Should machine-translated chapters be appropriately tagged?

[WillLi]

Complete misconception. "If you can't operate independently of the MTL you can't audit it or 'proofread' it to make it more accurate! Proofreading is not the same as TLQC! Most of the people who proofread the MTL are just guessing at what the true character of the text is, rather than making an informed decision."
Is a non searchable, opt-in only vanity tag for chapter uploads really so bad? You can't say that having a chapter marked as MTL is a punishment if it's the uploaders own personal decision.

This is all I'm going to reply to, because this is all that needs to be replied to to show just how off base you are.

Even without actually having enough knowledge to use Japanese on your own you can audit MTL, you know why? The freakin' internet exist, and the manga doesn't exist as just text. Any time you look at what the MTL gave you and it doesn't seem to fit the flow of what is going on and being said, you can use other translation services, as well as asking forums and real people for what phrases mean and how to use them. It's part of the process for self teaching yourself a language.

The fact that you can't think of that shows you're not actually thinking you're just spouting anything negative that comes to mind. That's also why you can't see the negativity loop of this tag. Because if people use it a lot of people will not even click on their chapter. And yes it's opt in, but if they don't use it, people will start to complain and start reporting (even though it's non reportable) and people will start treating the group even worse on the premise of them being 'deceitful' even if in chapter they say it, just don't use the tag. People are foolish like that.

But that's all I need to hear from you. You're non productive in this discussion and you're not here to discuss you're only here to rant, so with this showing the scope of your willingness to actually think, I wont be replying to you anymore.

 

[happypurple]

Even without actually having enough knowledge to use Japanese on your own you can audit MTL, you know why? The freakin' internet exist, and the manga doesn't exist as just text. Any time you look at what the MTL gave you and it doesn't seem to fit the flow of what is going on and being said, you can use other translation services, as well as asking forums and real people for what phrases mean and how to use them. It's part of the process for self teaching yourself a language.

As I have said before, the process you are describing is guesswork, not the creation of reliable output. are you even bothering to read the text you quoted?

The fact that you can't think of that shows you're not actually thinking you're just spouting anything negative that comes to mind. That's also why you can't see the negativity loop of this tag. Because if people use it a lot of people will not even click on their chapter. And yes it's opt in, but if they don't use it, people will start to complain and start reporting (even though it's non reportable) and people will start treating the group even worse on the premise of them being 'deceitful' even if in chapter they say it, just don't use the tag. People are foolish like that.

Your speculation seems probable, but I hardly think it's foregone conclusion, or even a net negative. If anything, chapters having a vanity tag might signal to users that MTL is allowed on the site, which would lead to less false reporting for MTL or "bad" scanlations. By giving uploaders the ability to add a tag to their uploads users will be able to recognize the chapter is potentially low quality and not get their hopes up before reading it and feeling disappointed, which seems to me to be a big driver of dissatisfied comments. People who know the chapter is MTL before even clicking on it but still complain in the comment section will just look unreasonable if there aren't any extenuating circumstances.
People can choose not to disclose they've used MTL at all. If the uploader writes a note in the credits page to say its MTL there are going to be people who complain regardless of if the upload is tagged, as there are already people who complain when groups put the MTL disclosure on the last page instead of the first page, adding a tag wouldn't make people get their hopes up. Since this would be opt-in by the uploader, there will be plenty of old chapters without the tag that have disclosures, so I don't imagine tagged chapters will be seen as the norm instead of the exception. People like to complain no matter what, so in my opinion the effects of adding a vanity tag are by far a net positive.

But that's all I need to hear from you. You're non productive in this discussion and you're not here to discuss you're only here to rant, so with this showing the scope of your willingness to actually think, I wont be replying to you anymore.

You may stick your head in the sand (haha) if you like, but I believe Ignoring ones problems will not solve them, so I will not stop criticizing any future posts you make until you bring counterclaims more convincing than "I was in an argument on reddit once". This issue is important to me so I intend to advocate for my perspective because I am confident it is right.

