Login
 
Not a member? Register now!
Announcements
Like us on Facebook, follow us on Twitter! Check out our new tumblr page. :)
Manga News: Toriko has come to an end; but be sure to swing by and share your favourite moments from the series!
Manhwa & Manhua News: Check out the new Manhwa and Manhua shout-out (March 2016).
Event News: One Piece Awardshas concluded. Check the results Here!
Translations: One Piece 913 by cnet128 , Gintama 693 by kewl0210

What can MH realistically do about translation quality?

+ posted by zindryr in Quality Control on Apr 16, 2009 10:45


Recently, there was a news post here at MH that cross-posted a well written article by Pomax in which he spoke about the history of improper English in manga and fan translations. This sparked some discussion, and I felt that the discussion needed a nice response from the administration here at MH before we moved on to what we CAN do about this problem, or if we can even call it MH’s problem.

What exactly is the problem?

I’d like to start off by defining the problem we are dealing with. If you’ve already read the previous article then you can probably skip this section as you are already aware, but I’m going to summarize what Pomax said. The English used in many scanlations these days is not in fact English, but a sort of hybrid English using English words, but not what anyone would ever really say in a real life situation. The problem with this is that it has lowered many readers’ expectations as to what to expect when they read manga, and has supposedly had a bad influence on official manga releases.

One thing I’d like to point out before moving on is that this problem is not entirely accurate. The problem mainly manifests itself in the world of speed scanlations where release speed is the key, and not quality. If the group is going for speed, will they really care if the translation they use is not completely proofed, revised, and mostly finished? No, probably not. Is there anything we at MH can feasibly do about this? I don’t believe so, but I would love to be proved otherwise. Anything that adds red tape and bureaucracy to the process of posting a translation to the public will only annoy both readers and the translators who post them.

Let’s face it, most people would rather read their favorite manga sooner rather than later, right? Anyway, now that we know what the problem is, let’s play the blame game.

Who is really at fault for the current state of English in scanlations?

Most fingers lay firmly pointed at random guy with one year of Japanese under his belt trying his best to translate his favorite series for the masses. To be honest? That was me five or more years ago. The very first manga I translated was Naruto for one of the first few speed scanlation groups (Bakafish) to hop on the series. Some people may remember that, some people may not. Regardless, I was translating a series I loved at the time out of love for the series. The response to our little team picking up the series and releasing a little more quickly than the current team scanlating was lukewarm to somewhat happy. I’m fairly certain the English I used back then would fall under the whole “not exactly English” category as well, but we never received any comments about that since it was mostly understandable.

Where am I going with this? My point is that the audience never complained about slightly weird English issues. They didn’t decide to not read our scanlations because the English was weird. Most of the complaints we got were about the scan quality. If there had been more complaints about the English I used back then, you can be sure I would have tried to fix it to the best of my abilities. I like to think I’ve grown as a translator since then, but the audience has not changed. If you’ve read this far then you likely do care about the quality of the English used in your scanlations, but being the cynic that I am I can’t imagine anyone who stopped reading several paragraphs ago actually caring that much about what I have to say. I’m pretty sure most of MH’s userbase didn’t even read the first sentence of this post.

Now that’s the cynic in me talking, and I would love to be wrong.

The point here is that the translators aren’t entirely at fault. If the audience doesn’t care enough to point out weird English sentence structures, then the translators might be less likely to improve in future chapters. Both parties end up being at fault here. I’m not so sure I want to blame someone who’s just started translating with little to no experience in the field though. Translating isn’t exactly something you can just do well from day one.

What can MH do about this problem?

Well, the main solution that has been put forward is one of instituting some sort of proofreading team. I’m not so sure this is the end-all-be-all solution to our problems, but it does have its merits in helping translators to improve. But once again, the main problem here is speed scanlations, which brings us back to audience apathy. Will adding in proofreaders honestly help with this problem that much?

Translator A posts his translation to Shounen Jump series B. Speed scanlation group C uses the translation right as it is posted and copy-pastes the translation before the proofreaders even have a chance to take a good look at the translation. Audience is generally pleased.

I won’t deny that proofreaders will help in the long run as the translator starts to get more experience, noticing their own mistakes since he or she has had the experience of being told that something is in fact a mistake. That sort of experience is invaluable in our line of work when we’re just starting out. It even helps those of us who have been doing this for a while. Everyone makes mistakes.

There is also another problem with this. Where are all these magical proofreaders with perfect English supposed to come from? The staff here at MH is stretched fairly thin as is, whether it’s with their own lives or the work that needs to be done here at MH. A lot of our staff (despite their remarkable skills in English) are also non-native speakers. Proofreaders don’t just appear by the dozen with endless amounts of time and an insatiable lust to proofread manga at our behest.

That’s where you come in. Yep, you. If you’re a native English speaker and you read the translations before they’re on speed scanlations, someone who doesn’t like the quality of translations these days, or someone who hates MH because they don’t police translations, I’m talking to you. You don’t have to proofread a whole translation. Just, if you notice something that doesn’t exactly sound correct then make a comment about it on the translation for the translator. Every time you do that, it will likely help the translator to improve in the future.

I say that because in all actuality, MH is a community of volunteers. We, the staff, can’t do everything by ourselves. The staff doesn’t post every translation you see on this site, the large majority are posted by members. The same goes for scanlations. The only reason MH has grown to what it is today is because we have some awesome members.

Another solution that was raised included the rating of translations or ranking translators by experience or through a series of tests. Rating systems lead to people questioning why they got a certain rating and general unhappiness if you continually receive low to mediocre ratings. This would also lead to a classification of “bad” translators. Some people might not want to continue if they fall under that classification even if they try to improve. We’re not here to make people quit because they think they’re bad at translating, we’re here to help them get better at what they want to do. Ranking translators through tests is practically the same, and experience isn’t always a good indicator as well. A somewhat lazy translator could do several hundred translations without ever trying to improve.

On top of the problems with rating systems and tests, there’s also the fact that the staff at MH would have to create the rating system and all the tests, enforce them, and take flak for certain decisions regarding who gets what rating. That’s a lot of overhead for something that users will decide for themselves based on just reading the translation and talking about it in the discussion threads. Also, that’s a lot of extra work for our already overworked staff. We’d probably have to bring more volunteers on to help with something like that.

This post is too long. Can we stop now?

I was thinking the same thing. The main reason for this is to point out that while a lot of the suggestions have their merits, MH implementing any of them would create a lot of extra work and bureaucracy for everyone involved, and that takes a lot of fun out of the process of translating. I would love to see proofreaders coming out of the woodworks to help translators improve the quality of their translations, but I’m not holding my breath. I was hoping to steer the conversation away from that and into a hopefully different direction. Something more community-based with translators helping each other would be a nice suggestion for a change.

NJT plans to start something more along these lines where we don't rely too much on a person or a single group of people to resolve the current issue. For starters he's running this series of news posts about translation quality to encourage people to more actively seek help (or more actively give help) on translations. It's not going to be a drastic huge immediate fix because let's face it, there really won't be any such thing, but it will be a start.

*Note to translators: If you don’t want people to proof your work then let them know. Also, just because someone doesn’t speak Japanese doesn’t mean they’re not going to be a good proofreader. I personally think someone with no knowledge of Japanese is going to be more apt to point out when something sounds weird than someone who knows the language well. And lastly, remember, this is all to help everyone here improve. There’s always room for improvement.

Have you shown your appreciation today? Click the thanks button or write your appreciation below!

0 members and 1 guests have thanked zindryr for this news
Comments
#1. by zidane ()
Posted on Apr 16, 2009
The title is a little misleading, but else that's more like the tone that's needed here, thanks zindryr.
#2. by zindryr (誰も知らない)
Posted on Apr 16, 2009
@1:

Oops, forgot to fix the title. Thanks.
#3. by Espling ()
Posted on Apr 16, 2009
I promise to help anyone I can when it comes to this problem and I hope anyone would do the same for me. This means that I agree, I guess.
#4. by Janina07 ()
Posted on Apr 16, 2009
To be honest fandom needs both the speed scanlation and the other group that scalates with care. Any red tape is sort of not needed because even if the scans are not posted here they will be posted somewhere else. Really people just need to read many scanaltions before complaining.
#5. by HisshouBuraiKen ()
Posted on Apr 16, 2009
What you've described here is basically one part of a three part problem.

1) The audience is to blame for being satisfied with whatever they get, whether it's gold or garbage, as you mentioned. As the biggest online reader, Onemanga is a huge culprit in this because they upload the first scan they find every time there's a new chapter and never update it with newer/better/fixed versions.

2) The translators themselves (assuming they are actually native english speakers) are to blame for not looking at their work once it's done and realizing how bad their dialogue is. There are numerous "translations" where it feels more like I'm reading the raw because it's done so stiffly and literally.

3) The whole "public" aspect of the subculture is to blame for putting pressure on everyone to get their stuff out ASAP. When a series has someone uploading its raws to the public, you're left with few options. You either rush your tl/scan to get it out their first and get maximum exposure, or you take your time to do a thorough job, release later or even last, and hope people still care enough about quality to give you a shot. Unfortunately the only way to resolve this would be more direct collaborations between the three corners of the triangle, which is a hassle at best and completely unfeasable at the worst.

I doubt #1 will ever change to any significant degree. #2 could be fixed, but #3 will severely hamper its progress. Let's say I PM translator X and say, "Hey X, listen, your translations are terribly unreadable in English. Here are some tips."

