The Great Localization Q&A

Hiya, everyone! A while back, NickyD answered some questions about localization (and the Hyperdimension Neptunia games) over on the Niche-Gamer forums. It started out as kinda specific to stuff he did, but he went on to talk about the entire process that was practiced where he used to work. It's pretty interesting, so I wanted to take it and post it here, too!!!!

It's a long read, but it's totally worth a look if you're interested in the game industry, publishing, translating, and/or localization. Thaaaaaanks~

~*~Tsukarechan~*~

Umm, so below is a copy of NickyD's posts. I guess I should link to the forums, too… It's kinda neato to read the posters' attitudes before and after the Q&A. I dunno much about this stuff, but I think it might be 'cuz a lot of people in the industry, and especially Daddy's industry (he's a Japanese salaryman, so he always says he totally has to do stuff "the J-way"), don't actually answer these kinds of questions 'cuz they're not allowed by their bosses. Or they don't wanna explain themselves to people who disagree with their choices. But Daddy says that's not what Mr. Tired Media is supposed to be, so we're always gonna be around to answer the super tough questions that bigger companies ignore! Yeah!

The first answer is very, very long. It helps set everything up to make answering the other questions much easier. I know it’s a long read, but please give it a go. If nothing else, you may walk away knowing way more about the localization process than some people who actually work inside localization companies!

Why so many memes? I personally feel like the times in memes in official translations have sort of passed by compared to the days of the Ace Attorney games on DS, although I ain't gonna lie, I did chuckle at some of them. But still, I do feel like the time for using them in retail releases has passed, as they sort of date the games' humor.

I’m glad this is the first question listed, actually. Before I can answer the question itself, it gives me a good opportunity to talk briefly about the entire process that leads up to the “why” bit. While I’m not throwing anyone under the bus here, it’s probably for the best that I don’t name names—although it’s no mystery where I used to work, it’s often seen as better to be discreet when talking about how stuff worked on the inside while I was there.

The next few paragraphs talks about how Neptunia first came over; it’s not a complete account, just my own. Feel free to skip it and go on to the paragraph that starts with TL;DR if you’re not interested. I think it’s a fun story, though. Haven’t told it quite like this before.

So. Go back to sometime in 2010, when the very first game in the Neptunia series was being pitched to my then-employer. Games from the company pitching Neptunia were not viewed highly at the time, and after average to harsh reviews of previous titles (yes, reviews do carry weight with certain things within such companies), the default response was to pass on the game. I thought it looked like a novel idea, so I kept bugging the then-president to consider the title; he asked if I could show that there was a market for it. I started posting about the game (after the Japanese announcement of its existence, of course) on the company forums and a lot of people joined in. It quickly became one of the most popular threads on the company forums… Next thing you know, the company decided to pursue the license and Neptunia made its way to the states. I’m NOT saying it was thanks to me, but I do feel like I lent a helping hand when the developer didn’t have a stateside presence of its own. Maybe my actions had no effect, but the fact remains that myself and many others wanted to see it come to the US—and it did.

Now then, the game is licensed. It’s time for the localization team to get together for a “kick-off meeting,” where the game’s genre, tone, mechanics, and characters are described. This meeting was often led by the producer of the US publisher. During this meeting, it was decided by everyone that the Neptunia game was a modern pop-culture title steeped in Internet lore and fanboyisms, due to its paper-thin console war allegory and based on the jokes and references in the Japanese script (which I can’t read). I have some links to share that support these conclusions in a bit. Also during this meeting, depending on how on-the-ball the localization manager was, the team members working on the game would also be announced. For this game, the editor was chosen to be me because of my presence surrounding the game already (forums, championing its localization in the first place), and the translator was chosen to be one of the senior translators at the time. We began to work on the game shortly after, and since this was the first title, we had to talk about name localizations, the game title, and how to bring their personalities across between languages. I had no idea why we couldn’t call the game Hyperdimension Neptune. There were fears of copyright claims and lawsuits (this was during the time when people were getting sued for using the word “Edge” in their games—one such game was sued at this publisher for it and was taken off shelves for quite some time), so we did what we had to do and settled on a new name using non-words. My internal effort was to keep it as close to the original name as possible, and changing an –e to –ia seemed to appease the upper management, so there that was. This was also where we decided on Arfoire for the US name of Magiquone, Console Patron Unit (CPU), and kind of sets up where the next paragraph is going: the localization process where I worked was focused on “giving US gamers the same experience as JP gamers,” which is very different from a straight translation, or “giving US gamers the JP game in English.” That was the company process, and I was a part of it. Not the producer, not the manager, not the translator—an editor.

