Search

Recent Posts

• It’s that time again
• Eroge review: Muvluv Alternative
• Eroge review: Dra†KoI
• Dickinson and SCA-JI
• fucking anime, fucking :words:
• The Problems with Bakatest
• i changed my mind
• Winter 2010 in haiku form
• Why do people think Railgun has good production values?
• 2009 in review

Topics

• Anime
  • After Story
  • Akasaka
  • Bakemonogatari
  • Chaos;Head
  • Code Geass R2
  • DCIISS
  • ef (melodies)
  • FMA
  • Hayate
  • Kannagi
  • KKN
  • Rakkyo
  • Random Shit I Watch Because I'm Bored
  • Sora Kake
  • Tales of the Abyss
• Comics
• Fuck
• Manga
  • Happy World
  • Negima
• Other
• PUBLIC SERVICE ANNOUNCEMENT
• Ramble
• Umineko
• Uncategorized
• Video Games
  • D3
  • Gust
  • IIDX
  • P4
  • Touhou
• Visual Novels
  • CROSS†CHANNEL
  • Sharin no Kuni

Archives

• July 2010
• June 2010
• May 2010
• April 2010
• February 2010
• January 2010
• December 2009
• November 2009
• October 2009
• September 2009
• August 2009
• July 2009
• June 2009
• May 2009
• April 2009
• March 2009
• February 2009
• January 2009
• December 2008
• November 2008
• October 2008
• September 2008
• August 2008
• July 2008
• June 2008
• May 2008
• April 2008

Blogroll

Blogrolls are for suckers

How to run an eroge localization project

By Moogy | September 11, 2009

An associate of mine requested that I write this up.

This post will, as the title implies, contain some general guidelines for establishing, managing, and maintaining a group of persons for the purposes of localizing an eroge or visual novel into the English language, as well as the localization efforts (henceforth referred to as “projects”) pursuant to such an organization. (As I am unfamiliar with foreign communities, I will not claim that this advice applies to or is useful for them. However, most of the tenets contained within should be fairly solid regardless of your language, so feel free to read it anyway.)

It will not tell you how to run a fansubbing or scanlation group or project. I suggest you look elsewhere or kill yourself if you are interested in one of those tasks.

Regarding the audience of this guide (as it were)… I suspect it will be fairly varied. However, I am tailoring this post to a few different sorts of individual in particular:

• An aspiring translator who wishes to translate a game in order to have fun, practice his Japanese, give back to the community (yeah right), or such, but feels at a loss as to where to begin with the whole task. Even if you know Japanese, it can be a bit daunting at first, especially if your game isn’t hacked yet! This guide will lead you through the entire process, step by step, so don’t worry.
• An individual who is capable of a significant contribution other than translation (be it hacking, editing, or project management in conjunction with one of the other two) and is interested in starting up a project. Please be aware that this guide will not allow you to magically find a translator or something – at most, I can provide you with tips as to how to go about locating one. Instead, this guide will most likely serve you as it would the aspiring translator. Please note, however, that you have a rather tricky task ahead of you. However, there is precedent for this sort of thing, so don’t get too discouraged!
• Other established fan translators interested in my take on things. Hi, guys. Maybe I can help you out a bit, eh?

And now, the people who this guide is not intended for:

• Freelancers who know their shit. I’m mostly referring to request-based hackers like Nagato here; people who have no interest in what this guide has to offer and likely don’t need it either.
• People who just want to join a group and QC/”edit”/image edit/do whatever. I do not intend to be rude by saying this, but your services are frankly expendable and you would not have much to gain by reading this in the first place.
• Those who are otherwise uninterested in what I have to say here.

Of course, if you’re merely a random passer-by who happens to be interested in what I’m talking about, then, by all means, feel free to peruse this guide and offer your commentary. In fact, I invite anyone who reads this to offer their thoughts on it either to me personally on IRC or via the comments field on the post itself!

Oh yes, please note that this guide will seem rather idealistic at points. Don’t worry; I’ll address that issue in Section III, wherein I will explain how to adapt these concepts to the real world, so to speak.

With all of that said, let’s get this show on the road, shall we?

