Improving BT Quality Discussion

[obsidian pegasus]

This has been very interesting from my perspective, which is one of the low tier translator's.

Translator standards- I recently took a test from one of the scanlator groups, read the pages, sent it back, 20 min no problem. Could I have done that a year ago? Not a chance in hell. why? Because Like Larethian said earlier that he's noticed people get better the longer they work at it, that's what's happened, but without working at it the past year, I wouldn't have gotten better. I started taking 4 hours to get one page done, now I can read through 10 pages an hour.
What should we do about newbies trying to learn, well I'm biased, so please don't shut us out, help us keep learning. Most of the time were off on a side project that wasn't getting any attention anyway's that no one cared enough about to deal with, so its not like were hurting anyone, and if we are, point it out please, since were learning we know we need to get better and will take critiques well. Putting up a wall though and saying, here send in this chapter TLed as a test before we let you work on a project, (equivalent of a scanlator test), would be extremely difficult for a nube. Maybe we can strike a balance? Doing just a prologue maybe, they're usually only 10 pages or so...?
Another TL problem I've noticed reading is even if the TLer speaks wonderful Japanese, they often don't have the same vocabulary in English, even if they have great conversational skills, face it, conversational vocabulary=2000 words that's why dialogue is so easy to TL. Discriptive vocabulary....20,000+ words? With this in Mind, even good translator's often can't word things that well, while nubes that grew up speaking English can grasp the sentence's essence rather than just its Translation, so there are perks to Nubes....(reminder: I'm a biased nube).

Editor's- Since I'm new(er) at this, editor's like 神 or other's have saved my life. When I'm translating, I think in Japanese sentence format, and for whatever reason, my English comes out the same way, so sometimes the verb will be at the end of the sentence. Even with this, a quick TLC by these guys cleans the sentence and it makes perfect sense. So This makes up for a lot of the "Poor quality" translations, its not that we read it wrong, (well, sometimes), but rather we worded it wrong, (even though I have a minor in English and grew up in California...go figure)

Rules- Well, If its incoherent, my vote is the translator responsible be put on a kind of notice and their work have a note by it saying it needs TLC to warn the readers.

Translator support- Someone mentioned there should be a place where TLers should work as a community to help the Nubes...um, there are several translation threads in the Forum already where I've gotten lots of help, and learned a ton. There are also how to learn　日本語 threads. Heck we even use it in the "banned" thread for fun.

Quality-? I haven't heard the same rumors you guys seem to. Most all of the comments I read are grateful happy people looking for more on all the feedback threads, I know bad one's aren't being deleted because I help approve or deny the post here when there's backlog. Heck, I've even seen 2 or 3 offering to pay translators to speed up their work...(nothing I was working on, bummer, (jk)). So where have you guys heard these rumors? Or is it scanlator groups who spread the rumors because they, (understandably), have pride in their own group? Or is it those random spammers that show up everywhere that can't be made happy no matter what you do?

[decemdies]

Quality-? I haven't heard the same rumors you guys seem to. Most all of the comments I read are grateful happy people looking for more on all the feedback threads, I know bad one's aren't being deleted because I help approve or deny the post here when there's backlog. Heck, I've even seen 2 or 3 offering to pay translators to speed up their work...(nothing I was working on, bummer, (jk)). So where have you guys heard these rumors? Or is it scanlator groups who spread the rumors because they, (understandably), have pride in their own group? Or is it those random spammers that show up everywhere that can't be made happy no matter what you do?

I actually don't find much complaints about our quality but then again, i do not visit the forums much.
Personally, I only commented here because even if there aren't any complaints, wouldn't it be nice to have a system that encourages low-tier translators to improve their own japanese? That's what i'm actually hoping for anyway. When i first started translating, i just stopped midway because it was too hard for me to fathom finishing even 1 chapter. I managed to finish that chapter after a long hiatus but you know, i think the community can really benefit from having a culture or a system that encourages quality as long as it does not discourage people from becoming translators. Which was why I proposed the culture thing.

[obsidian pegasus]

wouldn't it be nice to have a system that encourages low-tier translators to improve their own japanese?

Yes, It would.

That's what i'm actually hoping for anyway....i think the community can really benefit from having a culture or a system that encourages quality as long as it does not discourage people from becoming translators. Which was why I proposed the culture thing.

I've gotta agree with that too, but how would it work?

[decemdies]

That's what i'm actually hoping for anyway....i think the community can really benefit from having a culture or a system that encourages quality as long as it does not discourage people from becoming translators. Which was why I proposed the culture thing.

I've gotta agree with that too, but how would it work?