X may be aware and trying to work on it, he may be appreciative, or he may be a bitch who can't take criticism. But inevitably part of his response will be "If I don't get it posted right away someone else might do it and get their scan out first, and then no one will read mine."
#6. by Umiko ()
Posted on Apr 16, 2009
Should I include a line telling readers to point out my errors then?

I'm a translator, but I'm also the last to check before the release. So #2 is out of question.

#3.... I believe there are still lots of people out there who will never be able to stand those speed-scanlation and prefer to wait for a HQ version (which, unfornately, may never come)
#7. by shrimpy ()
Posted on Apr 16, 2009
Ah, there you are HBK. Wonderin' when you'd chime in...

In response to what he pointed out:

1 - yes, OM is awful. In fact, not to plug a site or anything, but I've been very impressed with http://www.mangatoshokan.com/, who update and post higher-qual versions, allowing you to choose which you'd like to read. It's pretty cool, just give the naruto section a peek and you'll see what I mean. If only we could get viewers to go to a site like that, it would effectively force OM-style readers to change.

2 - I think the vast majority of translators are too literal. But I see it as a growth process. I used to be that way too, and I've evolved.

3 - If anyone's doing this for 'internet fame' or ego boosting, they're in the wrong place. You get jack for this. And the fame is fleeting. Plus you're using a fake name! What kinda fame is that!?

The only way anything will improve is a big push from both sides, translators and proofers, and everyone has to have an open mind. There will be people on ego trips who won't want to listen. And, we still face the insurmountable apathy of readers everywhere. Even if we run a campaign to get more people involved in our community (which is essentially what's being said here), how can we convince people to chip in? The internet panders so strongly to voyeurism that we'd better have one heck of a carrot on that stick to pull them in.

I just want to say, I don't intend to be pessimistic, rather, I want to get this conversation into the issues and how to tackle them.
#8. by Nantalith ()
Posted on Apr 16, 2009
I agree with a lot of this article and the previous article but I also think translators of fiction [and I include manga in this] should have some fiction writing abilities - they should know how a story works so that they can retell the story as a story in another language rather than a translation of a story. And they should have a great vocabulary.

Umberto Eco makes mentions of translation as negotiation - a translator is not translating words, he/she is taking a cultural phenomenon and changing it, essentially forcing it, to be something else. I think translators might not have the 'right' attitude to want they are doing.

But I also think there are other factors that contribute to the quality of translation such as the age of the translator, their grasp of general knowledge and the translator's interest in the manga they are translating.

There is also the issue of readers just wanting to know what is going to happen next with no real interest in how it's portrayed because they either just want to talk about it or they are going to buy the books later.

Ultimately, unless the fans start to demand better there is not going to be much change - basic consumer principles. This is sad as I love quality translations and am quite happy to wait.

#9. by luisalirio84 ()
Posted on Apr 16, 2009
About apathy: just leave out naruto and bleach out of any system you implement and this will be a null consideration. As for fun zindyr... by "fun" you meen nethin goes lez all be friens and pasta ouar translations and stuf an u no... like do the same quality transliteration again next weak... u no... and like have leachers(fans) read ouar engrish an kizz ouar assets. Is this what you mean? Don't get it... most good translators i know like constructive criticism and don't mind not getting to jack off to comments about their awesome use of the word awesome in the latest chapter of shounenX. -_-Who is this not fun for again? oh wait you must mean ppl for whom fame=fun. Most new or learning translators I know also want help and have a desire to improve... again i fail to see the argument about this not being fun for translators... you need to correct it to Not fun for half-ass translators. neways...

Back in #translators njt kept asking me what could be improved at MH. I'm giving him my opinion/answer here. So the rest of my post is directed at him and anyone else who doesn't wish for MH to keep on trucking. >_> My 15cents:

In the previous post one scanlator commented in a way that highlights my point about scanlators being resistant.
http://mangahelpers.com/news/details/159#comment_3045

as I predicted njt ... not sure if they will be (of course i'm sure), but your biggest challenge will not be setting up a large quality control team... and it won't be getting translators to submit their translations for private review before going public.

Your biggest obstacle is the obstacle that has always existed. It is editors like Zidane21ps (and i apologize if this sounds like a bash I am not picking on you, just pointing out your attitude) These "editors" or "scanlators" will resist any attempt to improve their quality and if anyone tries to talk to them about improving their quality... this person is an a** farmer and is arrogant etc. You see there are some of us who want to improve and constantly try to improve (i play around with making mag scans look hq all the time even though i am in an hq group and have no need to do so) and then there are some of us who just want to do a slap text job and get some quick fame (zidane i'm not saying this is you - I haven't read your scans. they may be Hq, i don't know. i'm just making a point of your reluctance to embrace mh's attempt to be more than a home for crap scans)

Your argument is the same old tired one that has flown here for a long time. They're all doing this for "free" so who are we to tell them what to do. That is fine and good. But if MH decideds to have STANDARDS then you will need to 1. improve 2. post your stuff on Mediafire and give the link to MU(finding a translation elsewhere of course).

Here's what has to be done. I don't want to outright tell you what to do but here goes nothing. (feel free to ignore it but this is probably the only way that MH will make the leap forward it needs)

1. Once you get your translators (this site has many ... but it will be hard getting the majority to agree ... probably) to agree to post their translations in a private forum which will be accessible to the LARGE proofing and qc team. (you know you can exclude naruto and bleach from this if you wish and you'll barely lose any of your site activity or popularity)

2. The translations will then undergo a quality check. Some translators may find this too harsh and drop out (this is a price you may have to pay... send them to translate naruto and bleach while they grow a brain). Some translations may have to be outright rejected for being too awful (I don't know if MH rejects anyone but surely you wouldn't allow me to post translations?) A lot will have huge qc files with lots of reccomendations and suggestions which the translators who volunteer for this (can't force neone god forbid they melt) will have to take note of and fix their translation. Once a translation is approved and ready (Really for most of the translators here it won't that bad.. ok i lie.. that was my sad attempt at motivation) then it can be released to the public for scanlation(not exactly). (also consider whether new translators should be given tests to place them in a category. the categories should be simple and not insulting. Maybe just Beginner, Intermediate, Experienced) Beginners would be of a level that is un proofable and could post translations for "help" with them... i realize this will scare away all the translators that want fast fame but watcha gonna do? stay the same? ok i don't mind.

Alright you've gotten the translators on board, everyone is improving... the proofing machine is like a well oiled machine. You're pumping out translations at an amazing pace and quality(relative to now). All's well? WRONG. back to zidane21ps(sorry buddy xD). Your releases are still horrible to look at (at least for me and this applies to some not all of the stuff released here... probably the majority imo).

3. the next step is where you may just lose half the leechers (you know i mean fans). Of course you could exclude naruto and bleach from this and keep 95% of your readers. What to do about those speed edits... Well buddy i love your can do attitude but i know you can't do this ^^. You have to tell ppl like zidane21ps to close up shop and GTFO (sorry again). What you'll eventually have to do is rate the scanlators or give them a test (you can make it a test where good MQ and above pass) and if they don't pass the test send them to the training forum or to frankyhouse. Then (this is why i said that the trans don't exactly go public) only allow approved scanlators to view the awesome translations(meh) and have access to them. Some translators already only let one group use their translations so there are translators who can support this. Then later allow translators the ability to pick (in conjuction with the qc team) the best scanlator if there are multiple and work with them. (you don't have to do this but it would be amazing if you could and i'd take my hat off even though i don't wear one) Harsh, i know... but necessary... i think that anyone who isn't fly by night op. will be willing to put in the necessary work needed before they get all that glorious fame.

4. The last thing is a qc of the edits... by the time you've implemented 1-3, the survivors of the battle royale will be good enough to submit to this. members of the proofing team may also take the qcing tests and can do both. If the proofing and qcing teams are streamlined and you get everyone familiar with the system... it will be possible for the prrofing and the qcing to slow down MH's releases by 1 day only or less. And depending on the editors about another 1 day of time for better quality edits. (i'm just making assumptions here based on having run a scanlation group and been a member of another huge group- you can't hold me to any of this) OMFG 2 whole days slower... what does this mean... well it means ppl can read wsj on Sunday instead of Friday ... and it'll still be a weekly wait. The only difference is that if MH has competitors ( prob the lamers that don't want to improve) they will have a two day jump on MH .. however i'm sure lots of people would be willing to change their reading day to sunday for a much better product.

Ok some questions i know ppl will want to attack me with:

What about us translators that like to post our text translations for public adoration/comments? You translate for fame and not improvement/love of manga? GTFO Also the proofing team will be your admirer... they'll admire your work till it becomes amazing >_> This will separate the girls and boys from the men and women (that means those who are translating for good reasons from those who just like to have their e-penis/female equiv. stroked)

Won't this kill MH since franky house and manga share will beat them every week? Hmm i wonder about that. It is ultimately a decision of njt or whoever runs this place to ignore or consider my reccomendations but i suspect that since most of the translations on this planet seem to come from here they will have a hard going of it releasing stuff ... (if they choose to become what MH is now .. well again that's a decision best left to the owner of this site) but again if you exclude naruto and bleach from this experiment... you won't lose a lot of leechers and Frankyhouse and MS won't get that many MH leechers and thus won't have that large a pool to find translators. MH has the power now and they should use it while they still can.