While I’m on it, the whole idea of CPU as Console Patron Unit was decided on because I was told that, in the original game, the kanji used to describe the girls as “hardware goddesses” was just that—it was both a technological term and a deification. So instead of using the direct “Hard Goddess” (heh), we settled on acronyms that exist in technology while giving them a more deified meaning. This trend, once established, has to carry through all of the titles for consistency within the company, which is why you see things like CFW (Criminals of the Free World) and DoS (Deity of Sin) crop up. Tonal consistency, I guess is what we called it.

TL;DR: That brings us to the part where text localization begins in earnest. First, you must understand that I am not a translator. The Japanese script is given to the translator, who turns it into English (kinda-sorta, depending on their fluency) and adds notes to help the editor when polishing their translation. Notes like “this is a pun having to do with the kanji used in her name and a fish” or “this is referencing a Pepsi commercial that aired in Japan during the 1980s” (yes, that was an actual note), or even “2ch meme.” So to answer your question of why there were so many memes: ask Japan. Given that I’ve described our process to this point, keep in mind:
· The tone of the game had been decided as pop-culture, Internet geek, and console war
· The company process is to give US gamers the same experience as JP gamers
The second bit in particular is important here. Memes, when directly translated from one language to another, lose all meaning and, almost universally, the context in which they are used falls flat. So to give US gamers the same experience as JP gamers, we had to find memes that exist in the US and put them in the dialogue as close to where the original meme was brought up. Sometimes this resulted in heavy rewriting to get the joke to work, but that’s localization; if there’s a joke in Japanese, it needs a joke in English. A bad joke for a bad joke, a good joke for a good joke.

Don’t believe that Japan inundated their script with memes? Here’s a link to a JP wiki by Japanese gamers that was used as a reference point for the translator on Neptunia Victory: all you need to do on there is ctrl+f “2ch” and there are your Japanese memes that originated from that online message board. Other memes exist on that wiki, but they may have originated elsewhere; the 2ch search is the fastest way of showing that memes were used in the US because memes were used in Japan. Here’s a link to the original game’s wiki, also.

I agree that using memes in retail titles is dating. When you also consider that the time the game is received by the US staff and the time of the game’s release are nearly one year apart, the game is actually dated by the time it comes out. Not to mention when a game is re-released several years later with much the same script (hint hint). One major thing to keep in mind, not just with me, but with translation teams everywhere—we are not supposed to write ourselves, our opinions, or our feelings into the games we work on. But that doesn’t mean we don’t fail; everyone has done this to some degree, even if it’s as internalized as word choice or syntax. There is no 1:1 translation because everyone has a different vocabulary or will piece sentences/ideas together differently. And because I’m not a translator, I have no idea how sincere or direct the translator’s translation is, either. My point is that I don’t like using references as humor and I’m trying to write more original content in my own game; there are movie and game references, but not as punchlines. When criticizing or critiquing the work others have done, try to keep in mind that they may personally agree with you, but they also have to do their job and have to do their best at it.

What is it you feel when you decide to insert a Bill Clinton joke, for example, in a game project when said person won't be relevant to culture knowledge 4 years from that game's release?

I think I answered this above to some degree. I wouldn’t do this for the exact reasons you outlined in the question. Is it different when a Japanese game references a comedy duo who haven’t been relevant in years and you have to use an equally outdated comedian reference in English (since that was the localization process used as described above)? What if that game takes place in a different fantasy world where Japan doesn’t exist?

That last question, in particular, is what I believe in. Don’t make jokes in a game that aren’t affiliated with that game world or timeline. I remember when I was doing some spot-checking work for Ar Tonelico Qoga (I had to help with a few events because the team working on it was falling behind—it was massive in terms of text), the translator and editor described a character as “being Spartan in her methods.” I asked them, “Why would you say Spartan when Sparta never existed in this world? They wouldn’t even have this word.” But it was ultimately their call, and that’s just how it is sometimes.

I know the Bill Clinton joke is probably from Lunar: SSS, courtesy of Working Designs. WD ascribed to a similar style as the places I’ve worked, and I agree with parts and strongly disagree with others. I found the localization of that game very charming and fun as a kid, but having grown up and seeing the process, I can say that they took it too far with political humor from a world not the game’s own. Unless, of course, Japan agreed with the style choice—this was the case with Neptunia, actually. Our tone and style were very well-received by the developer (yep, they have fluent English staffers), so we were urged to continue on down that road.