Section I: The Basics

I-I: Choosing A Game

All right, so you’ve picked out your game and you’re ready to-

What’s that, you say? You don’t have any particular game in mind that you would like to translate? Well, there’s your first red flag.

My advice in this regard is going to be at odds with the stance of Amaterasu Translations in particular, but so be it.

Put simply, you should pick a game that you have a vested personal interest in. Please do not allow yourself to be swayed by ignorant masses (/jp/) and end up translating a game that you hate.

To this end, if you are a translator, I would highly, highly recommend that you only translate games that you have both played to death and loved to death. This will prove to be a great help in two regards: First, the quality of your translation will be higher if you’ve played through it in the past (preferably more than once) and possess a comprehensive knowledge of its text. This will help you identify areas which prove problematic to translate, understand the author’s writing style (or lack thereof), and generally be more familiar with just what the hell you are actually doing. Second, you will much more motivated to actually finish the translation project if you enjoy the game and enjoy the process of bringing it to a wider audience.

Now, on the flip side, if you’re looking for a translator to work on your pet game, you likely already possess a rather vested interest in whatever game it is, yes? If not, I would highly suggest rethinking your course of action. In any case, you can generally assume that any translator who is willing to help you out fulfills the above criteria, so you don’t need to worry about this too much.

If you are an established fan translator, you know what you’re doing already. I would advise skipping this section entirely, in fact.

Also, one thing I would like to make clear: If the game you have selected is not hacked (i.e., in a state where insertion of English text and/or images into the game is possible and the end result of this is pleasing to the eyes), don’t worry about it. If you are truly enthusiastic about your undertaking and have a dedicated translator, you can find a hacker to assist in your efforts with but a modicum of communicative prowess. You should, of course, locate one ASAP, since it’s no fun translating a game without being able to see your work in action, but you shouldn’t be dissuaded from tackling a project in the first place if you don’t have a hacker.

Anyway, I am going to assume that you have a game that you are interested in translating at this point. Let’s move on to the next step.

I-II: Assembling A Team

For the purposes of this section, I shall assume that you, the reader, are the only person interested in translating the game you wish to translate. If you have some friends on-board already, that’s great, and will likely make this step much easier for you, if not nonexistent in its entirety.

First and foremost, I’m going to introduce who you need, with guidelines regarding locating them to follow.

I shall divide the positions you need to fill into two categories: Critical and Non-Critical. Put simply, I could not in good faith support a translation project lacking any Critical role (with certain exceptions), whereas I find Non-Critical roles to be not unnecessary, but rather more dispensable.

So, without further ado:

Critical Positions

• Translator – The big one. The ultimate arbiter of all translation projects, past, present, and future. There’s not really much more to say here – without a translator, there is no translation project.
• Editor – Despite what you may believe, an editor is absolutely necessary. (Note the singular; this will come into play later.) Avoiding poor English is doing a service to both the original work and your readers – though some leechers may be satisfied by substandard efforts in this regard, you owe it to the original author, yourself, and the fans who really care to present a polished script. You can skip having an editor if you possess the literary prowess of Seung Park, but you likely do not.
• Hacker - AKA reverse engineer. Again, this is obvious. You need some way to get the text in the game, after all. Note that this role can be skipped if you’re working with an engine  that is already conducive to translation efforts (NScripter, KiriKiri, etc.)

Non-Critical Positions

• Proofreader/QCer – Although someone dedicated solely to ironing out typos and grammatical mutations provides an important supplement to the task of editing, your editor should be strong enough that such errors are nowhere near common. Proofers are incredibly easy to find anyway.
• Image/Video Editors – Translated menus are very nice, but they’re not essential, and very few games require enough image editing that finding an image editor becomes a real concern. (To my knowledge, only Chaos;Head has necessitated image editing to date.) Regarding videos: First and foremost, it should go without saying that karaoke is essentially pointless bling. Only include it if you have some free time during the localization process for whatever reason. As for other forms of video editing… I cannot say I have ever encountered a game that would absolutely need it, but use your discretion. Subbing animated cutscenes is probably important, yeah?
• Anything Else – Throw any bright ideas you might have about new and exciting forms of task delegation out of the window. If a position isn’t listed here, you don’t need it. Exceptions are as follows: You may optionally recruit a translation checker, but your translator should be good enough that he or she does not require one. You may also elect a member to the fabled office of “Project Management,” but this should be naturally filled as the project and group wear on anyway. (More on this in section III.)