Wait... how is this not making MH into a large scanlation group like null or Kmts? You're right... but answer me this: Isn't MH already a scanlation group in reality? the only difference is that it has 0 standards :/ so right now it's just a huge scanlation group that has mostly LQ ... MQ and a few Hq releases. and that's if i judge by ordinary, average reader standards -_-

Wait... but ... what about freedom to post and freedom to share? Why share something so awful it hurts my eyes? GTFO (you're just lazy and want quick fame and you know it)

Well that is all i have to say. Njt, I said i'd stop badmouthing MH and tell you what I thought could be improved and i've spoken my 8cents. If you do something similar to this... well... cheers ... if not... meh MH can stay a place for crap scans :x don't bother me none.

Feel free also to continue posting translation articles that only me and a few hundred other people will read and pretend that you're changing something.

Once again: njt i'm only saying here what i did not have the time or energy to say in #translators in the way I would have said it there and i apologize if I came off in an offensive way and offended you or any of your crew/people/scanlators/readers/groupies/fanbois. Also i tried not to use offensive language (i would have used a lot more on irc) but i couldn't help myself in certain places.

Why not just pm you this? Cause i want to be elitist and arrogand and snobbish and no fun at all i guess. ^^

Proof this as a test. I'm sleepy.
#10. by Hyperworm ()
Posted on Apr 16, 2009
luisalirio84's approach is one way. By effectively making MH into a massive scanlation group, MH has control over quality and it could produce good work. But it sounds far too drastic to me. I have to admit I haven't read it over slowly and carefully, though, so I'll do that in a minute. Trying to get this posted first :)

The OP post here asks those who "read the translations before they’re on speed scanlations" to "make a comment about it on the translation for the translator".
But isn't this a misdirection of effort, really? If you want to improve speed-scanlations, not only do you have to notice the translations quickly, you have to work in that short timeframe, be faster than people who have been trying for ages to be as fast as possible.
(Yeah even if we don't make it before the speed-scanlations, the translator can learn from it for next time even if nothing gets corrected, but I get the feeling we'll have a very low rate of progress that way.)
And will improving speed-scanlations really help anyway? By definition they care about speed more than quality - even if we improve translations swiftly for them, the typesetting will probably still be iffy, and things won't have been checked over carefully on their side, and so on. That isn't the kind of scanlation I want to read anyway, so... :/

And what good does fixing translations do if no-one's going to edit them in? Particularly older translations. If someone notices a mistake in a two-month-old work, is anyone going to care?

Maybe we can achieve good progress if we focus on the people who care about improving their work. Not just translators who will edit their work to corrections, but scanlators who, if they're using translators' work from MH, will promise to update their scanlations if the translation is changed.
For instance, can the scanlator be notified if a translation has been edited, at the minute? Can they partner their scanlation to an MH translation? We could give such partnered scanlations special focus, make them more visible. We could try to form a culture of fixing and updating work. This would make MH an active part of producing the scanlations, so in a way it would also make MH into a huge scanlation group, but all the work of editing the scanlations still happens within the individual scanlation groups; nothing is done inside a private, closed MH system.

That isn't to say we can't help speed-scanlations... perhaps there could be a separate rank (as Translator is a rank) for people who aren't actively translating or scanlating, but are willing to help check translations, or proof wording, etc. And a list of people who have that rank (or are actual Translators) and are online or something, and something saying what they're willing to help with. Some body of people that could quickly be turned to for advice, even in the timeframe demanded by speed-scanlations. Make it easier for the people who need help to turn to people who can provide help, rather than expecting the people who can provide help to stumble across those who need it.

Maybe I should stop suggesting ideas. Is any of this remotely sensible?
#11. by b0mb34man ()
Posted on Apr 16, 2009
I, as a scanlator, tend to proof through the translations to see if spellings are correct and see if it is confusing and what not. I sometimes contact the translator for example kewl if anything is wrong. I know that the scanlator has some control over this, but fails to do so because they want speed. My group, Sugooi Scans, try to incorporate speed, quality, and coherence. I know my post isn't as long as everyone elses, but I hope my post contributes to something. ^^
#12. by Maxy Barnard ()
Posted on Apr 16, 2009
well all i do is gripe saying that scanlators should proof their releases even if the translation's been proofed (seeing the text on the page helps context a shitload), but hey, if all the checkers and proofreaders from groups were offered to proof a particular translators releases then we'd have a nice system going. IDEALISM WINS
#13. by Sariachan ()
Posted on Apr 16, 2009
Interesting read, and I agree with the choices of you and the rest of MH staff, both what you do and what you decided to NOT do (I would hate a rating system for translators as well, for the same reason you fleshed out).

I'm not a native English speaker or I would help, anyway thanks for all the efforts and dedication you put into this site. :)

About the translation quality, I can't say much, of course, but I naturally try to read (or watch, in case of fansubs) what sounds better for the little I can understand, but an accurate, even literal when necessary translation is more important for me, after all I couldn't really say if the English used is completely correct or not anyway. ;)

But, that said, fansubs and scanlations are done for free by people who use their time and knowledge for the sake of not only themselves, but also, if not more, for other people, so I would respect and appreciate their "work" even if they used the worse English ever.

In other words: I respect and appreciate anyone efforts, but I'll naturally search for the best, and I think there is place for everyone... speed releases, MQ releases, and HQ ones... I don't really see where the problem is.

Again, maybe this is just because I'm not an English native speaker so maybe I don't, or better I cannot understand / be aware of the problem, but I also read few scanlations on top of that, and buy the official manga too, hoping that the professionals' work will be better (which isn't always right, unfortunately, but the phenomenon is worse with anime than with manga imho -I'm talking about the Italian manga/anime scenario, here-).



P.S. About literal translations VS liberal translations, my stance is this: the first are more for "active" readers who don't only want to enjoy a story, but to grasp the mechanisms of the source language too, while the second are usually better and more correct, too, but are more for "passive" readers who only care for the story.
Someone quoted Umberto Eco's opinion on the translation process, and while I would never write something against a such great writer and language professor, I think I can say safely say that most manga and anime fans aren't your average audience you translate for, but fans that usually care about the culture and the dynamics of the source language (in this case, Japanese) at least as much as the translation sounding like "an English you would really use".
#14. by zidane ()
Posted on Apr 16, 2009
Hahahaha, that was a great one, you kinda lost creditability when you mentioned that people should go to FH to learn cleaning, lmao. Seriously buddy, I got your point and I'm not angry for using me as an example for a shiity speed-scanlator which, by the way, I am not - but that's not the point here, I've been around long enough to not take this serious, especially if a person that obviously has no clue of the weekly scene is saying it. Anyway, you should re-read the post you were refering to, you seem to have misunderstood somewhat there. I wasn't, at any point, refering to scanlation quality ie. cleaning, but only translators. By no means would I ever support shitty-speed-scans which are just done for fame or stupid reasons like that... heck, that's the most stupid reason to scanlate shit for...

Point being, re-read and understand my post that I did in that other news, and well, for the future, don't use people's names for examples if you don't know them, that's just not nice :D
I'm not going to rant on how great or bad I am, it doesn't matter... but I can assure you that all we aim for is quick (relative) quality. If that includes waiting till a wednesday for hisshouburaikens translation, then we will do that, and did that just some weeks ago, with the cleaning being done much earlier already... you probably don't even know the group that I'm talking about, so I can tell you. It's binktopia, and we are some strange mixture of a speed group, though not as fast as in the past, and a quality group, but even that doesn't matter, again, my points in the other post were all concerning translators, and not even SPEED translators, but beginners, whom Pomax fucked with by saying to not release anything at all, which I just can't agree with. I mean it, people who claim their way to be the only right one don't have any creditability for me, fuck off.
#15. by zidane ()
Posted on Apr 16, 2009
Quote:
I, as a scanlator, tend to proof through the translations to see if spellings are correct and see if it is confusing and what not. I sometimes contact the translator for example kewl if anything is wrong. I know that the scanlator has some control over this, but fails to do so because they want speed. My group, Sugooi Scans, try to incorporate speed, quality, and coherence. I know my post isn't as long as everyone elses, but I hope my post contributes to something. ^^

Sorry for being an ass and pointing at someone directly...but, let's take a concrete example for once, since we came to talk about scanlations now for some weird reason. Let's talk about the effort that groups put in a scanlation - for easier comparison, here are the 2 chaps of the latest Gantz, once done by the guy I quoted, who claims to aim for speed and quality, and and the other one done by the group I'm in- may sound jerky, but heck, if I'd care.

http://www.sugooi.org/manga/?m=Gantz&c=296&p=1
http://read.mangashare.com/Gantz/chapter-296/page001.html

People with some knowledge of typesetting and cleaning (the two things one should think off when comparing these two scans) might be interested in the trans (done by cnet) and the "MQ" Raw by Heiji-sama, which was unfortunately only 6mb small, containing 2000px LQ-jpegs.
Both to be found here:
http://mangahelpers.com/m/gantz/chapters/
Both scanlations were posted on MH, if you want to download them...

Once you compared it, you'll see how different people's opinion is on quality, since both groups claim to aim for speedy quality... Just like the one before me used me as an example, sorry sugoi-scans, I'm not trying to insult you or anything, just pointing out that the understanding of quality is diffferent for everyone... what might look MQ to the actual scanlator, is cheap LQ for others. Again, I don't know your group, just saw this single Gantz chapter - nor am I going to judge you or say to stop scanlating, as I still don't see anyone here in any position to forbid doing work for free.

