Should machine-translated chapters be appropriately tagged?

Dex-chan lover
Joined
Mar 18, 2019
Messages
116
Dex-chan lover
Joined
Jan 8, 2023
Messages
193
I mean if some off-brand manga websites have a [Rough Translation] tag, why shouldn't regular aswell
 
Dex-chan lover
Joined
Jan 18, 2018
Messages
687
That doesn’t prove that the scanlation isn’t MTL, it could mean the officials are also MTLed, and I think I’ve seen accusations of this being a thing
Very close means nothing. Not only do any puns get lost in translation without explanation, there's no way for English readers to know that it was originally some sort of joke at all. In general, all nuance is lost when machine translating. Subtle foreshadowing, jokes, carefully-chosen language; it's all gone. It's an insult to all writers to tell them that their words don't matter as long as the gist is being conveyed.

I heavily advocate for it to be disallowed from the site, but I know that doing so isn't the general sentiment of the staff.
You two are very much missing the point. Nothing you are talking about actually has anything to do with "MTL" and has everything to do with effort. Puns are lost in translation in human translation too unless someone puts in the effort to try and make it work.

And Lobuttomize, nuance is not lost on all MTL, nuance is just lost on low effort MTL not all MTL is just mindlessly copying and pasting what the machine puts out. And what about all the low effort human translations on the site? And word choice nuance? Are you kidding me? That literally still exist with human translation. hardly any words have 1 for 1 the exact same meaning in different languages. Because culture is also part of it. Two words could have the same literal translation, but could have wildly different vibes they give off using them. And when translating, unless you know the CULTURE not just the language, you might get it wrong. And the number of translators who have lived at least 5 years in the country they are translating for is very slim. Are you complaining about their possibility to miss cultural nuance? No, because if they put in teh effort they can figure it out, just like someone doing MTL.
Your opinion doesn't change the fact that machine 'translations' aren't translations, they are inaccurate words spit out by google translate and then glued together by some random guy that doesn't speak the langauge that is being translated. They are not translations.
This is just an uneducated take completely. Firstly, your 'knowlege' of a language, unless you speak it natively, or learned it from a person who speaks both your language and the other language natively, is based on the same kind of thing as an MTL database, essentially a dictionary of word translations and rules for how to use said words. Just because a human is doing it doesn't suddenly give it a magical power. And just like translators have to edit, a good MTL chapter has editing and proof reading to make sure things make sense. And if something doesn't seem to fit the context, you know what they can do? They can use this magical thing called the internet to look it up and find people who can explain it to them. Literally MTL has every ability to be literally as accurate as a human translation, it's entirely up to how much effort you are willing to put into the other steps to make sure it is right.
 
Dex-chan lover
Joined
May 18, 2019
Messages
3,734
Don't forget the poor grammar tag
You can't say poor, that'll hurt the feelings of people who got tagged with that tag. You gotta say "rough grammar".




No that was the actual argument in that stupid fucking thread.

I mean if some off-brand manga websites have a [Rough Translation] tag, why shouldn't regular aswell
I have seen how well that tag worked on the site. It doesn't. It's way too much fucking effort to tag something that most people don't give a shit about. I stopped doing verification for those poor translation/grammar tags after a month. People just love to slurp up all the low quality shit it's not worth my time. Like besides daruna and cutegyaru, who else are doing all the shit to keep the tags valid?
 
Last edited:
Dex-chan lover
Joined
Oct 9, 2019
Messages
1,970
You can't say poor, that'll hurt the feelings of people who got tagged with that tag. You gotta say "rough grammar".
You know what's rough? The amount of people that seem like they're one bad away from sending death threats to the poor translator
 
Fed-Kun's army
Joined
Jan 16, 2019
Messages
116
You two are very much missing the point. Nothing you are talking about actually has anything to do with "MTL" and has everything to do with effort. Puns are lost in translation in human translation too unless someone puts in the effort to try and make it work.

And Lobuttomize, nuance is not lost on all MTL, nuance is just lost on low effort MTL not all MTL is just mindlessly copying and pasting what the machine puts out. And what about all the low effort human translations on the site? And word choice nuance? Are you kidding me? That literally still exist with human translation. hardly any words have 1 for 1 the exact same meaning in different languages. Because culture is also part of it. Two words could have the same literal translation, but could have wildly different vibes they give off using them. And when translating, unless you know the CULTURE not just the language, you might get it wrong. And the number of translators who have lived at least 5 years in the country they are translating for is very slim. Are you complaining about their possibility to miss cultural nuance? No, because if they put in teh effort they can figure it out, just like someone doing MTL.