Refer to section II-II if you want more information about what exactly each member of your group should be doing to properly fulfill their role.

Now, I would like to stress, and not for the last time, the importance of a small group size. It is significantly easier to stay motivated and organized when are you dealing with five people or less – as such, five people is a good size to aim for when starting up a project. If you have less, that’s cool, go ahead, but, if you have more, you may want to consider cutting some chaff. Remember, ease of communication is important, and one of the easiest ways to complicate things is by having too much communication! There is a great deal more regarding this philosophy in section II-I, so skip there if you want some more reasoning regarding this suggestion.

Also, you’re probably wondering how to recruit people, right?

Well, I’m going to be frank here. Recruiting a translator off the street, so to speak, is likely impossible. Don’t think about starting a project without having someone willing to actually translate the game beforehand.

However, if you need editors, QCers, image editors, etc., then you have a much better chance (as in, an extant chance) of finding some. I would suggest posting a topic about your project on Gemot, then swinging by #Ensue or #tlwiki on Rizon. If you can prove that you’re competent and dedicated, you should have no problem recruiting a scurvy crew to aid you with your translation. You might also want to put a page up on the TLWiki itself, since there are random helpful people hanging around there as well.

/jp/ is probably a not great place to announce a translation project or ask for help, given the transience of posts and the board’s anonymous nature.

Anyway, just ask around a bit and you’ll find people to help. It’s not that hard, to be honest.

So, I’m going to assume that you have a team together now! On to the next section!

Section II: Getting Down to Business

All right, so you have a team together now.

Your first order of business is to get the game hacked and ready for insertion. Dump all of the text, figure out a word wrapping solution if need be, write an inserter if need be, an installer for your patching solution (if necessary), etc. This is the hacker’s job, and is explained a bit more later. The long and short of it is that you should finish hacking before anything else, though.

You may be wondering what to do afterward, though! Though I should hope that it’s mostly intuitive, I’ll write a bit about proper workflow and how to avoid version hell and all. So, onwards!

II-I: Workflow

Basically, your standard workflow for a game script looks like this:

1. Translation
2. Editing
3. Proofing
4. Insertion
5. Testing (You probably don’t want to do these last two for every single script. It’s more efficient if you do them in batches.)

Pretty simple, right?

You’ll want to be sure to use an SVN or wiki to manage files, by the way. Sending around a bunch of files to people’s harddrives is a recipe for disaster – they will get lost. Explaining how to use an SVN or wiki is a bit beyond the scope of this post, but feel free to bother people in #tlwiki about either. You’ll probably find someone willing to help. Or, hell, just Google it. Don’t let relatively simple systems intimidate you, all right?

Anyway, I don’t have too much to say here. This stuff should be common sense, really. So, let’s move on.

II-II: General Expectations and Guidelines

Here are some tips for running a localization project. I’d like to state again that this guide is just my personal spin on things – but I’ve been around for a while and I think most of my advice is at least pretty solid.

What To Do:

• You must have an IRC channel where members of the group can (and are expected to) congregate and discuss the project. You can leave it open to public or make it private – the important point here is that IRC is by far the most effective form of group communication on the internet, especially for a project such as this. Instant messengers and forums are not a substitute in this regard for a number of reasons. Forums, for example, have clunky interfaces and lack true immediacy. Instant messengers are unsuitable because they focus on one-on-one discussion (a huge no-no; everyone needs to see every discussion pertinent to the project) and thus lack robust multiple-user chat features. The long and short of it is that you must have an IRC channel and every member of the group must be in it. Anyone unwilling to participate in the channel should be dropped from the project if at all possible. Trust me, this will save you a lot of hassle in the long run.
• Building on the above: Members of the group must remain in contact with other members of the group and should be expected to report their status regularly and when questioned about it. This means no sudden month-long breaks from the internet or their job. If something comes up, then something comes up, but it needs to be reported to the rest of the group ahead of time so they can cover your ass.
• You should remain relatively transparent to the outside world. A release date of “when it’s done” is complete bullshit and everyone knows it. You don’t need to have a fancy automatically generated statistics page or anything, but members of the group should be willing to report their progress, as well as the overall progress of the group, to anyone who asks politely. Seriously, do not be a dick about this. Remember that, in the end, you are providing a service to others, and they have a right to know your progress if you’ve publicly announced the project. Also, try to establish at least a preliminary ETA of some sort. It will keep the fans happy and provide you with some additional motivation.
• Have a project page of some sort. If you don’t want to go to the fuss of setting up a website, TLWiki is very accommodating to up-and-comers (and we’ll be happy to help you with anything in general). This is a pretty simple way of tracking your progress and presenting a face to the outside world.
• This bears repeating: Use an SVN or wiki (or both). No arguments – you need to have your scripts online and backed up, and you also need a central location for people to commit edits to. Generally, smaller groups are more suited to wikis, while larger groups are more suited to SVNs. SVNs are quite flexible, though, and you may wish to forgo a wiki in favor of them even if you have only two or three people!