Stop daydreaming of quality control, there won't be any. (talking about scanlation quality here, not for translations) Why do you care to begin with? Find your groups that you trust with the quality and stick with them, do the same for the translators, good groups will also do the same and stick with a translator, as well as a RAW proviider, not using some LQ/MQ thursday raw but rather wait till the MQ/HQ friday raw etc. There are just groups that care, and others that don't, nothing will change that - YOU as the reader have the choice to decide to wait or not to wait, so whatever.

Edit: Sugoi's scan wasn't even that bad, this is more like the horrible quality if we speak about bad scans: http://mangahelpers.com/s/pi_te_ku/readonline/22069/2

That one makes one really think about forbidding LQ scans, but I'm still against it as long as everyone is free to choose what he/she reads.

My honest apologies to everyone I may have insulted, it wasn't my intention, I'm just describing the situation as I see it, and I could go much more into details about the things that are going in the wrong direction in the weekly-scanlation world, but that'd be more like worth an own article, as the purpose of this one was quite a different one. Sorry.
#16. by njt (Last Boss ♪~( ̄。 ̄))
Posted on Apr 16, 2009
Quote:
MH, will promise to update their scanlations if the translation is changed.
For instance, can the scanlator be notified if a translation has been edited, at the minute? Can they partner their scanlation to an MH translation?


Yep, translators can notify scanlators if there is a change in their translation ^^;

http://mangahelpers.com/photoplog/images/41102/1_trans_guide_7.png

see :3

That was added so translators can notify those that are using their trans (hopefully those using are subscribed >.>) so that they know there is an important change~ :3
#17. by HisshouBuraiKen ()
Posted on Apr 16, 2009
Implementing a large-scale "QC" process that translations have to survive to even see the light of day really violates the core principle MH was founded on, which was a chance for everyone to put their work out and contribute. It also runs the risk of too many cooks spoiling the broth - multiple proofreaders can easily have conflicting opinions.

One of the reasons Dattebayo works so well is that we have a streamlined process where one person is given a single job and trusted to do it well (after proving they can in fact do it, of course), while many fansubbers get bogged down with a ton of people clammering to get involved and producing a big cluster of changes and opinions that need to be sorted out (at each stage!).

No, what I think needs to happen is changing the idea of "commenting" to "critiquing." Make it so only mods and other translators can post text to someone's release and let the masses stick to the "Thanks a bunch, I really enjoyed this translation!" button. As a smaller group, it will be easier to get the producer-members into that mindset than the whole user base.
#18. by Hyperworm ()
Posted on Apr 16, 2009
@njt
That's good ._.

We could do more than merely letting the scanlator know there's a change, though - we could offer an incentive that they actually do change it.
I don't know if what I suggested would be enough - that is, MH-partnered status offering better visibility on the site, and being accepted/recognized as using a translation that a huge MH community has hammered any problems out of. And of course, the fact that they'd lose that status if they didn't keep their scanlations up-to-date. Would that be enough to make groups want to partner with MH? :|

An advantage would be incentive for helpers. If people knew for sure that their correction to a translation would get fixed in a popular scanlation in a reasonable time, no matter how old the scanlation was, then perhaps we'd get more people offering corrections. Maybe. :x

Ah, and of course translators themselves need to know when people have pointed out problems. Do they get notified whenever people post comments? If not it could be done by PM, but a system where the discussion is more in the open would work better.
HisshouBuraiKen makes a good suggestion. Or critiques could be higher-priority or displayed in a different area or something.
#19. by zindryr (誰も知らない)
Posted on Apr 16, 2009
@luisalirio84

I believe you’re missing the point of MH. We’re a platform; we help people to share their work regardless of quality. It isn’t our place to say who is skilled enough to translate on our site. We are not a group of elitists. The most we can honestly do is to encourage people to improve. Everything you are talking about adding goes against practically everything that MH is. I’m not opposed to trying to get a more active group of people to proofread and help translators. Far from it. As I stated above proofreading gives translators invaluable information that they can use in their work in the future. The more times a translator is edited for the same mistake, the less likely they are to make that mistake again. Experience.

MH is not here to discourage people from doing what they love to do. We also cannot force the audience to do anything. We can only try to encourage the people producing for the audience to give them higher quality work.

As for your suggestions, I ask here like I asked in the above post, where are all of these skilled proofreaders going to come from to fill in this large proofing and QC team? What if the people in this large QC team start dropping out and increasing the release time between when the translation is posted and when it finally makes it through this QC phase and out to the public? Why do we want people to stop translating, especially if they’re the only ones working on a series? Who says beginner translators are just in something for the fame? Isn’t one of the best ways to get advice by publishing something to the public at large?

Saying that your way is the only way to improve really does not sit well with me. I can see it lowering the productivity of the site, creating an elitist attitude among those still around, angering readers who have lost the only people doing a particular series, and turning our hobby into more of a job than it should be. Something more productive would be to have translators help each other, some proofreaders show up and help with that process, and gradually raise translation quality over time.

Also, I’m not going to talk about scanlation quality. That doesn’t belong in this conversation.
#20. by d4v1d_su ()
Posted on Apr 16, 2009
German people say "lange Rede , kurzer Sinn"
What's the plan ?
Let's not judge whether it's a bad idea or not, or that the idea is very difficult to realise. First of all, what are the ideas ?
I like the idea that everyone should take part of this and post something if they find some mistakes or something they don't understand. Translators should also be active as well, like asking the scanlation groups about what they think about the translations.
And I want to confirm one thing. Are we trying to kick out "bad" translators or are we trying to make them better ?

And I also hope that the fight or dispute or argumentation or whatever you call that is over now. Besides, it's already off topic.
We are talking about translation and not scanlation.
#21. by AscE ()
Posted on Apr 16, 2009
To luisalirio84:
"You have to tell ppl like zidane21ps to close up shop and GTFO (sorry again). What you'll eventually have to do is rate the scanlators or give them a test (you can make it a test where good MQ and above pass) and if they don't pass the test send them to the training forum or to frankyhouse."

I recommend, in the future, you actually investigate who the person you are addressing is so you avoid embarrassing yourself by making false accusations.

This isn't about reluctance to improve people's translation skills or to improve the quality of the translations. It's about the impossibility of both maintaining a rigid criteria for quality translations, as well as maintaining the freedom for anybody to post his/her work. You're suggesting that some translators are downright rejected for their "unsatisfactory" work and an army of QCers be around to "improve" all the other translations. Not only would this slow down the scanlation process as a whole CONSIDERABLY, but it would also cause many upstarting MH translators to simply leave. MH will become an elitist site where 10 translators and 500 QCers post their work. I'm fairly sure that is quite the opposite of what MH was built to be.

Furthermore, the notion of QCing each translation before they go public would not stop the release of poorly translated scanlations being released. Going through MH to release a chapter is not a requirement. While many, if not most translators DO post their work on MH to make it public, scanlation teams who only care about speed will inevitably go around the system and you'll see a crap release being posted anyway. Unless of course you delete all these releases in which then MH would lack both translators and scanlators.

You base the fact that all will be will if you slow down scanlations by two days because the majority of people will wait for the higher quality scanlation and unfortunately, you are wrong. The existence of shitty scanlations proves that these scanlations have support from the fanbase, thus proving that quality is not the main concern of many fans.
#22. by shrimpy ()
Posted on Apr 16, 2009
HBK hit the nail on the head: we need constructive criticism.
The question still stands how we're going to motivate the few of us doing this to read over each other's translations. I for one have found RL demands completely destroying the time I once had, and I anticipate it's going to get much worse. We'll just have to settle for what each person can contribute, no matter how little that may be.

Also, question to HBK - you work professionally, correct? Can you comment on what the commercial model is to getting out the finished product? Do they, as you said about DB, only have 1 person at each stage?
#23. by HisshouBuraiKen ()
Posted on Apr 16, 2009
@shrimpy - They do in my (limited) experience. You may have someone who wears multiple hats (e.g. timer/editor), or multiple people doing the same job (translators, encoders), but you never have two people working simultaneously on the same stage of any single episode.
#24. by Zakk ()
Posted on Apr 16, 2009
I am not a scanlator/translator, however i would say that as many people have pointed out Quality control would be hard to do and kicking out bad scanlators impractical, maybe you could try promoting HQ scans with banners and threads in more obvious areas over the downloads page, for example have a special section nearer the top for groups that release HQ and then have LQ ones lower down.
#25. by luisalirio84 ()
Posted on Apr 16, 2009
uh... how did I embarrass myself I believe I stated I didn't know what his quality was nor do I have the time to donwload some stupid shounen manga to find out. I was refferring mostly to his attitude about any system that would not allow editors that didn't want to improve. First of all look. I get that this is a place where 95 % of the people are narutards and bleachedbrains... but... I was more adressing njt since he asked for my input.

This was because we had a discussion in an irc channel in which I told him that manga helpers would never be more than a place where quality goes to die. He was adamant about improving and how they were trying to improve. He kept asking me what I would do to make it better and I said it was impossible. Now, I didn't say that out of hatred or pure evil (maybe a little) but I said it because as HBK said ... the changes I talked about go against the core principles/values of MH. So imo I already knew then that trying to control the quality of the products that MH cranks out would be impossible. But njt wanted a reason and he wanted suggestions on how to improve. So i tackled both problems with that one post... I bluntly laid out what i thought would have to be done in order to seriously improve MH ... (if he likes MH the way it is he wouldn't have asked me all that) and then as I knew would happen most ppl did the expected action. They rejected it for the same reasons I knew they would reject it. And njt can now fully understand why I said it is impossible.