Well, if we really get down to the specifics, then I'm still not sure how the entire system should be planned out but basically:

The idea is to be more friendly. That's all. If we can get translators to motivate and learn from each other, then I think new translators could presevere long enough to translate substantially. Of course, we should not force translators to do it. Rather, we should have maybe a "Welcome new translators" commitee. This group is not in charge of making sure that new translators do the translation work. It's supposed to be in charge of making sure new translators understand the rules of the forum and the resources available to them when they need help. It's just telling them: Hey! You need help? Go to the lingua franca lexicon page. Hey, need to discuss editing? Go to some part of a forum.

Note that although every translator in BT has access to these resources, not every translator(especially newbs) knows that other translators are quite willing to assist you in translated related stuff. Even some of the current translators might not know of these resources. Of course, it's best if you could search the answer from the resources yourself but for a person that is unable to read japanese with kanji,(especially from sites like weblio which provides excellent information) then it's very difficult for translators to improve with great stride.

So, maybe they could be directed to a forum where anybody can ask questions and translators or other knowledgable people who roam around the forum can just pitch in their opinions. There could also be a page for translators where FAQs are compiled for translation reference. Like the use of the word 分からず which is a slang for
分からない. Correct me if i'm wrong for the meaning of 分からず. I occasionally come across these little things that just make translating more difficult. So i think it would be beneficial if anyone could access these FAQs.

[Hiyono]

As I understand it, project supervisors have quite a bit of leeway when it comes to managing their project. As such, why not put forth a proof of concept of sorts by running this on a test project? Instead of rehashing the same arguments time and again, why not just do it and let that speak for itself?

[cloudii]

As I understand it, project supervisors have quite a bit of leeway when it comes to managing their project. As such, why not put forth a proof of concept of sorts by running this on a test project? Instead of rehashing the same arguments time and again, why not just do it and let that speak for itself?

Originally, there was a part of the Tsuki-chan proposal that called for optional diagnostic tests, and a "stamp of approval" to be placed on top of pages. These two aspects would require approval from higher-up before implementation, since they would be extensions to the Baka-Tsuki system.

Many of the other suggestions, however, could be implemented right here and now. However, it wouldn't really be much of a proof-of-concept, since no one ever really notices when a translation is superbly written and stylistic. People really only notice when a translation is sub-par. I'm sure all of us ("us" referring to translators and editors) are very well aware that our main readership on the Facebook group generally could care less about quality as long as the gist of the plot gets through. In order to produce change that has an effect on a community scale, it needs awareness--and what better way of gaining awareness through a debate?

To be honest, I started this thread with the sinister goal of changing the culture on Baka-Tsuki. I'll be very frank. I honestly do very much hope something like the Tsuki-chan Challenge will yield an effect on the greater community of contributors. Why? Because currently in our culture, there is no respect for quality. Everything is about speed, because our readers care about speed more than anything else. Our readers wouldn't even mind if we sacrificed accuracy, style, or comprehensibility in order to produce faster translations.

If you've been in this business for as long as you have, translators (<3 your dedication), I'm guessing you're very likely a goal-oriented person. You aspire to reach the next goal: such as finishing the next chapter, or finishing the next volume, or reaching the next benchmark. These goals drive you and keep you going. You like the series, but rather than keeping the story to yourself (b/c you've read it already), you want to share it with the community. There's an end-point that we're all reaching for, and it makes us (or at least me) quite proud of myself when I can finish a volume of translations from my own effort. It's something I can honestly take ownership of.

For the most part, I'm driven by the motivation of "finishing". Quality of English generally sits pretty far back in my mind. From what I seem to understand, most of the not-so-amazing translators are pretty consistent in keeping their translations very very literal out of fear or screwing something up badly. However, that's not necessarily the best English to read. In fact, those are the situations when people do notice the translations are not-so-great.

The question is, how would you feel if there was a Golden Trophy of Sorts, sitting at the finish line? If there was a Golden Trophy of Quality sitting at the very end, would you be motivated to think more about quality? Would you be drawn to it? Would you want your project to eventually reach that point?

Actually, to be honest, as a translator, it hardly appeals to me. XD But then again, that's why I envisioned this as something for editors to aspire to. I don't know why, but everyone kept derailing the conversation back towards the translators. Honestly, in the ideal model of the translating workflow, translators shouldn't be expected to make a polished, finished product. The translator's job is make a translation that's understandable by the editors. Editors are the individuals who make a project "High Quality" (though good translators make the job a lot easier, and technically a incredible translator could solo it all... but we can't all expect to become "incredible").

This is a discussion that should theoretically be about editors, not about translators.

Is there anything we can do to help editors do their job better? Or any way we can attract more editors to baka-tsuki?