This is just an uneducated take completely. Firstly, your 'knowlege' of a language, unless you speak it natively, or learned it from a person who speaks both your language and the other language natively, is based on the same kind of thing as an MTL database, essentially a dictionary of word translations and rules for how to use said words. Just because a human is doing it doesn't suddenly give it a magical power. And just like translators have to edit, a good MTL chapter has editing and proof reading to make sure things make sense. And if something doesn't seem to fit the context, you know what they can do? They can use this magical thing called the internet to look it up and find people who can explain it to them. Literally MTL has every ability to be literally as accurate as a human translation, it's entirely up to how much effort you are willing to put into the other steps to make sure it is right.
You're wrong. Nuance can be lost in human translation, obviously. But it is always lost in MTL. It's simply not good enough, yet. If you speak to any legitimate translator; they will tell you this. You can't argue against it. A machine translator cannot add nuance with effort, because they have no idea that it's there, as they can't read it themselves. All they can do is clean up whatever nonsense Google Translate spits out. DeepL does this automatically, making it seem like it's a better translator, but, it's just a less direct translation.
 
Last edited:
Dex-chan lover
Joined
Jan 18, 2018
Messages
687
You're wrong. Nuance can be lost in human translation, obviously. But it is always lost in MTL. It's simply not good enough, yet. If you speak to any legitimate translator; they will tell you this. You can't argue against it. A machine translator cannot add nuance with effort, because they have no idea that it's there, as they can't read it themselves. All they can do is clean up whatever nonsense Google Translate spits out. DeepL does this automatically, making it seem like it's a better translator, but, it's just a less direct translation.
You're the one who is just wrong. You're speaking in absolutes and that's a clear indication of your lacking. Firstly, "legitimate' translators outside of our little world of scalation use machines ALOT there are problems no doubt but they are not so flaws as to be entirely unreliable. The reason nuance is lost is because of lack of effort most the time. If the machine has a 0% chance of giving you enough to work with to identify and incorporate the nuance into a translation, then literally that means the data for it doesn't exist in it's system. And for that to happen it would mean you are claiming that people just straight up aren't putting it in.

Further showing how uneducated you are. DeepL is not just "Cleaning up what google translate says" they don't even use the same databases.
 
Dex-chan lover
Joined
Aug 15, 2020
Messages
676
I don't it matters really unless it's really bad (which you should just block and forget), or someone is profiting off of (which they shouldn't because piracy is a crime). This is just a non-issue honestly, people have been using Google translate for a long time already and no one bat an eye
 
Fed-Kun's army
Joined
Jan 16, 2019
Messages
116
You're the one who is just wrong. You're speaking in absolutes and that's a clear indication of your lacking. Firstly, "legitimate' translators outside of our little world of scalation use machines ALOT there are problems no doubt but they are not so flaws as to be entirely unreliable. The reason nuance is lost is because of lack of effort most the time. If the machine has a 0% chance of giving you enough to work with to identify and incorporate the nuance into a translation, then literally that means the data for it doesn't exist in it's system. And for that to happen it would mean you are claiming that people just straight up aren't putting it in.

Further showing how uneducated you are. DeepL is not just "Cleaning up what google translate says" they don't even use the same databases.
You misunderstand what I'm saying and you don't know what you're talking about
 
Group Leader
Joined
Jul 15, 2019
Messages
6,564
You misunderstand what I'm saying and you don't know what you're talking about
gfg life
darjeeling-girls-und-panzer.gif
 
Dex-chan lover
Joined
Mar 18, 2019
Messages
116
This is just an uneducated take completely. Firstly, your 'knowlege' of a language, unless you speak it natively, or learned it from a person who speaks both your language and the other language natively, is based on the same kind of thing as an MTL database, essentially a dictionary of word translations and rules for how to use said words. Just because a human is doing it doesn't suddenly give it a magical power. And just like translators have to edit, a good MTL chapter has editing and proof reading to make sure things make sense. And if something doesn't seem to fit the context, you know what they can do? They can use this magical thing called the internet to look it up and find people who can explain it to them. Literally MTL has every ability to be literally as accurate as a human translation, it's entirely up to how much effort you are willing to put into the other steps to make sure it is right.
MTL is not real translation. Stop mental gaming yourself
 
Group Leader
Joined
Jul 15, 2019
Messages
6,564
MTL is not real translation. Stop mental gaming yourself
let the bro imagining himself in a dream where Art isnt translation.
images


How can a machine have a complex thought like human?
you dont have to go as far as it can translate as natural as a human would, it will never be real in such a short time. @WillLi
 
Dex-chan lover
Joined
Jan 18, 2018
Messages
687
You misunderstand what I'm saying and you don't know what you're talking about
Says the person who doesn't even know that DeepL and Google translate work on entirely different mechanisms for translation and DeepL LITERALLY was praised for being more nuanced than google translate by professional translators. But says "All nuance is lost with MTL." Yeah I'm the one who doesn't know what I'm talking about huh?
MTL is not real translation. Stop mental gaming yourself
You don't even know what real translation is. MTL is used all teh time professionally and I'm not just talking about about 'cheap work" or anything. We're talking about serious high level translation, of way more consequence than teh stuff going on on this site.
let the bro imagining himself in a dream where Art isnt translation.
images