You can all play around with your small ideas about critiques... but HBK are you going to compel say... zidane21ps to incorporate changes made by a translator after "critique"? You all know very well that most scanlators will want to typeset as soon as a translation is posted. That brings us to the translators of this site... there are exceptions... but are you really telling me that you're going to compel "weeaboo translator of the day x" to take your criticism and try to improve? of course not. that translator will most likely do whatever they want and ignore or just give you the royal run around. In the end... the translators who already are improving on their own will maybe improve a little faster but the vast majority of translators here will continue with the curent MO.

In the end, MH will remain a site with a few good translators and a majority of butchered scripts with terrible expression and flow. As for the quality of the scans :/ comeon? you haven't even dared to talk about qcing the editors' work. >.> Few good groups... but for the majority of them, i will remain incapable of looking at their scans without wanting to gouge my eyes out.

ok... too many posts to read i'll comment on some necessary ones:

@zidan21ps.. Uh... zidane21ps... you weren't so nice to sugooi scans yourself. And you're right! I don't know you nor am i nice to people who I don't know... it's too time consuming. Also if you are such an awesome scanlator why the reluctance? A little bit of qc should be no problem for you? neways ... let's not make it any more personal. -.-'

Where did I say you should go to Frank House to learn cleaning? i don't even know if they have a trainging area... could I really have said that.. i can't seem to find the place. You've been around long? I remember your first scanlation... though i didn't waste my time reading it. You're right again! I don't know the weekly scene... why would i waste my time with some crappy shounen mangas? =_= you may not have been referring to editing but your general attitude seemed to say "i'll do what i like" which is fine and which is why I used you for an example of a scanaltor that would resist change. If this is not you... well i apologize but i'm not gonna reread yuor old post since your reply was quite sufficient. Don't worry i'll fuck off after this post. I was not dictating anything to you as gospel it was my advice which was requested and which i gave albeit knowing full well that said advice could never be followed. So yeah.. it wasn't really meant for you. Wow you still don't understand that the reason there are so many versions of the WSJs every week is because of fame and glory and e-penis? neways wasted too much time on you already.

@zindyr yup I know that MH is a platform that allows ppl to post anything that sorta looks like it might be a translation and anything that somehow did something with the raw and the translation text. And I knew that would never change therefore MH will always be what it is now... craptastic. I know it'll slow speed and that's what you care about... which is what ticked me off with njt's original comments... if you're a home for crappy speed scans accept it and take your ribbing properly. I don't see how it turns your hobby into a job? most ppl in the established groups do it as a hobby too... You act as if you're being held at gunpoint by the readers... nvm i agree with you .. you are. Lastly for every one good translator you can find about 4 good proofreaders ... they're seriously everywhere...

neways your post as well as AsCe's post prove another point... the people here don't want to change. they like things the way they are. They like it that anyone can post their trans no matter how awful >.> They love that the scans come out super fast and super nasty. They love that there's a legion of rabid fanbois ready to pass bleach and the rest of them through neat image and slap on some text as fast as possible. You guys love this. I do not.

In any case.. njt sorry for the trouble I caused... It's my nature... i'm not gonna comment anymore.
#26. by zidane ()
Posted on Apr 16, 2009
If you...say so?

But I still don't see how you make MH responsible for "supporting LQ crap", since this is just a neutral place to submit stuff, it's everyone's personal decision on what to download / read / comment upon, so where exactly is the problem? MH isn't a scanlation group, it's just a place to release things for scanlation groups, controlling that is simply impossible. As "hard" as it may sound, people will have to look for the good stuff themselves here, unlike in Mangatraders, Null or Illuminati-manga, etc. where one can always expect high quality. This website just serves a different purpose, and I don't see how it's wrong. Also, there are some solid "institutions" like cnet, hisshou, or molokidan, just to name some, translators that I fully trust and for whom simply checking for grammatical mistakes is most often enough...

What's wrong with shounen btw... dare you to bash One Piece :/ oh, and how about we name the series that have "bad" translators?

Point being, most serious groups have an own site to release things, and MH is merely the place to announce their scanlations to a bigger audience, just like MU - though lately MH went on a new road with storing DDL's and OV's, which I don't really support, well actually I don't care, since I won't use these services anyway. Yet, a site that's for announcing new releases can't be taken responsible, it can only offer help for those who are interested in it, and I'm exclusively talking about translators here... support shouldn't be forced on anyone though. The "my way is the only right way" idea is too often used by fundamental religious fanatics already, let's try to not spread this plague here.
#27. by pangoo ()
Posted on Apr 16, 2009
I'm glad that MH are trying to improve the quality of the translations. I would like to point out that the quality of the translations is REALLY important, especially concerning English translations. This is because they are often used to translate to other languages.

Another important point is that people actually talk about the chapters later on. And if the translation is wrong, they the whole discussion is based on one person's misunderstanding.

Therefore I believe that anyone who translated has a duty to proof read their work. After all they are translating for others.
#28. by Xaeo ()
Posted on Apr 16, 2009
I'm gonna go out on a limb here - would it completely unrealistic of me to suggest some kind of tutor group where existing skilled translators organise some kind of 1 day a week councelling session for newer translators to bring their work along and get support? Now I understand the implications of this - already busy translators have to donate even more of their free time, and beyond that both the tutors and the tutee's have to have a mutual respect and understanding, but there are many benefits. The system would build on quality, strengthen bonds between scanlation groups and provide people with a pool of known talent.

Please note, I'm in no way demeaning the work that is going on, or saying the skilled members of this community arn't doing enough - I think the work some of you guys take on approaches a full time job. I do, however, think that if it became popular it would be a nice social event which would help everyone grow
#29. by r@z ()
Posted on Apr 17, 2009
I spot these errors quite often, and I find them to be very bothersome. I think that the best way to get rid of them is to add (in the very spirit the MH community was founded upon) more helpers. Raw providers provide raws, translators translate from one language to another, and scanlators put the translated text onto the raw. Editing the raws (or cleaning), editing the translated text and typesetting are all left to the scanlators, and out of these three processes it's only natural for the least important (editing the translations) to be the most neglected.

MangaHelpers offers a great decentralized and distributed open-framework for the manga-releasing process -- but it stops just short of providing much help for the other intermediary manga-releasing processes (like allowing Joe sixpack who's capable of editing written English an outlet to help.) I'd be glad to edit translations whenever, and I'm sure many others would be also. There's just no way to do so now...
#30. by b0mb34man ()
Posted on Apr 17, 2009
@Zidane21ps
I was not offended because I clearly knew what quality I had when I had released it (mainly for speed purposes like you said). Since it was Cnet's translations, there is little or no proofing necessary. Not to mention, the chapter had almost few words which I could easily proof while doing it on PS. Normally, I would copy/paste onto Word and read through the whole thing and edit the text or contact the translator to see if this is the correct meaning. Also, in your Gantz release (Binktopia) I clearly stated I was put to shame when you had released the chapter. ^^ On a side note, this thread is clearly about translations and not quality of work in terms of cleaning.

Quote"Edit: Sugoi's scan wasn't even that bad, this is more like the horrible quality if we speak about bad scans: http://mangahelpers.com/s/pi_te_ku/readonline/22069/2 "

That was amusing for me because I too thought the same thing.....

#31. by Elkin ()
Posted on Apr 17, 2009
I notice some scanlation quality discussions above... but for the purposes of this article I'll stick to the translation-side of things.

Realistically, I don't think we can ever implement a translation grading system beyond the 'translator' and 'translator-in-training' (TT) one we've got at the moment. For a whole variety of reasons that you can see in the comments above. Of course, whether I agree with all the reasons is another matter... Anyway, moving on.

To be frank, the 'approval' system for translations by TT is rather broken, and I'm not talking in terms of English quality. I'm not even sure what the approval system is there for at the moment. Not pointing fingers at anyone, I'm finding a whole lot of problems with translation's interpretations of Japanese. One particularly horrific one even had dialogue spun out of the air. Yes, nothing whatsoever to do with the text in the raws. And these get approved anyway and circulate into scans *sigh* So, why not just use the existing system and improve upon it?

Anyway, what would be helpful is to have a button/option for requesting for a critique, which will then post their translations to a critique forum those willing to help can browse. This should be particularly useful for TTs, because frankly, I don't even know what forum their translations go into. Nor do I have the patience to click through every translation to see if they have a line requesting for critiques.

Personally, I'd appreciate it if TTs had a 1-2 day lag between posting their translations and their translations becoming available for scan use. This gives translators at least some time to check if there are any glaring problems and make suggestions (if they want to). I would go further to suggest that critiquers be able to grade the quality of the translation according to how many % of lines have been interpreted correctly.

Likewise, allowing translators to request for proofs would be immensely useful as well, perhaps allowing them to choose the time frame they're willing to wait a proofer before the translation gets posted, proofed or otherwise. This would probably be a great deal more efficient than the current proofer request thread, since anyone who's interested can just pop into the forum and proof whatever they feel like.