Why were some characters, like Neptune in Neptunia Re;Birth 2 and 3, altered in terms of how they speak? How come some characters have completely re-written dialogue when sometimes a basic straightforward translation seems the easiest, like was shown in a fanmade Re;Birth 2 retranslation on PC? Is that really what the audience for a game like Neptunia wants?

Three questions; I’ll answer in order.

First! Altered how, and compared to what? If you say Re;Birth1, then I would remind you thatfourother Neptunia titles had already been released with a set tone, style, and characterizations. After the release of RB1, the Japanese developer actually reached out to their new US branch and told them to ask me back because they noticed the changes in that game versus the prior titles.

Also, Re;Birth2 and Re;Birth3 were remakes of Neptunia mk2 and Neptunia Victory, both of which I had worked on and were released years prior. What makes this interesting is that the RB games were released by a different publisher than mk2 and Victory, but in Japanese, the scripts were word-for-word the exact same. I bring this up because when I was asked to work on RB2 and RB3, I was asked to rework the text. I imagine this is because, legally speaking, one company cannot use the same text that another company paid the work for. Thus, I ended up doing double the work when Japan only had to do it once. If you think the text changed between, say, mk2 and RB2, then you’re right. And that’s why.

A final thing to keep in mind is that, in the company, we didn’t know which lines were going to be voiced and which would need to get chopped from the English dub until way later on. While it seems like a completely separate issue, there are editing and writing techniques that help bring character personalities out more, whereas those techniques would not be needed in dubbed scenes thanks to intonation and delivery (sarcasm, for instance, is best used only in dubbed scenes). We would use those techniques throughout because of the uncertainty, and it would sometimes result in lines being long-winded or sentences getting swapped for clarity between languages.

Second! Take all the information you’ve been given so far and come to your own conclusion. I didn’t play the fan translation, but I have seen the translations that I was given from translators for games and quality of English is the main concern (breadth of vocabulary, for instance, is not a concern in very, very direct translations). If those fans are serious, they should open up a business and see if they can start to shift these ideologies held by the other companies existing today. Different styles can’t hurt; I always felt that a lot of the niche publishers have a very “samey” style to them, and I actually think it’s because everyone’s in the same area. Everyone talks to one another, makes in-jokes, and has a unique writing style local to that area. If there were a company in, say, Massachusetts doing the same thing, just by nature of where the employees live the text would read different. I’d be curious about the process used by these fan translators, because I guarantee you it’s different from what we had to do inside the company. For instance, we hardly ever were given the time to play the game prior to working on it (I’d take the “test kit” PS3 home and play them on my own time; why I'd do this is a story for a different day). If we were given the time to do line-by-line work as a team, where we watch the scene play out in the game and base the character emotions/expressions from there, I bet some lines would be different. But it was a business, and they have timetables and schedules, and it had to be completed in short order to keep the margins higher.

Third! This is going to be a bit of cold truth. Sometimes, companies don’t care what you want, especially the English-speaking audience. They don’t care when they design the games knowing ahead of time that certain mini-games might have to be removed for a US release. They make it known right out of the gate that they make their games for Japan only and it’s up to the publishers to get it released overseas. So a lot of the heat falls on them while Japan can shrug it off; I don’t have an opinion on whether that’s right or wrong, it’s just how it is. From a B2B perspective, the developer wants to license their game to the publisher who a) offers the best deal for profit potential, and b) promises to give their game the royal treatment and make its localization top priority (or give its release quality Limited Edition versions). You may not like the result, but B2B, sometimes the deal is to focus on making the text as strong as possible. Based on company policy, that may or may not allow leeway into liberal translation and editing. So I’m really sorry to reiterate, but on the backend you guys are not on the contract as points of interest. Money, quality, and turnaround time are. G-Generally speaking. I was never allowed to see these contracts, and even if I did, they were in Japanese and I can’t read them. It’s what was reinforced through working at a company and being friends with the guys whodosee the contracts.

Isn't a comprehensive dialogue rewrite just more work for negligible or negative gain?

See the answer to the third question above. It might be as you say, but that’s part of the wheeling and dealing of business. If you want an example of a company who took this idea to heart, boot up Lux Pain or Arc Rise Fantasia. Excellent games that were given such shoddy text work that it actually hurt them in reviews, in sales, and in reputation. That company no longer works in localization because, based on experience with the business side of things, I reckon that no Japanese company would license their work out to them because of that shoddy quality and reputation.

Do you think your translations do justice to the original script? Can any translation ever do full justice to the original work?