How can a machine have a complex thought like human?
you dont have to go as far as it can translate as natural as a human would, it will never be real in such a short time. @WillLi
That might be the dumbest thing yet. Whether or not it can think like a human is entirely irrelevant to if a translation is 'real'. And even if thinking like a human was necessary, guess what, there is a human involved in all MTL. MTL doesn't have to be just copy pasting whatever the bot spits out and calling it a day. You can read what it put, cross reference words, compare the dialogue to the flow of the chapter and make sure it seems like it makes sense, and even ask other people when you're lost on how something messed up. MTL doesn't need to 'think like a human' if just needs an accurate dictionary, then a human comes behind it to check it. And they have access to the entirety of the internet if they want to to make sure they get it right.

You guys act like normal translation is perfect, but I have yet to see a single manga get translated by two groups and they translate the dialogue exactly the same. (Really easy to see with a lot of Korean series) They can't both have the nuance 'right' by your standards. Maybe just maybe nuance never could be perfectly translated in the first place?

All you guys have done is come up with stupid criteria and hurdles to say MTL is bad, but normal translation doesn't even meet your standards half the time but you ignore it because "it was done by a human." Most of you probably can't even read the languages of the comic's you read in the slightest, and no nothing about the work of translation. You've just heard "MTL bad" and seen a few bad ones and decided that's it they must be bad.
 
Group Leader
Joined
Jul 15, 2019
Messages
6,564
Says the person who doesn't even know that DeepL and Google translate work on entirely different mechanisms for translation and DeepL LITERALLY was praised for being more nuanced than google translate by professional translators. But says "All nuance is lost with MTL." Yeah I'm the one who doesn't know what I'm talking about huh?

You don't even know what real translation is. MTL is used all teh time professionally and I'm not just talking about about 'cheap work" or anything. We're talking about serious high level translation, of way more consequence than teh stuff going on on this site.

That might be the dumbest thing yet. Whether or not it can think like a human is entirely irrelevant to if a translation is 'real'. And even if thinking like a human was necessary, guess what, there is a human involved in all MTL. MTL doesn't have to be just copy pasting whatever the bot spits out and calling it a day. You can read what it put, cross reference words, compare the dialogue to the flow of the chapter and make sure it seems like it makes sense, and even ask other people when you're lost on how something messed up. MTL doesn't need to 'think like a human' if just needs an accurate dictionary, then a human comes behind it to check it. And they have access to the entirety of the internet if they want to to make sure they get it right.

You guys act like normal translation is perfect, but I have yet to see a single manga get translated by two groups and they translate the dialogue exactly the same. (Really easy to see with a lot of Korean series) They can't both have the nuance 'right' by your standards. Maybe just maybe nuance never could be perfectly translated in the first place?

All you guys have done is come up with stupid criteria and hurdles to say MTL is bad, but normal translation doesn't even meet your standards half the time but you ignore it because "it was done by a human." Most of you probably can't even read the languages of the comic's you read in the slightest, and no nothing about the work of translation. You've just heard "MTL bad" and seen a few bad ones and decided that's it they must be bad.
Do you even read the thread's title? Do you need some machine to read it for you to be able understanding what it is to be having a decent conversation on how to make mangadex great again? None of your shit are make any sense and you are talking with your mind based on how in general scanlation works.

Of course you can use MTL and reviewed what the "machine" says, but thats how it is basically with someone who did it without a machine. Oh, my god it is fast because it read from dictionary so we just make the sentence good enough for human to comperhend what it means.

I didn't even say a thing about it's bad. I'm just saying we're both are stupidly equal because you are replying to non sense instead of helping why would it be good to have a tags for MTL.

gurren-lagann-kamiya.gif

You are basically as bad as us shitposter trying to derail whether you think it's a make sense argument in this thread.
 
Dex-chan lover
Joined
Feb 4, 2023
Messages
115
Do you even read the thread's title? Do you need some machine to read it for you to be able understanding what it is to be having a decent conversation on how to make mangadex great again? None of your shit are make any sense and you are talking with your mind based on how in general scanlation works.

Of course you can use MTL and reviewed what the "machine" says, but thats how it is basically with someone who did it without a machine. Oh, my god it is fast because it read from dictionary so we just make the sentence good enough for human to comperhend what it means.

I didn't even say a thing about it's bad. I'm just saying we're both are stupidly equal because you are replying to non sense instead of helping why would it be good to have a tags for MTL.

gurren-lagann-kamiya.gif

You are basically as bad as us shitposter trying to derail whether you think it's a make sense argument in this thread.
I can't decipher a single thing in this post. Holy crap, the cringy weebs from the early 2010s are back.
 

Users who are viewing this thread

Top