Of course, this entails creating two new categories, as well as have an automated system of status changes... Anyway, just my two pence about the best compromise between all options.

EDIT: as a side-note, I can't read the comments at the very bottom for the earlier translation quality article. Maybe we should split the comments into 2 pages?
#32. by Ex-Shadow ()
Posted on Apr 17, 2009
Maybe I should say that people will appreciate a speed scanlation with good English language, with grammar and etc. Will proofreading team make slow scanlation or not? We must try it first.....
#33. by serizawa (ならぬことはならぬ)
Posted on Apr 17, 2009
What HisshouBuraiKen (HBK) proposed is very similar to what I proposed in the previous thread on this subject: have a smaller group of competent and trustworthy people to handle translation QC.

I agree that establishing a rating system may make people offended/sad/etc and leave MH if they get a poor rating. Well, if the goal is to provide quality translations only, then perhaps such filter ("you're bad and don't want to improve? Ok, feel free to leave") will be necessary, though it will sound a bit elitist.

However, if MH is aiming at having all kinds of translators and at the same time improve the quality of the translations, then perhaps a suggestion could be a "certification system", so to speak.

In other words, have a "MH seal of quality" for good translators who show that their work is reliable (not necessarily perfect or outstanding, but always keeping a good level).

So, MH could let all translators/scanlators, both good and bad, to post their work at MH as usual, but a translator/scanlator may aim at getting higher credit in the site by submitting their work for evaluation, in order to get the "seal of quality".

The evaluation could be carried by that smaller group HBK and I have proposed. There is an interesting detail here: the group would only have to check the works of the translators who apply for the "seal of quality". This means less work for the group's members.

Not having the seal wouldn't be any negative status, since there will be good translators who won't go through the trouble of getting that seal. Speed scanlators won't bother to get it too, if speed is what they prize the most.

Nor will the seal be a proof that the translator/scanlator is the best. But it will be an evidence of trustworthiness.

A newcomer, for instance, would have a clearer idea of whose translation he/she may trust by checking if the scanlator/translator has that "seal of quality".

So, we would still have an ocean of translations, both good and bad. But a small portion of them would carry the "seal of quality", meaning that they can be relied upon safely. For those who are acquainted, the general concept would be similar to ISO certification - in this case, MH certification for manga translation.

Obviously, we cannot forget that the improvement (education) of translators should be stimulated by creating a convenient infrastructure (forums etc.) where one can get constructive criticism.

But I would like to hear people's opinion on this.


@HBK and @shrimpy:

I work as a translator for a manga publisher as well. In my case, there is one single person assigned to one single task too. But it ends up with several people who proofread the translations, making it very prone to introducing errors and/or inconsistencies, unfortunately.
#34. by death13a ()
Posted on Apr 17, 2009
How about allow people to post translations' revisions. So when someone posts more or less direct translations from Japanese to English, another person who doesn't speak/read Japanese but has proper English grammar can post their best shot of revised translations. If translator approves of revision, then when someone comes to see translations they see approved revision. If revision is not approved then user needs to click link to go see all "unapproved" revisions for that translation.

This system just adds extra posts (i don't see anything new has to be invented for this) and least insulting to translators ( through they do need other to allow to modify their work)
#35. by zerocharisma ()
Posted on Apr 17, 2009
I wanted to add support to Mariachan's comments, since this is my situation and why I enjoy MangaHelpers so much.

I can now identify myself as an "active" reader and a connoiseur not only of the culture of Japan and the dynamics of the language, but also of you people, and of the dynamics of the international manga subculture.

As an author of speculative fiction, I often find myself using incidents I observe here, and insights I derive from trends I see happening with you its participants to illustrate points I make to fellow authors about how to identify and foster a vibrant artistic milieu. (I'm completely serious here, folks.)

I also agree with the saying "one who does not know another language does not know his own". Hey, and I have to write in this language, so at least watching the translation process helps me to understand English better.

Just keep in mind that all of these problems you're debating in this thread are the same things that define a lively, growing art form. As you prescribe strict behavior and belief codes, the art and the culture surrounding it starts to choke and become moribund. Like American literary fiction. Bleh.

Just a thought...

z.






#36. by fxu ()
Posted on Apr 17, 2009
My Science, some of these comments appeared to have been written by a six year-old.

You all talk about QC and editing (grammar mistakes), but please read your own comment before you hit "Submit"...
#37. by Dark-san ()
Posted on Apr 17, 2009
It makes a little no sense here. Whether if it is scanlation or translation, it is just something we all do to pass the time. It is after all meant to contribute a little back into the manga scanlation community. For short, it is just pastime. Why so serious?

I have been kept amused by the recommendations put up here. From setting up a team of proofreaders to sample the raw translations first before releasing it into the public and to introducing a rating system that promotes elitism. We are jolly well know that Mangahelpers is not meant for this purpose neither nor they have the resources or the capacity to do so. So why aim for something that big and yet no achievable?

The only sane suggestion from someone (I can't be bother scrolling down and look for his name -.-; ) I see here is for the administrator and the rest here to adopt a wait and see approach. Get a reliable proofreader and QCer, make them come together and do up a powerpoint slide that contain critical guidelines on how to improve their translations and release it to all the translators here.

Yes, it is small scale but it proves effective in some cases. Besides no matter how busy a translator gets, it wouldn't hurt his schedule or his pride just to read some slides. I meant hey I am a translator myself, and I won't mind doing that.

If that wouldn't help to improve the situation, then maybe Mangahelpers should start doing something drastic.
#38. by owlish ()
Posted on Apr 17, 2009
It's pretty cool to see so many people speaking/writing so passionately on the subject, and coming up with a wide variety of suggestions. Here's my idea: incorporate the wiki concept and allow people to collectively improve the translation.

- The setup: A translator is open to the idea of having other members critique and edit his/her translations. The translator opens up a group that will collaborate on translating a specific manga. Anyone is allowed to join the group, and members can sign-up to participate in it with a single click (similar to how they add a manga to their Favorites list.) The translator who created this little group is in a mod-like position in the group.

- How it works:
Each time a new raw is released, a wiki-like sandbox page for its translation is automatically generated. Anyone who are interested in participating (aka, the people who signed up for the group) can make alterations to the translation, anything from correcting glaring mistakes to editing typos. (However, contributors are discouraged from making the less meaningful cosmetic, stylistic alterations that are a matter of preference.) Then, the designated leader-translator (the one who created the group) makes all the final touches and releases the translation to the general public. If errors/typos are discovered post-release, members can notify the translator to correct the mistake (same as how it is now.)


-Possible benefits:
1) How it can help translators: The final decision is still yours, but you can get a lot of input from the work of other people. You may even have much of your work taken care of, if the contributors translate a bulk of the text before you even get started. :3
2) How it can help people who want to help out: You can take part in the translation process even if you are lacking in time, Japanese skills, or even English skills. Skilled but busy translators can translate a few lines on the go, without feeling the pressure of having to translate the entire chapter when real life calls. Junior translators or students of the Japanese language can tackle lines that they feel comfortable with, knowing that the mistakes, if any, will be caught by people with more experience. They can build up skills and confidence without agonizing over the portions that they are not skilled enough to translate. The champions of the English language can exterminate the typos and grammatical errors that have been overlooked by the other contributors.
3) How it can help readers: Many of the less popular manga have few, if any, translators who can keep up with the raws every week. Having the contributors wiki-ing the translation can hopefully speed up the process if the contributors can at least take care of the the simpler pages (where all they do is shout "Wait!" "Stop!" "No!" "Agggghhhh!" ) that requires very little Japanese proficiency, leaving the main Translator with much less manual labor. Also, translations May have better quality, since more minds worked on creating the translation, but the speed of the release will not be much slower since it is the same old Translator who is releasing the translation when they are able to.
4) Best of all: No hard feelings about pointing out errors- you don't need to feel bad about a silly typo, or create drama by pointing out a disputable line : anyone who sees a problem can fix it, and if that's still a problem, fix it again, and again, until the lead Translator finalizes and releases the translation. Don't like the final touches of that translator? Start or join another translation.

- Other notes:
1) Because wikis generally keep a log of the changes that have been made, and also because The Translator does a final check, pranks and sabotage to the translation-in-progress can be avoided. (The log can be deleted a few weeks after the translation's release.)
2) Setting up a group for the sandbox and having a lead Translator for the group allows for multiple translations for the same chapter. Also, translators who feel fine with their current process can keep doing things their way without being forced into participating.
3) If necessary, incentives can be created. For example, active contributors can be designated "translator assistant", Santa's little helpers, quality checker, proofreader, etc. etc. Tally up the number of chapters that they contributed to that can be easily displayed, require that all future translators need to contribute in other translators' translations before they get their full license, etc. etc.


That's just what I thought might be a useful system to have around. I came up with the idea because I am not fluent enough to translate whole chapters without spending a ridiculous amount of time, but "sometimes" fluent enough to understand and translate small segments more smoothly than a skilled, but busy and tired translator... And also because a wiki format would be easier/quicker to correct typos than leaving a comment to point out the error (and coming across as snobby/elitist/unappreciative), and making the translator feel bad. :(

Where would the army of contributors (occupied translators, semi-translators, and proofreaders) come from? Well, if there is no contributor, this is exactly the same as a translator translating. But if there is at least a single person willing to chip in a little bit, such a set-up would enable them to help. If there are serious flaws to this idea, I'd love to hear it. Thanks for reading, and thanks for creating a discussion like this. :)
#39. by cnet128 (MH's Best Translator)
Posted on Apr 17, 2009
Personally, I think a huge part of this equation is distribution. The issue isn't so much that people aren't willing to write high-quality translations, or to improve their existing translations... the more fundamental issue is that they have to little motivation to do so, since so few scanlations will use these new versions. And those scanlators have little motivation to use the new versions because so few people will read their new scanlations if they do.