What Not To Do:

• Do not include in-line translation notes. They are an eyesore and completely kill the flow of the text. For the love of god, do not do this. In general, translation notes should be kept to a minimum and confined to the readme. Keep your audience in mind here – you don’t need to explain honorifics, you don’t need to explain Japanese foods, and you don’t need to explain Japanese customs. Jokes or references that cannot be conveyed effectively without a TL note should be rewritten entirely or otherwise localized.
• Do not release an excessive amount of partial patches. A patch for each chapter of episodic games like Sharin no Kuni is fine. A patch for each week of CROSS+CHANNEL is fine. A patch for an individual route of a chara game is fine. A patch for each 200 lines of random nukige XYZ is not fine. Use proper judgment to avoid jerking people around.
• Do not make non-ero patches. One of the beautiful things about eroge is that they are firmly restricted to those who are mature and open-minded enough to deal with explicit depictions of sexual intercourse. Catering to prudishness and immaturity is doing a disservice to the medium as a whole and the work you are translating.
• Do not credit yourselves in the game itself. I’ll allow a certain degree of leniency here and say that it’s okay to add your names to the credit roll if you really, really want to, but I absolutely cannot stomach group splash screens at startup or alterations to the text in the title bar of the game window. Let the localization speak for itself and save your e-penis for the readme and website.

Got all that? Good.

You don’t need to follow all of this, but please keep it in mind.

And, above all – use your damn common sense!

II-III: Job-specific Expectations and Guidelines

To put it simply, there are minimum standards of quality by which people involved in any translation group must abide. I am going to be excessively blunt in this section. I apologize in advance if any of you are offended, but that’s the way it goes.

I’ll break this down by job as well.

