical engineers more valuable, not less. The work got easier, so what we do got more complex.
They’re all just tools. Use the tools or become obsolete.
Kind of a false dichotomy. A great example is debuggers vs print statements. Some people get by just fine with print statements, others lean heavily on debuggers. Another example is IDE vs plain vIM.
Becoming obsolete is a fear of people who are not willing or able to learn arbitrary problem domains in a short amount of time. In that case learning to use a particular tool will only get you so far. The real skill is being able to learn quickly (enthusiasm helps).
Re: What Google Translate can tell us about vibecoding
#120Earlier quoted context omitted.
Good, but not exactly what I expected for "easily fixed". It includes a lot of steps and constant human evaluation between them, which implies that decisions about tone are ultimately made by whoever is prompting the LLM, not the LLMs themselves. > "If they are generally in the style I want..." > "choosing the sentences and paragraphs I like most from each..." > "I also make my own adjustments to the translation as I…
> It seems like a great workflow to speed up the work of an already experienced translator, but far from being usable by a layman due to the several steps requiring specialized human supervision. Just like programming. And anything else "AI" assisted. Helps experts type less.
Most of my time as a programmer is not spent typing.