Okay, I'll come out and say it: The problem lies with OneManga.

Now, MangaHelpers is well on its way to being a powerful distribution system in its own right. It's a popular site, with a large archive of scanlations for a ton of series. It has an excellent read-online system and a user-friendly interface for locating the scanlations a reader is looking for. But at the same time, it's still too complex to take the place of OneManga as the one-stop archive-trawl-latest-chapters-shop for the average leecher.

So what I think would be a beneficial change in the long run (though I don't know how feasible) is to go that extra mile and make OneManga obsolete. To make sure that every series has read-online scanlations available for every chapter, and to make a new idiotproof interface that caters to the average leecher in the same way that OneManga does. Obviously, this interface wouldn't be a replacement for the interface MangaHelpers has right now, but rather a companion to it. Like a companion site linked directly into the MangaHelpers backend. For argument's sake, let's call it "MangaReaders".

If a user went to this hypothetical "MangaReaders" site, he would be presented with a bare-bones, user-friendly, let's-get-started interface. A list of the latest chapters to be released, sitting below a prominent "choose-a-manga" segment similar in functionality to the Browse Manga page we have here today. Just start typing in a name, and the list of manga magically narrows itself down. Each row displays simply the name of the manga, and three buttons: "view downloads", "read from beginning", and "read latest". Clicking the latter two will take the reader to a page similar to the read-online pages we have on MangaHelpers right now, only with more options. With the option to skip directly between chapters of the series. With a mini-widget sitting out-of-the-way allowing the reader to pick a different manga. With that alone, MangaReaders would be a site with all the functionality and ease-of-use of OneManga; all you'd have to do is spread the word and draw the users in.

And drawing the users in shouldn't be too hard, because you can provide more functionality than OneManga has ever had. Examples? Add a "download this chapter" button on each page, tied into the MangaHelpers download system. Boom: Instant useful functionality that OneManga doesn't provide. And of course there's a downloads page for each series for those people who don't care for online viewing at all, and just want quick-and-easy downloads for all the chapters of a series in order. Want more? Add a button, prominently visible on each page, reading "add to favourites". If the user goes through a quick registration process, he can have a quick-access list of the latest chapters of all his favourite manga right at the top of the MangaReaders homepage. And what's more, this registration and favourites list carries over to MangaHelpers if the user ever visits that site, and vice-versa. Need anything else? Include a comment system for each chapter, that appears at the bottom of the page as you're reading it. (This could be tied into the MH comment system for scanlations, or just an entirely separate set of comments for the chapter as a whole). Still not enough? Add a drop-down menu, allowing the user to select between different scanlations for the same chapter (naturally, with the translation used in each indicated). If the reader doesn't like the scanlation he's looking at, an alternative can be only a click away.

But he probably will like the scanlation he's looking at, because the whole system is tied in to a rating system implemented back on MangaHelpers. Scanlations are rated by MH users, and the chapter with the highest rating is the one that appears by default for that chapter on MangaReaders. And this is the important part, because it means that the translators and scanlators have motivation to create high-quality chapters – because not only will their chapters automatically be available to the hordes of leechers browsing MangaReaders as soon as they're posted, if they're highly rated, they'll be automatically stored for posterity as the default scanlation for that chapter for anybody going through the MangaReaders archives! And with archives of this quality, and with all that functionality (which is basically just existing MangaHelpers functionality given an idiot-proof, minimal-click interface), why would anybody go anywhere else?

There's really no end to the functionality that could be included in a site such as this. You could take advantage of the chapter title fields that are built into the MH translation system, and have the chapter titles (where available) visible alongside the numbers in the "select-a-chapter" menu on MR – now that's what I call ease of navigation! You could include easy access to the MH reviews for a series. Any small selection of the features I've mentioned would serve to make the site significantly better than OneManga whilst retaining all of the user-friendliness that makes OneManga so popular.

...Just a thought ^^
#40. by emer ()
Posted on Apr 17, 2009
Is dat sum drama on my mangahelpers, oh dear...

Seriously, lots of talk, but will there be any action(s)?

I'm glad that I'm not translating : )
#41. by Nihil ()
Posted on Apr 18, 2009
@cnet: I do like the idea of a multiple-versioned scanlation reader, but simply creating an online manga reader with tons of extra functionality isn't going to overtake OneManga. There are already manga readers with favorites abilities or with download abilities, and the reason why those remain dominated by OneManga is pretty simple -- the average manga reader doesn't want to download chapters, or track their favorites, or anything like that. They mainly just want their Naruto fix, I suppose.

But the primary reason why OneManga is on top, I think, lies with three reasons: first, it has a simplistic interface aimed towards these casual readers; secondly, it has a limit on what releases are posted, so that it's easy to catch all the updates just by regularly visiting the site; and thirdly, most importantly and where I think is the crucial area, it's fast.

By fast, I'm referring to how fast it loads the pages. I've seen people request series to be added to OM that were on MangaFox just because MF was too slow. OM occasionally chokes up -- especially Thursday nights =P -- but for the most part, it loads pages as fast as you can read them.

If MangaHelpers wants to build a new manga reader to kill OneManga, it will have to go much, much faster than Read Online currently does. Now, I know that a large part of the reason OM can serve images so quickly is because of how much it compresses them (which is why Read Online at MH is a nicer alternative, if I'm not in a hurry), but the average reader doesn't care about a bit of JPEG compression.

However, to improve the speed of loading images, it unfortunately mainly comes down to a matter of cost. More servers, faster connections, etc. However, there are some other ways to increase speed. It occurred to me, the other night, that very few online manga-readers bother to prefetch the next page -- that is, load the next page while you're reading the current one. Seems to me that it's an obvious improvement, when spatial locality is virtually guaranteed. Of course, to implement this, you'd need an AJAX-based reader, but a degraded version could always be provided for Javascript-less visitors.


This seems largely off-topic, so, I'll end with a comment my translator made to me. It would make most sense, really, to put this control for proofreading in the hands of the translators -- why not add an option to mark any released translation with tags like "do not proofread," "needs proofreading before scanlation," "would appreciate help with my Japanese," "would appreciate help with my English," et cetera.
#42. by ruggia ()
Posted on Apr 18, 2009
.... I think the problem is slightly exaggerated....
The translation these days is pretty good, and except for few exceptions, I think the fan translation is mostly better than the official ones...
#43. by Arhazivory ()
Posted on Apr 18, 2009
Took me a while to read through everything, but after reading I really appreciate what MH is trying to do. If I had an issue with a translation, in terms of grammar or anything else, I'd opt to PM the translator about it and find out if there is an alternate way a particular sentence can be phrased to provide more clarity. For small grammatical errors, like a misspelled word or an incorrectly placed apostrophe...the TS process would take care of that.

I join with those who do not agree to making any and everybody able to edit a trans or start a BIG proof group. I think the scanlators are in the best position to directly talk to a translator about his/her script, because said script is (or is supposed to be) carefully examined in the TS process.
#44. by fxu ()
Posted on Apr 18, 2009
"It occurred to me, the other night, that very few online manga-readers bother to prefetch the next page -- that is, load the next page while you're reading the current one. Seems to me that it's an obvious improvement, when spatial locality is virtually guaranteed. Of course, to implement this, you'd need an AJAX-based reader, but a degraded version could always be provided for Javascript-less visitors."

@Nihil: Mangashare's reader has that. It caches the next page.
#45. by NymphStealer ()
Posted on Apr 18, 2009
gonna try to put my thoughts out coherently on translations, but im not a very good wordsmith like some are...
reading the latest articles concerning translations, helping improve translations, and basic calling for native english proofreaders to help out in this process..... just one thing i noticed is not mentioned which i thought of when reading cnet's guidelines for using his tranlations...
which english are ya looking for?
im native english speaker but imma american english speaker w/ southern dialect. which is a bit different than what one find on the west or east coast of the states...
if lookin for by the text-book english...
whose english?
british english, american english, and whatever else is gonna be different in some ways... like spellings and how we word things to get points acrossed... what might sound good to me or nother american english speaker might sound completely off to those who speak british english...
dunno how big of a deal this is to some, but thoughtd i mention it since there are so many variations to english... if looking for proofreaders it might be good to know which variation to use when proofing translations...

and yeah i know all sorts of grammar mistakes littered through my post... lol... but i dont have to be perfect on the internet... XD
#46. by njt (Last Boss ♪~( ̄。 ̄))
Posted on Apr 18, 2009
@42 & 45

MH does too :p

Anyway, I'll be writing up the next post for this here soon >.> (been out drinking, b-day and all ._____.)
#47. by Nihil ()
Posted on Apr 18, 2009
@njt: Ah, I didn't notice that. Then it would seem that "beating" OM in the casual manga reader "market" would come down to a matter of resources and money, which doesn't seem like a fun game to play. I vote MH stays like it is, with a little more emphasis on improvement and quality ;)
#48. by kouros_9 ()
Posted on Apr 18, 2009
A rating system would be a good thing. In fact a lot of french scanlation groups that use english translations, don't translate anymore according to the released translation but according to a translator. And bad translators are well known as well as very good translators. However it is your current system which is at fault. Because it is a star system for translators too, even if your organizational system may be good. I could see that some translators released a lot of translations at the same time (sometimes it is 6 or 7) in order to be seen and read. And we can't deny that creates a competition between some "english" translators. To well-known (we could tell famous or star) translators like cnet it is not a problem because everybody acknowledges their skills, but...
#49. by Kioras ()
Posted on Apr 18, 2009
Really, the best thing is often, to do nothing at all. Having a rating system on releases would not be too bad... But I feel that actually going ahead and decreeing something on high is often a negative.
#50. by proscientia ()
Posted on Apr 19, 2009
I am pessimistic about what MH will be able to accomplish in terms of improving scanlation -- and possibly also commercial translations --, but I think it is a good idea to try.