• Translator - Ideally, the translator should know a substantial amount of Japanese – enough to play any eroge without doing more than consulting a dictionary here and there. They should not require a translation checker, either; acquiring a second opinion on tricky lines is certainly acceptable, of course. Basically, a translator should be worthy of trust – trust that they will not mangle the meaning of the original lines. Now, this next bit should go without saying, but, unfortunately, I must restate it:  You are translating from Japanese into English. Translations from Chinese or Korean are not acceptable. TLC and editing may be able to repair JCE or JKE fansub scripts, but they are not sufficient for eroge, due to the volume of text and the fact that eroge translations are dealing directly with literature. Speaking of literature, the translator should also be able to preserve the original author’s style to a significant degree – enough to allow the editor a feel for it, at the very least. (A native speaker of the English language is most likely ideal for this.) Also, multiple translators is a recipe for disaster. Don’t try it unless you are incredibly confident that the translators can work together effectively and efficiently.
• Editor – This will be a bit long, sorry. Above all, your editor must be a native English speaker. No ifs, ands, or buts; for, without exception, non-native English speakers produce fundamentally awkward English. A native speaker of the American English dialect is also preferable, as American English is more prevalent throughout the world than British English. Also, you should have noticed that I have been using editor, singular, exclusively throughout this guide. This is because half of the editor’s job establishing consistency throughout the scripts. Consistency in what regard, you may ask? Maintaining a consistent writing style (which is an adaptation of the writer’s original style fostered by the translator, of course) would be my answer to that. So, it should follow naturally that multiple editors will only detract from this consistency, yes? And, no, a GRUE is not the solution to this problem. A GRUE is, in fact, a laughably absurd concept, for it is merely doing the same job twice. In the interests of efficiency, I will discard the idea of this position entirely. Anyway, since we’ve established that half of an editor’s job is establish and maintain a consistent writing style, what is the other half? Why, what you would expect from an editor, of course – making the game script read like it was written in English, as opposed to translated Japanese. To this end, the editor should not only be capable of, but should be willing to rewrite entire sentences so that they sound natural in English. Remember: Correcting grammatical errors and typos is not editing. Editing is what I like to refer to as the art of localization, for it is the editor’s job to make the script shine in English as opposed to Japanese. As such, an editor needs to be a good writer and familiar with fiction, so that he or she can convey the writer’s original ideas in a way that makes sense to English speakers! (Of course, he or she should also avoid making things up entirely (unless he or she has an exceedingly good reason to), since that’s a bit contrary to this philosophy.) Ideally, the editor should know at least a bit of Japanese as well, so he or she can consult the original line for guidance when the initial translation is a bit too strange to make proper sense of. By the way, it is perfectly acceptable for an editor to request a QC or two. Personally, I feel more at ease when someone else has looked over my editing and made sure I haven’t introduced any strange phrasing or the like. Anyway, to sum this wall of text up: Native speakers only, writing skills required, more than one editor = bad.
• Hacker – Honestly, you probably know your job better than I do, but I’ll give it the ol’ college try, eh? Hackers (or reverse engineers) should be expected to capable of reversing encryption on archives, making unpackers and repackers for said archives, cracking DRM, and providing automatic word wrapping solutions. Obviously, more or less tasks may be required according to the engine you’re dealing with, but these are the basic stuff you’ll be encountering. I’d especially like to express the importance of automatic word wrapping – it can be done via an external DLL or via a parser run on scripts, but it should not have to be done via manually inserted linebreaks. Manual word wrapping is an excellent way to slow any project down, and should be avoided in all cases. That’s all I have to say here – apologies for the short section. More relevant info will follow in Section III, since real-world examples are pretty much necessary here. Oh, by the way – you’re also responsible for release engineering, so try not to pull a Message with the installer.
• Project Lead – Whoever assumes this role has a fairly weighty responsibility. Not only do they need to accomplish their own task, be it translation, editing, or hacking, but they also need to make sure everyone else stays in line. Some real-world leadership skills are required here, and those aren’t something I can teach in a paragraph or two, unfortunately. I will say this – you need to be patient. Assertive, but patient. Honestly, it’s probably best to avoid having this position entirely by sticking to small group sizes, if you ask me. However, if you absolutely need someone to manage all of your members, then, for god’s sake, make sure they’re someone who won’t lose their temper over random bullshit. It’d help if everyone got along with them in general, really. God, this sounds like a management seminar, doesn’t it? Anyway, basically, a project lead is responsible for setting timetables for tasks and holding participants to said timetables. They’ll also probably end up being in charge of PR – your IRC channel and any web pages you may use, pretty much. So, obviously, people skills are a must. This is a job that is much easier said than done, so to speak, so real-world (haha) experience is likely required!
• QC – Be anal about grammar and spelling. Try not to break the scripts when you fix errors (hi, newbie). This is a very easy job, so… that’s about it.
• Image Editor – Don’t suck at Photoshop. Get shit done on time. That’s all you need to do, so don’t fuck it up.

Whew, that was quite a read, huh? That about wraps it up for this section, though, so let’s move on to the last part of the guide – real-world examples!

Section III: Application, Application, Application

I’m getting kind of lazy so this section isn’t really going to be organized. I apologize!

Anyway, this section will just be some various anecdotes I’ve picked up along the way that you may or may not find interesting and/or useful. Feel free to skim through.

III-I: Distro

I didn’t cover this anywhere else, so I guess I might as well put it here.

