Agree. My family was watching Vagabond, and the guy said, “야! blah blah blah.” It should’ve been “Hey! blah, blah, blah,” but they translated it as “F**k! blah blah blah” instead. While the guy was mad at the other guy and was yelling at him, I don’t see why the translator felt the need to add a cuss word in the translation when the actor did not curse at all.
I don’t know how it looks on English NX, but managers of our regional branch don’t even know who is translating their shows. There are subcontractors of subcontractors and those subbers who are doing the translation get really low payment. Language of subs is very simple. And they are getting shorter and shorter. Soon we will be a nation of mumbling and grunting illiterates.
Something that bothers me is that some Viki subtitles in my native language (Brazilian Portuguese) are too literal, this word-by-word makes the sentences unnatural. For those who have a good knowledge of English, it is possible to see the “original English sentence” behind the translated one.
A teacher of mine used to say: “if you can tell what is the original sentence without looking at the source text, it is an incorrect translation”. Not all English verb tenses have a direct correspondence in pt-br, but quite often I see this transliteration (especially with phrasal verbs).
This interferes directly with the line length. I don’t say people here should use the CPS / CPL guidelines from Netflix or other streaming services, some of them are insane and unnecessary. Sometimes, a shorter synonym would help with readability. People feel inclined to translate all words, even if in Portuguese that word is not necessary, or if its usage configures a grammar mistake.
Grammar. That is another key problem. So many people eager to help (and become a gold QC), but in reality, don’t know basic rules well. I’m not saying that these people can’t contribute to translations, but they should avoid dramas with a complex vocabulary/subject. Like, if you are not acquainted with law / medical / business jargon, please, don’t engage in these dramas, unless you really have time to do proper research on the subject. A proofreader should not spend 10 times the length of the “part” to correct a subtitle.
I’m one of the people that would like to see “Oppa” and"Unnie" at Netflix translations instead of the first name of the characters, but I do understand the choice.
I remember the first drama I watched that had a word like that every few seconds, and it took me a while to get it straight. But I really love the culture, so I wasn’t mad by the fact that I needed to do research and sometimes even watch a scene a second time. I was fascinated by the familiar drama of this huge family I was seeing on the screen. My older sister didn’t like her Viki experience, but during this pandemic watched quite a few Korean and Chinese dramas on Netflix. Maybe, in a few years, she will be joining me here…
yes same…because I’m so invested in learning the culture too, I prefer Oppa and Unnie to first names, but someone just sitting down for entertainment will obviously find it weird and like it better if it was tailored for their own culture
I feel like since I was a native and got what everyone said for “oppa” “unni” etc, I was really worried on how other people might feel if I use that term. Hearing this has made me super happy knowing that you guys prefer that terminology.
To me, some terms carry an extra meaning scope that a simple “brother, sister, auntie…” can’t translate. But again, this is not important for those who are only after some entertainment. Maybe one day…
Ha, don’t worry, this phenomenon is to be found in every language. It is in Italian, it is in Greek and it is also in English - I used to wonder why some English sentences were so weird, and when I started learning Korean I understood that the Ko-En translators had been translating word by word, keeping the original word order and the idioms. Since then I have made it my business, as an English editor to hunt and exterminate all those. And because of that I risked misunderstandings with a couple of people I like a lot, like sophie2you, who is a friend of literal translations.
Some interesting threads on this - including hilarious examples - can be found here:
I hate it when they use Gods name in vain, AND all that swearing< gosh and to make a point? think they could find something better than those four letter words, and I’d like to know when some one says F-self would like to know how they do that! well sorry it still sounds funny to me!!
It’s not linked to the subtitles in themselves, but in the way they organize their work:
KNP refers to Key Names & Phrases Master Glossary (KNP)
They collaborate on spreadsheets for notes.
This the formality table:
It kinda looks like what other languages on Viki do for formality.
Then information on each character:
y’all just too much for me! not a computer wiz like you guys!
On their website, they say they do a quality control that they call QC:
Depending on the title (popularity or NX Original), they will do more or less quality control:
One type of quality control is focused on translation quality: localization QC.
Localization QC qualifies translation quality, consistency and style guide conformance. This process involves a QC operator reviewing the timed text asset, implementing changes and categorizing the reasoning for the changes.
Localization QC is performed on the following asset types:
- SDH/Closed Captions"
In their Subtitling Originator tool, they have warning messages for reading speed:
Zoom in (we can see 20c/s, 15 c/s, the final version on the left and the original version on the right with corrections the editor made like at the bottom “Literal Hard at work?”…):
– Comments from QC: they are communicated to the authors:
Zoom in, it has 6 columns:
Original language, Targeted language before QC, QCed version of the targeted language, change types (text, time), reasons and comments.
It kind of looks like our Bulk Editor, but we don’t have the last 3 columns for Quality Check + the column “Targeted language before edition”.
"Localization QC - Metrics Evaluation:
Error rates for each asset type are defined as follows.
- Subtitle error rate:
- Events with errors / total # of events
- Language error rate:
- Events with translation errors / total # of events
- Technical error rate:
- Events with technical errors / total # of events
- Audio error rate:
- Number of audio flags / total run time
Common Localization QC Errors:
- SGP (spelling, grammar, and punctuation)
- Objective Translation
- Subjective Translation
- Timing to Audio
- Reading Speed"
The last comment I found on their blog:
QC: How Does it Look?
"We have automated (or are working towards) a number of quality related measurements: these include spelling checks, text to audio sync, text overlapping burned-in text, reading speed limits, characters per line, total lines per screen. While such metrics go a long way towards improving quality of subtitles, they are by no means enough to guarantee a flawless user experience.
There are times when rendered subtitle text might occlude an important visual element, or subtitle translations from one language to another can result in an unidiomatic experience. Other times there could be intentional misspellings or onomatopoeia — we still need to rely on human eyes and human ears to judge subtitling quality in such cases. A lot of work remains to achieve full QC automation."
It’s info they publish on their website! There is a bunch of technical terms they use that I didn’t understand at all
Interesting. Thanks for sharing the info.
This layout assumes that any two people use the same level of formality with each other. Which is very often not true. For instance a subordinate will use formal speech level with his superior and the superior will talk to him casually. Or a child with an adult, student with a professor etc.
It says on top: “The table reads horizontally from left to right, i.e. Character 1 is formal “F” with Character 2. Character 2 is informal “I” with Character 1 . “F/I” indicates both are used depending on the context.”
It’s not the same cell when we take a closer look (C1 to C2: 3rd column and C2 to C1: 2nd column).
I think they probably update it like OL, people who are not very familiar at the beginning, but get closer.
I think it would be nice if Netflix can use Mr, Mrs, Miss etc. when polite language is being used. English is limited in that way unless extra effort is made to show honorifics. Somethings like Oppa, Unnie, Hyungnim etc will not translate properly into English due to cultural differences. At that point, they have to use proper names.
You cannot say Oppa as brother when girl is addressing someone she likes. It doesn’t make sense to first time viewer…
I zoomed my browser to 200% and I still see one shared cell. Okay, never mind, you’ll explain to me in person when we meet tomorrow.
This table is read from the left column to the line on the right side:
How will Irmar (left column) speak to… red line.
Yes!! See you tomorrow Nightmare pattern (sequel)
How are you guys meeting tomorrow? Is Piranna in Italy, too?
Neither of us is in Italy, LOL. We meet online twice a week for crochet lessons. We are making some rather complicated stuff these days. Fortunately we only have one hour difference.