Well, first off, I never translated. It’s the translator’s job, after all. That said, yes and no, considering Japan’s response and even requesting me to return for later titles despite starting my own company. I know that, internally, I did a good job and that’s why I kept being slotted for Neptunia titles. Reviewers enjoyed our work and Japan liked what we were doing—some tiny tweaks were even made to characters on their end to help the JP and US characterizations align between the original game and mk2. It was a bit of an instant reboot, since the first title was supposed to be the only one (one piece of evidence is its true ending).

I’ve always loved this quote from RPGamer’s review of the original Hyperdimension Neptunia, and I think it’s part of why we continued down the localization path that we had created:

“It's honestly rather shocking how distinctly disparate the localization and actual game are in terms of quality. NISA really outdid themselves with Hyperdimension Neptunia, as it is an extremely verbose, story-heavy game, and it's sad that Compile Heart didn't live up to their end of the bargain. In the end, it's difficult to recommend staying away from Hyperdimension Neptunia, because despite its flaws, it may just be the most humorous title NISA's ever released. If you do decide to brave the world of Gamindustri, at least now you'll know what you're getting into.” —RPGamer review

That said, the “no” part is in response to any game giving full justice to the original work. As I’ve stated above, everyone’s unique, so no script can be the “same” or give “justice” to the original writer’s script because they are two (or more!) different people. The Japanese script is often not perfect, and likewise the English script is rarely perfect (my grammatical errors in the original Neptunia are cringey; I mixed British English rules with Chicago rules with AP rules all over the place). What you can hope for is a solid interpretation that has you smile when a Japanese gamer smiled, or groan when a Japanese gamer groaned. Good for good, bad for bad, all you can strive for is bringing the spirit and emotion across; this often quickly gets you away from a word-for-word straight translation. And when titles are worked on by more than two people, well, did you ever play the game Telephone in school? There’s a reason for it… It’s true and it happens even to adults.

Who is the best CPU, and why is it Noire?

Not even asking about the CPU Candidates?! I can’t say Nepgear, then… Oh, and like it or loathe it, people really seemed to embrace her “what the goodness” quirk and I’m super happy about that. Little liberal changes like that can help to bring a character’s personality to life in a way that copula and such can in Japanese. Nepgear is so cute.

As for best CPU… It’s strange, my thoughts change based on the game. During Neptunia Producing Perfection, I was able to see so, so many more interactions with Vert and how she lives her life that I was definitely her biggest fan in that game.

In general, I like Neptune. I especially enjoyed the dynamic between Neptune and Plutia (not Iris Heart, just Plutie).

B-But of course, I mean, Noire is the greatest because she’s the greatest. She only does everything (ugh, but it’s such an easy game joke to make).

Is using translator's notes to explain difficult/impossible to translate concepts ever acceptable?

I assume you mean placing them in the game itself. In anime, I absolutely approve of this concept, especially when the anime takes place in Japan or a specific period of Japan (or heck, something like Samurai Girls where the names are so obtuse that it’s practically required). In games, it’s not that easy. Let me explain.

When we are given a game to localize, we have a series of preexisting documents with all of the Japanese in them. That’s it. A translator puts the English translation in the column next to the Japanese, the editor puts in an edited form of the English translation in the column next to that, it gets checked over internally by the people working on it and then in a “script party” with the rest of the localization staff, and then the team sends the files back to Japan for approval and that is the extent of localization’s role. Putting in notes would either a) diminish an already painstakingly small character count per line (fewer characters than the average tweet!) to place notes in the text box with the dialogue, or b) require programming a whole new subsection of the game, which will need its own glossary, graphic UI elements, activation triggers, and more. Localization companies often defer programming needs to the original developer because developers hate giving out their source code (I assume, as a dev), so this is pretty much an impossibility in games unless there is a big budget and lots of time (hint: there is never either of those).

That reminds me… About changes from JP to English. Take a 3-line text box and fill it with kanji. Fill it to the max, as many of Histoire’s lines will do. Now translate it. Now keep all of the ideas presented in that line, along with key terminologies, and file it down to under 120 characters, with a limit of 40 per line (font and type size are determined by the developer). It can be pretty tough! You guys, sometimes it’s just not possible to keep everything. We’re not allowed to “just add another line/text box.” If we were, maybe I’d sing a different tune, but we weren’t.

Now that we've entered a dark age of censorship masquerading as localization, is there any hope at all for people who just want to play games in a language they understand without any further changes? Square Enix, Nintendo of America, and even NIS America have made it clear that they want to appeal to wider audiences. While this probably isn't the intended effect, what this tells the niche [small, but profitable] audience that has always loved their games that they are no longer the target market. Do you understand or agree with the backlash they've caused?