 

[doppler]

I for one don't like machine translations. They get cultural references all wrong. Only someone from the country would get the subliminal references correctly. As far as I suspect got it right. DeepL looks like Korean to English is good to go. Maybe something to do with the way they write. Sort of like German requiring a whole paragraph sentence, just to say "It's green."

 

[KiraEllion]

A warning label
For potential eye cancer
Would be welcoming.

i came here to actually address this.

Define eye cancer. not all MTL's are shit jobs not all non MTL's are legible.

people already do the notice thing that's more then enough "warning" for something that does not indicate quality at all.



I do however think that there should be a way to tag uploaded scans that did a terrible job

Edit:

I for one don't like machine translations. They get cultural references all wrong. Only someone from the country would get the subliminal references correctly. As far as I suspect got it right. DeepL looks like Korean to English is good to go. Maybe something to do with the way they write. Sort of like German requiring a whole paragraph sentence, just to say "It's green."

Thats where Experience on the part of the translator comes in. I can't speak or write Japanese but cultural jokes I have a good understanding of using a MTL does not change if a "cultural reference" is understood or not it just makes translating more accsessable.

 

[EOTFOFYL]

i came here to actually address this.

Define eye cancer. not all MTL's are shit jobs not all non MTL's are legible.

people already do the notice thing that's more then enough "warning" for something that does not indicate quality at all.

I do however think that there should be a way to tag uploaded scans that did a terrible job

Tis a harmless joke
At atrocious uploaders
Who never proofread.

 

[pandascepter]

Tis a harmless joke
At atrocious uploaders
Who never proofread.

eks de

 

[sterven]

 

[Aureliano]

I think MTL or not is not relevant, what is relevant is if it is a shitty translation or not, it just happens that as of right now, the vast majority of MTL ends up being shitty.
People known to have a good understanding of the languaje being translated to should have moderating powers to apply a "Rough Translation" (i would use shitty translation) tag.

 

[pandascepter]

I think MTL or not is not relevant, what is relevant is if it is a shitty translation or not, it just happens that as of right now, the vast majority of MTL ends up being shitty.
People known to have a good understanding of the languaje being translated to should have moderating powers to apply a "Rough Translation" (i would use shitty translation) tag.

Japon becoming a mainstream.

 

[AlexEliot]

people post probing questions in the suggestions forum all the time to figure out whether people are in support of it or not

Every thread in the suggestions forum is halted once it's been rejected, defeating the entire purpose of this post—which invites discussion.

 

[Gwyndolin-chan]

@happypurple and others don't like MTLs because of translation quality.
MD does not police translation quality.
Therefore any reasoning that revolves around MTL quality is irrelevant to the tag's implementation.

So, I think the benefits of having such a tag should be weighed strictly on the technical and use count merits of it.

Will groups include an MTL notice regardless of ability to tag with MTL? (if groups can crosspost to places other than MD, I'd imagine they'd include the notice in all of their release packages, in which case, the tag would not free up any data space. Thus, no tag should be implemented.)

If it was guaranteed that all releases on MD would replace image-based notices with a tag, there would probably be a data savings.

Would such a tag help readers? If most readers don't even care/can't tell if a translation is "bad" according to the standards of those who've posted in this very thread, then the tag would be a waste of time for everyone but those who ardently oppose MTL irrespective of quality. in which case, the tag would only serve that specific group of MTL abstaining readers. (Is that worthwhile?)

Given that apparently most readers don't care about quality, and MD itself does not police quality, I think the more important question is if such a tag would be automatically included or excluded in a the most-used search tool (quick search vs advanced search). Given that MD does not police quality, and the readers largely are indifferent, I would reason that an MTL tag, should it exist, should be automatically included by default in searches, and that the existence of an MTL tag would strictly only benefit the readers who specifically want to read translations that are not MTL (regardless of translation quality) and thus are willing to go out of their way to specify to exclude MTL tagged works.

If translation quality is truly, honestly an issue, then I think a human-driven translation quality auditing process that does not discriminate against either MTLs nor human translations would be faster and more useful than an auditing process targeting MTL tagged works given that it's already been admitted that human translations vary wildly in quality.