[Cthaeh]

Hmm... Thinking more about it from a goal-orientated incentive, I see new value in your suggestions to mark things as in need of editing. I suppose if I look at my case, I started editing on BT because I had a specific goal I wanted to work towards (convert early Index volumes to past tense). Having a well defined goal probably made it easier for me to start. And the editing I've done after that was completed has been somewhat goal orientated too, at least the editing that has been most effective. Giving new editors well defined goals for specific projects may attract more of them; and it might also help temper some of the 'over-enthusiasm' that new editors sometimes have when first starting (I could use myself as an example).

Of course it might not always be as easy to define specific goals such as converting tense, or updating terminology. But providing a framework for the editing side of things could give editors a greater sense of accomplishment. Though, the optional enrollment and translator engagement aspect might come back as important here, because attracting more editors could also mean attracting more 'bad' editors with revisions that would require translator review and undos

[Detalz]

Translation groups do not form over night. Most editors randomly apply based on sudden interest, and even then suffer from poor communication with translators. I think the problem here lies in awareness. Why not incorporate a poll/recruitment poster for editors/TLCs in the thread for each series? Translators interested in cooperating with editors can choose to place the poster on the top of the thread. The status can be modified by translators to tell what they are in need of or if all application help slots are filled. The applications may then be forwarded to Project Supervisors/Primary translators, who ultimately determine the criteria for accepting the editors(some more strict than others). And until the applications start rolling in, we(the translators) can only be patient.

I don't disapprove of this "stamp of approval" idea. But who's going to be doing it? It's going to take a LOT of work to have to review each submitted series for quality. And for free too. I think a category creation for "Edited Projects" as opposed to "Non-edited Projects" would work. Though there will still be a discrepancy in quality, I just don't think Baka-Tsuki will find a task force large enough to enforce this stamp approval idea. They'd need to be proficient in Japanese, have lots of free time, and be willing to read series they might not want to read. Though a merit system to indicate a translator's quality might work and simplify the process much better, I'm not too sure about this idea. Note this can be abused, as one chapter may be much more polished than others because a translator knows it will be reviewed and graded.

[obsidian pegasus]

I have a question about the TLC guidelines. I agree, being able to sit down and read both the 日本語と英語 to check for mistakes would be brilliant, Isn't that kind of person already likely a translator working on a project somewhere?
Now, On the flip side, I'm reading baka and test in Japanese instead of the TL and I keep itching to fix things I see, (all of Yuuji's insults were changed to "idiot" instead of what they really are). So I can see how someone taking the time to do this is invaluable.
Weighing both of these it comes down to not having enough staff and way to many amazing light novels everyone wants TLed, right?
Suggestion then.
Translator-Sama's such as Teh Ping (bow) and other veteran's do translations like normal.
TLC-Its been suggested we need to find some way to increase the level of new translators because though they (we) might suck now, they (we) will get better over time. With this in mind, why not assign them the job of TLC, reading through, comparing it to the English in order to learn as they go? Perhaps correcting minor mistakes and posting why they did so in the comment area of their correction.
Editor's- I believe these are they guys that make the quality. If the translation reads like pigeon talk, (which is the only quality problem I've really noticed) and its not Hina talking from papakiki, then Editor's do what they do and smooth the wrinkles in the translation like they already do.

(I guess the point of this post was to point out a way to welcome new translators and train them by giving them the TLC job to raise their level for future projects)

Also, as was pointed out earlier, some LN are MUCH harder than others. I graded the 13 series I own on difficulty and % words I don't know/page and One of my series is actually down in the 70% range, VERY difficult, while others are 95% or so, (魔王なオレと不死姫の指輪/問題児達が異世界。。。）Perhaps grading the series out there so nubes don't bite off more than they can chew. (For instance, Its taken me 7 month's to finish the chapter I started TLing because the level was difficult, during that time I've done 4 other chapters and read several other books. (oops) )

[Hiyono]

This is a discussion that should theoretically be about editors, not about translators.

Is there anything we can do to help editors do their job better? Or any way we can attract more editors to baka-tsuki?

I might be mistaken here, but I'd say the simplest way to make editors more important is simply to have them come into play before the "final product" is released. That's one of the major differences between groups like Nano's and B-T: when it comes to the translation process (translation -> TLC -> editing), groups ensure that the process is finished in its entirety before the work is ever given to the public at large. B-T on the other hand opens up the work to the public the second translation is over.