“Masquerading as localization” is a bit of a loaded statement… So many other factors are at play, but I digress.

Companies want to grow, and to grow they need a larger cut of the market. I guess this is how they’ve decided to tackle that issue. Especially given that niche games from Japan that these small publishers can afford are often filled with “questionable” content, they’re left with little choice. What you need to understand is that these companies are very, very risk averse. I mean, fear of lawsuits because we used “Neptune” instead of “Neptunia”? What else do you need to hear to know that risk is the last thing most of these places want to take?

And before someone says Senran Kagura (which I love), know that sometimes game localizations aren’t up to the US publishers. “HQ titles” as we called them are either pushed through by force or used as a threat: “well, if our US branch doesn’t want HQ’s game, then maybe we’ll start shopping ALL of our titles to different companies and phase you out…” I’m not saying they didn’t want Senran Kagura; several publishers did, actually. It’s to illustrate the point that not all risqué titles are brought over “despite the risk.”

How's the development of Undead Darlings going? Who is best Zombie waifu?

Thank you so much for asking! Development is SO HARD, you guys. I had braced myself for a difficult path, but this is like Dante Must Die Turbo Mode while wearing a blindfold difficult. Not to lament my situation, but I work from home and talk to contract workers via phone or messenger, I have paid myself nothing for over a year, and battle severe depression regularly. Haha, I’m not playing the depression card—I’ve come to realize why so many indie game talents cite depression as a thing happening to them. It’s an incredibly stressful and trying job, building a company and a game from scratch…but that’s what makes it so rewarding. This is a game that exists because of my actions and a handful of others; it would never have been born otherwise. It’s kind of neat.

As for development, it’s going to be a meaty game. I’m estimating the same amount of playtime or content as Demon Gaze, with the visual novel section and the dungeon crawling section reversed in terms of importance. Over 300,000 words and climbing—for reference, Hyperdimension Neptunia Victory was something like 180,000 words.

The dialogue script is nearly complete and all other aspects are being worked on. The biggest challenge is getting our dungeon graphics to look acceptable and not too terribly bland or empty. I’m hoping that some clever crowdfunding tiers can aid in this effort—like giving backers the chance to put their signature on a wall as graffiti. It will make every wall unique and bring more life to the game’s dungeons for sure. I’m loving the music, the art for enemies and the girls (and Buck), and I guess I have to say I enjoy the dialogue. Kind of biased.

The zombie waifu of your dreams awaits. Who that is will depend on your tastes! I’ve always been a sucker for the childhood friends, so my default thought is Pearl. I’m a huge fan of each girl though, because each carries a backstory with tragic elements that not only really bring their personalities to life, but also highlight how/when/why they turned undead and all the exciting subtext that can be gleaned from there. I didn’t want it to seem like one of those basic “it’s a visual novel…but with cardboard boxes!!!” projects, so I’ve worked really hard to give these girls an entire life that existed before the game begins and tie it into their current situation.

I know not everyone here is a fan of mine, but I really do believe that if you go into the game with an open mind, you’ll click with someone as well. You can still totally not like me, but these girls deserve a shot. Plus, what better way to criticize my work than to use something 100% original as a base to help me improve?

With reactions to your previous works, do you regret any directions you've taken? What have you learned, and will you do anything differently (now that you have more creative control) going forward?

Regret is part and parcel of growth. While it would be easy to hide behind a company motto or process, I do take responsibility for every project I’ve worked on and I used different styles based on the game’s tone and setting. Sadly, most of the projects that came my way were of the same type (comedy), so I didn’t get many chances to really stretch and take on a game that would demand a more serious and precise localization. Comedy is the most volatile of genres when translated, which is why you can spy changes with relative ease versus something like, say, Beyond the Beyond (protip: don’t play it).

That said, on a macro scale I don’t really regret the things I’ve done because they’ve taken me to where I am today. On a micro scale, yeeeeah there are some lines that I wish I did differently or that I wish other team members would’ve been like “hey man, that’s not really what it says” (this did happen, but usually it was an actual mistake versus a liberal localization choice).

Since I’m not involved with localization at the moment, the things I’ve learned and the changes I’d enact in a localization environment may never be seen, but now I’m confident I’d do an ever better job at something everyone could enjoy. I have the creative outlet to make something of my own, which helps to keep those urges at bay should a localization project come up. There is a time and a place for everything, and from all this, I’ve better learned to identify when it’s okay and when it’s not my call.