In this case, the auditing process would provide for an audit status field for each work with three levels: unaudited, audited (requires rewrite), and audited (acceptable). It would also be necessary to record the auditor and date of each audit, or some other way to establish the trust and quality of auditors.

and finally, as a reader, personally, I prefer a more direct translation (annotated as necessary) vs a liberal localization, legible over illegible text, and an ok translation over no translation at all. lastly, since i can only read one language, it's not like i could audit the quality of a translation anyway, so the MTL vs non-MTL quality/auditability debate is entirely moot to me. the ability to audit one's own output is entirely an internal issue between translators.

i see no reason to discriminate against MTLs if, evidently, there are humans whose output they couldn't even audit themselves correctly/accurately.

i dont use this site much as i don't read much manga and haven't for years, but just found this thread really fascinating after returning for two mangas randomly.

 

[happypurple]

I'd rather read a text wall than another half dozen off topic posts/complaints about text walls which add nothing to the convo

 

[WillLi]

Just found a great case example, "The Cost of a Broken Heart" the newest English chapters here on mangadex are machine translated. The grammar isn't great, but if you look at the official English chapters (get them where you want) It is VERY VERY close in 'intent' being conveyed. If they just had a better proof reader it'd be nearly 'perfect'. It shows why there is no real need to highlight that something is MTL.

 

[Remocracy]

Use the block button for shitty translation, mtl or not.

I’m already using my blocks for “Official” groups. If I wanted to read from your shitty site I’d just go there

 

[Remocracy]

Just found a great case example, "The Cost of a Broken Heart" the newest English chapters here on mangadex are machine translated. The grammar isn't great, but if you look at the official English chapters (get them where you want) It is VERY VERY close in 'intent' being conveyed. If they just had a better proof reader it'd be nearly 'perfect'. It shows why there is no real need to highlight that something is MTL.

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

 

[fasontales]

Yes they should be removed immediately. These 'translations' are not real, this isn't a debate it's a fact.

 

[Lobuttomize]

Machine translation needs to be marked as such so that it's visible from the list of chapters. This way, groups will be able to easily tell that chapters are machine translated without checking the contents of the chapter and—hopefully—pick up the series with a proper translation.

 

[Lobuttomize]

Just found a great case example, "The Cost of a Broken Heart" the newest English chapters here on mangadex are machine translated. The grammar isn't great, but if you look at the official English chapters (get them where you want) It is VERY VERY close in 'intent' being conveyed. If they just had a better proof reader it'd be nearly 'perfect'. It shows why there is no real need to highlight that something is MTL.

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.

 

[seekermoc]

There is no point to a tag that cannot be universally applied (as it would rely on the scanlator's honesty), nor one that only attempts to define the process and not the result.

I've read many series where the TL was done by and human, yet is so bad it's unreadable. Hell, I've proofread a series where the TL was not a native speaker of Japanese or English, and the initial draft of what he put out read worse than MTL.

I've also read many MTL series where the translation is so good that you'd never know it's MTL if the scanlator didn't tell you. What matters more is that the group has a good English PR (edit: or a native English TL).

Granted, there are occasions where there could be nuances or cultural references that a MTL just can't translate, but the same is true for any series where the TL isn't a Japanese native (and there's a LOT of those). Are we going to ask TLs to prove their native language and mark their chapters as "non-Native" TL? If not, then using this reasoning to stigmatize MTL holds no water.

Even if an MTL missed a pun or reference here and there, does it even matter? It might for some series, but the majority of manga isn't exactly deep literature, and a MTL group isn't likely to attempt a complicated series. As long as the translation reads well, and the missed reference isn't critical to understanding the story, then who cares?

It'd be infinitely more useful if users could nominate chapters for a "shitty TL" warning label, which would capture bad chapters, MTL or otherwise. Not that I actually expect the devs to consider implementing such a thing. Edit: nevermind this, it'd be way too susceptible to abuse.

 

[Remocracy]

Man people have a lot of strong opinion on MTL

Likely won't happen since https://forums.mangadex.org/threads/tag-for-machine-translated-chapters-rejected.1246834/