If the goal is to change the culture of not caring about quality, the simplest way is to not make that a choice. Arbitrarily assigning numbers here, if we say that a given translated text is 85% of a finished product, with editing contributing the final 15%, I would venture to say that most readers would be more than satisfied with the base 85%, i.e. the unedited work. (Given their predilection toward machine translating things for themselves to read and the sheer intensity they display in seeking out any and every spoiler, I would say this is a fair assumption.) That is not to say they cannot appreciate the value of the full finished product, but that it is not a requisite condition for them, and that, as the weights would suggest, there is at best a "minimal" gain when comparing an edited work to an unedited one.

By forcing the edit process to predate release, as with more traditional groups, the "culture" which does not value quality is simply no longer an option; whether you're fine with an unedited copy or not, you're not getting anything less than the best. This establishes the degree to which translators' esteem quality and to a large extent is the reason they are able to recruit editors who care, i.e. because they've shown they care.

To a large extent, this is one of the reasons the quality work we have come to expect from translators like zzhk, Dreyakis, and js06, among others, attains such a degree of quality - they personally review and revise their work before it ever sees release. It's still following the more traditional translation -> edit -> release process; it's just that as fluent English speakers they also serve as their own editors.

[obsidian pegasus]

Here's what I'm taking from this discussion-

A. we need some form of quality control (can come in the form of a test like scanlator groups do)

B. We need some form of process. (Translate, TLC, Editor etc...)

C. This is obviously not necessary for some projects (there are some amazing translators here), but
perhaps necessary for the minor projects flying
under the radar, (and the teaser section).

right?

Suggestion

1st- make a list of "questionable" translations that need this kind of review
2nd- Have a roving Editor and TLC as suggested above review these projects
3rd- encourage and help these still learning translators to improve, (after all, were here to have fun!)

Conclusion

That would take care of everything right?

[Tony Yon]

this is just an idea that pop up at my head a few hours ago. how about making a subforum for recruiting editors? the lack of quality sometimes come from lack of dedicated editor/s but sometimes it's hard to get an editor (though most likely those who interested just don't know where to start)
so it's like a sub forum where translators can ask if there's anyone who willing to help them to edit the series they working on. it'll linked with project main pages, and every editor (major one, not for minor edits) can apply as editor and check their compatibility with the translator (skill, style, etc..)
so, anyone free and feel like helping on editing, can enter that sub forum and choose a project that recruiting editors. and vice versa, for translator who need dedicated editor/s can post there and wait for response/s.
for those who have joined as dedicated editor, can discuss with the translator with whatever way they feel the best (google docs, or focus on forum is fine)

um... anyone understand what i'm trying to say..?

[Misogi]

this is just an idea that pop up at my head a few hours ago. how about making a subforum for recruiting editors? the lack of quality sometimes come from lack of dedicated editor/s but sometimes it's hard to get an editor (though most likely those who interested just don't know where to start)
so it's like a sub forum where translators can ask if there's anyone who willing to help them to edit the series they working on. it'll linked with project main pages, and every editor (major one, not for minor edits) can apply as editor and check their compatibility with the translator (skill, style, etc..)
so, anyone free and feel like helping on editing, can enter that sub forum and choose a project that recruiting editors. and vice versa, for translator who need dedicated editor/s can post there and wait for response/s.
for those who have joined as dedicated editor, can discuss with the translator with whatever way they feel the best (google docs, or focus on forum is fine)

um... anyone understand what i'm trying to say..?

The idea seems good on paper, but practical-wise, I think that some simple-minded people will use it to ask for translations, as it's a recruiting sub-forum.

[Hiyono]

this is just an idea that pop up at my head a few hours ago. how about making a subforum for recruiting editors? the lack of quality sometimes come from lack of dedicated editor/s but sometimes it's hard to get an editor (though most likely those who interested just don't know where to start)
so it's like a sub forum where translators can ask if there's anyone who willing to help them to edit the series they working on. it'll linked with project main pages, and every editor (major one, not for minor edits) can apply as editor and check their compatibility with the translator (skill, style, etc..)
so, anyone free and feel like helping on editing, can enter that sub forum and choose a project that recruiting editors. and vice versa, for translator who need dedicated editor/s can post there and wait for response/s.
for those who have joined as dedicated editor, can discuss with the translator with whatever way they feel the best (google docs, or focus on forum is fine)

um... anyone understand what i'm trying to say..?

Unfortunately, the problem isn't just finding editors, but finding qualified ones. To be quite frank, even being a native English speaker is not a sufficient condition.

[Fuwan]

If you want to improve the average quality, you need to make sure that the translators are proficient with at least one of the two languages they are using. Certain translations on the site do not meet that minimum guideline. I don't want them to stop translating, necessarily, but I would prefer as a reader if there was some sort of quality check before a given translation ends up on the site.