• Most groups just use a simple DDL for their patch, maybe with a Mediafire mirror or something. This should be fine for pretty much anyone, bandwidth-wise, since patches shouldn’t ever reach enormous sizes.
• I personally don’t care one way or the other, but people generally like it if you provide an installer for your patch instead of just the raw files necessary. Probably something to keep in mind.
• NNL decided to do standalone releases for their localization of ef (which I worked on parts of). I have no qualms regarding this policy, but be aware that many people feel rather strongly about piracy, even if their arguments may be a tad ridiculous. Issuing a standalone is a great idea from a logistic standpoint, but it may narrow your audience a bit. They also eat bandwidth like the dickens, let me tell you.
• Joke about the UBW installer goes here.
• If you happen to be a moralfag, I would suggest coming down from your ivory tower and realizing that RELEASING YOUR LOCALIZATION WILL MAKE PEOPLE PIRATE THE GAME YOU JUST TRANSLATED. If you can’t come to terms with this, you should rethink things from the very beginning.

III-II: Hacking Shenanigans

Hacking’s an interesting job because you could be dealing with anything from NScripter to rUGP. Here are a few stories from the field:

• For our (unreleased) Demonbane patch, we actually ported the PS2 version of the game to PC by grabbing the PS2 resources and inserting them into the PC engine. This is possible for any console port that runs on the same engine as the original, so if your game has an enhanced console port, you should check it out and see if you can get any of its features working on PC. See also our (unreleased) PC backport of Chaos;Head Noah.
• If your game runs on NScripter or KiriKiri2, you probably don’t actually need a dedicated hacker unless the archives are encrypted in funky ways or there’s hardcore DRM on the exe. For example, I was able to complete all of the hacking needed for Sharin no Kuni by myself, using only insani’s KrKr2 toolset and word wrapping solution.
• There are random games out there that use fucking StarForce, like the original Happiness, so have fun if you come across it. (lol)
• There’s a group out there with a position for “linebreaker.” Don’t be that group, seriously.
• Joke about the UBW installer goes here.
• If you’re stumped with archive encryption, there’s partial source for ExtractData available. asmodean’s tools also include source. I’m pretty sure Crass is closed source, though.
• I actually ended up learning the scripting language for Nitro+ games and using it to rewrite Saya no Uta’s script to lag less in English. It was pretty fun and useful. Basically, learning the scripting language the engine uses may be beneficial. They’re generally quite simple, anyway.
• Also, have fun with bytecode, har har.

III-III: When PR Goes Wrong

Hoo boy. These might seem like nothing more than funny stories from the past, but drama can and will rear its ugly head at you, too, if you’re not careful. Try to avoid any of the following scenarios.

• “Support JAST USA’s official localizations of Nitro+ games!” Yeah, sorry, but I totally dropped the ball on this one. Communication is a good thing – don’t just remove a bunch of projects and act all hush-hush about it.
• BREAK THE FUCKING SUPPORT DISC MOTHERFUCKER. Yeah don’t do this.
• Joke about the UBW installer goes here.
• Starting a poll is akin to starting a giant trollwar across the entirety of /jp/ and IRC. I would suggest not doing it, unless you are capable of completely shrugging off all drama.
• GamePatch hurr durr
• TH2 hurr durr
• Narcissu Side 2nd hurr durr

III-IV: Other Random Shit (AKA Trolling)

Ha ha!

• If you are a manga scanlation group named Hidoi, you probably shouldn’t try to translate games.
• In fact, if you pretty much suck in general, you shouldn’t try to translate games! Don’t be one of the shitty new groups popping up to produce Engrish “translations” of random bullshit. See: aninga
• Don’t translate random shitty freeware games that no one gives a shit about. See: insani, al|together
• Don’t leave up project pages for shit that will never be finished for fucking years.

Okay I’m done. Nice guide huh.

In closing, I would like to state that I am totally correct about everything and that if you don’t follow this guide you will get AIDS and die and no one will attend your funeral.

Appendix: Resources

AKA linkdump.

• TLWiki – Has a  bunch of resources on the tools page, contains lists of ongoing and completed projects, has a bunch of project pages to look at, and is generally a good place to get some support for your translation.
• Gemot – The biggest eroge forum in English.
• encubed – The biggest eroge localization news site in English.
• Crass – For all your extraction needs.
• /jp/ – Believe it or not, translation projects have gotten started here.
• #tlwiki@irc.rizon.net – Bother me and other people here.
• vndb – Some sort of listing of eroge. Might be useful.
• ONScripter English – Courtesy of Haeleth/insani.

Topics: Visual Novels | 8 Comments »