I think that a good start is to encourage native English speakers to give comments about the quality of language for various translations and others with good Japanese knowledge to give input as to how well the translation conserves the salient features of the original text. I liked the idea of a translator being able to inform others with information box asking for editorial input. Unless it commonly occurs that there are too many comments for a translator read, I think that the suggestion of limiting comments to people with special status on MH is unnecessary.

Although, I prefer that MH stays the kind of site it is now -- where any member is allowed (and encouraged) to contribute to the site --, I will not mind any extra features that do not limit the freedom of the members. I also do not mind a ranking system being implemented for translations and scanlations. I think that the ranking should be limited to members, so that no one can vote twice. If something like cnet's idea of a manga reader is implemented, I think that a Bayesian average scheme -- or another method to ensure that a translation with a high score and few, possibly biased votes does not replace a good translation with a lower average score -- is used. On a side note, I might attempt some translations with a friend in the future and I do not mind being told that my work is garbage if it is really bad. Getting the manpower to implement some of the schemes mentioned below might be difficult, though.
#51. by llamapie ()
Posted on Apr 19, 2009
WEEEEEEEEEEEEEEEEEEELLLLLLLLLL It would be nice if some of the speed groups actually read the mangas they're scanlating. "Sage tech!" .. so damn annoying.
#52. by chibidiono ()
Posted on Apr 20, 2009
I guess I'm just glad I can read the RAWs then
#53. by noob3d ()
Posted on Apr 20, 2009
So the speed scans I've been reading at onemanga were not really in english?
Strictly speaking some things just can't be turned into english when going from japanese to english without altering the meaning to a certain degree.I think most of the problem lies in the translator being reluctant to be more flexible with the language...either that or he just doesn't know any better.I think its unfair to place the blame on the translator when its a speed scan.The proofreader should have said something about the unorthodox use of language instead of giving the OK even if it means delaying the release by an hour or 2.
#54. by MagicWookie ()
Posted on Apr 20, 2009
It might be easier to simply have a "thumbs up" or "thumbs down" button to rate a translation, or even the translator, and leave it up to the translator to find out how he/she needs to improve. I think it will still help those who are actually wanting to improve and weed out those who proofers would be wasting their time with.
#55. by Hyperworm ()
Posted on Apr 20, 2009
@cnet, MagicWookie, maybe others
Wasn't the idea of rating translations/scanlations (by a team of an unspecified unbounded number of knowledgeable members) shot down in the previous thread, on the grounds that people would complain about being given a low rating and leave? :S

I certainly wouldn't trust Anonymous to pick out the best translation/scanlation. The vast majority are completely unable to judge accurate translation, so the opinions of people who know Japanese would be drowned out totally. Also, I don't think enough people care about good wording to balance out the variation you'd get from a mass of silly votes like
"1 because he used Soul Reaper and I prefer Shinigami"
"5 because they're the first out and I don't read any others"
"[1 or 5] all the time because they have a general reputation of being [bad/good]" (when this reputation may be self-perpetuating and incorrect, or outdated, or...).

I'd guess that speed, popularity and cleaning (easier to judge) would determine Anonymous's ranking of scanlations - everything else would be lost in the noise. Translation votes would be almost entirely on popularity. ._.
#56. by shirogami_tsukasa ()
Posted on Apr 20, 2009
Thanks for this post! Lately I've been considering learning Japanese so that I can help translate manga. I'm not sure how to volunteer as a proofreader, so I'm just going to do it here. I'm a native speaker, I have a bachelor's in English, and I'm also Korean, so I understand how sentence structure can be easily confused when translating. Let me know if I can help, otherwise I'll be adding suggestions to the translations on my own.
#57. by tradedaemon ()
Posted on Apr 22, 2009
(I only managed to get through half the comments before it was all redundant, but past that, someone might have already said this)

Why not just set up some group of people (I like the roundtable name from the last article comments)willing to help with the English problem, and have translators go to them if they want to improve? I'd use the service. That way, those who want to improve do, and no one is policing anyone. Keep the site a resource, rather than a measuring stick.

I mean practically, you're not gonna fix the problem by forcing anyone to abide by anything, so why not just focus on those who are willing? That'll be a step up at least.
#58. by Tsuchikage ()
Posted on Apr 22, 2009
I can see what you mean by this being a very much voluntary community, and I do think that proofreading in terms of general quality of English would be good to have. I also agree that it would be good to have those who care be the ones.

I also deeply agree with the fact that it would be much better for there NOT to be a rating of translators nor any kind of ranking system. I think one of the most beautiful things is that people, even amateurs, start off translating and receive quite a measure of attention (albeit in the form of scanlators), which then encourages them to keep on improving and working at it.

Also, in regard to the note about how just because someone has no knowledge of Japanese that doesn't mean they can't be a good proofreader, that's why there's a difference between quality checkers and t/l checkers. Personally, as a translator, I actually want both. For those people with only a year under their belts, not just am I worried about the smoothness of the English they have translated, but also whether they translated the thing correctly to begin with.

I also say this mainly in regards to myself, and have been considering, with more time, I will make sure to always include a transcript in Japanese with my translations. I'd really appreciate it if we had TL checkers, to pick out where we may have mistranslated/misunderstood the original Japanese. Accuracy of translation is one of the greatest reasons why I started to learn Japanese. I don't trust translations to 1. be accurate or 2. retain the richness of meaning (nuances etc.) It's also just plain disappointing when a joke is lost in a bad translation.
#59. by cnet128 (MH's Best Translator)
Posted on Apr 23, 2009
Quote by Hyperworm:
I certainly wouldn't trust Anonymous to pick out the best translation/scanlation. The vast majority are completely unable to judge accurate translation, so the opinions of people who know Japanese would be drowned out totally.
I know that voting for translations/scanlations isn't the perfect solution. But even in a worst-case scenario, it has to be better than the OneManga system of "first scanlation wins, everyone else can go screw themselves". I'm optimistic enough to hope that at least the obviously better scanlations would end up rated higher than the obvious hack-jobs.

Of course, an ordinary voting system doesn't have to be the only option. I'm sure there are alternative ways of identifying the top scanlation. For instance, if the public voting system really is that bad, voting could instead be restricted to a subset of members, essentially a small group of judges who can be trusted to choose fairly. Obviously this raises its own problems, but again, if it results in the manga archives consisting of something other than just the speediest hackjobs, it's a clear improvement.
#60. by ScaryRei ()
Posted on May 1, 2009
Aside from blatant mistranslations (and I HAVE see a few of those... "taiman" does NOT mean "Thai Man"), what constitutes as a good translation or a bad translation? What criteria would you use? I've seen translators blasted for being too literal, then again for taking too much liberties, using 'dialects', not using 'dialects', being too formal, being too informal, and the list goes on and on... Granted, at the very minimum, all translations should be proofread for correct spelling and grammar in the target language. J2E translation is never going to work if you're doing a straight word-for-word translation (Babel Fish, anyone?). The best translators that I've seen are those that not only have a good understanding of the source language, but also have the ability to write in the target language as well. Having a proofreader/beta is always a big plus.

About the Author:

Alias: zindryr
Rank: 誰も知らない
Registered: Mar 28, 2005
Message: PM |

Author's Contributions

Forum Posts: 1628

Latest Translations

Date Manga Ch Lang Translator
Aug 10, 2018 One Piece 913 en cnet128
Aug 7, 2018 Yakusoku no... 98 fr Erinyes
Aug 5, 2018 Mahou Shoujo of... 58 en Lingwe
Aug 5, 2018 Chaos;Child... 3 en Dowolf
Aug 4, 2018 Shokugeki no Soma 274 fr Erinyes
Aug 3, 2018 One Piece 912 en cnet128
Aug 3, 2018 One Piece 911 en cnet128
Aug 3, 2018 Gintama 693 en kewl0210
Jul 31, 2018 Shokugeki no Soma 273 fr Erinyes
Jul 31, 2018 Shokugeki no Soma 272 fr Erinyes

Latest Releases

Date Manga Ch Lang Uploader
Mar 1 MH Yearbook 2013 Mangahe...
Jan 19 MH Yearbook 2012 1 Mangahe...
Nov 14 Houkago 1 Osso
Nov 14 Oragamura 1 Osso
Nov 14 Kenka 1 Osso
Nov 14 101Kg 1 Osso
Nov 14 Murder 1 Osso
Nov 14 Doubles 1 Osso
Nov 14 Pinknut 1 Osso
Nov 14 Kimagure 1 Osso