in·ter·view / n. 1. Oral examination of an applicant. 2. Conversation with a reporter, for a broadcast or publication. 3. Meeting face to face, esp. for consultation. v. Have an interview with. [MFr: entrevoir, glimpse]


The following is an interview conducted via e-mail between myself and Richard Honeywood, a senior member of the Square Enix localization team. Mr. Honeywood has worked in some fashion on nearly every major Squaresoft/Square Enix game since 1998's SaGa Frontier, ranging from technical programming to overseeing the entire localization effort. I would like to thank Robert Rork for setting up the interview, and Kyoko Yamashita for handling the PR side of things.

Mr. Honeywood has provided me with a translation of a recent interview he gave to a South American magazine, which concentrates on his work on Final Fantasy XI. Because I didn't touch on these aspects in my own interview, I've posted it here as well in the hopes that FFCompendium readers would find it interesting. Read it here.


Q. In a typical big-budget game such as a Final Fantasy game, what sorts of jobs would be included in localization, and how would they interact with each other?

A. Depending on its size, each title usually has:

  1. One or two Localization Coordinators/Directors.
    They are responsible for the management of the whole localization project including budgets, schedules, staff assignments, communication between the localization department and the development, marketing, distribution, QA teams, etc.
  2. One or two Localization Assistants.
    They assist the coordinators and translators with any odd jobs that come up in the localization process. They may start off with familiarization and the writing of walkthroughs of an early (pre-release) version of the game. Then they help manage, convert, and check the translated files. They also help verify bugs and ensure those bugs are fixed, helping the coordinators to provide feedback to each language's QA department.
  3. Localization Specialists (Translators).
    Numbers can vary from one to four or five translators for each target language. We port most of our games into American English, British English (edited for PAL territories), French, German, Spanish, Italian, and occasionally Korean and Chinese. The localization specialists have to completely familiarize themselves with the game before commencing translation. We try to translate directly from Japanese to each language wherever possible, but we sometimes translate J->E, then E->FIGS. One translator will be selected as lead, and is responsible for communication and decision-making for that language. Once translation is complete, the translators remain for the QA process, making any changes necessary until all is ready for release.

    Our translators like to be called localization specialists as they do much more than just translate. They help write and record songs and spoken dialogue for the games. Often they have to arrange for changes in the game to better suit their culture (e.g. not just rewriting scenes' texts, but changing graphics, controls, or even levels of difficulty in some cases).

  4. Editors.
    As the name implies, they edit the translated text, fixing any language issues before the game is sent to the QA department. They also help to maintain a standard Square Enix style across all our titles. We almost always have one English editor assigned to each project.
  5. Optionally, a Localization Engineer (Programmer).
    These assist the development teams with any additional programming for the game that the original Japanese programmers cannot do. Recently it is more of an advisory role, as the original development teams have become more accustomed to the localization process. But over the years, the localization engineer's role has varied from making simple line checker or converter tools to taking over the whole source code and performing the programming conversions internally within our localization department.
  6. Quality Assurance Staff
    We usually have two teams per language. One team will consist of Japanese staff here in Tokyo who look for technical and gameplay issues. The other team will be either in our L.A. office or London office, and is made up of native speakers for that language. They focus mainly on language and cultural issues. (But you'll also have the US and UK staff finding occasional technical issues that the Japanese QA staff missed, and more surprisingly, Japanese testers reporting spelling mistakes or inconsistencies for a language they can't even speak!)

    On top of the time difference between the three continents, during crunch time in Tokyo we can even have three large teams each working eight-hour shifts to perform around-the-clock QA on our larger titles.

Q. How do you think the job of localization has changed over the years since you started working in it, both in terms of technological and societal/sociological factors?

A. Well, when I first joined Square about eight years ago, localization was an afterthought. No one expected the games to sell very well, and the foreign language versions were done on the cheap to gain a little bit of pocket money while the team prepared to move on to their next title. There was only one person serving as a go-between with America at the time, and no in-house translators in Tokyo.

A single translator (in our US office, or outsourced for European languages) would be given fixed fonts and restrictive letter limitations (due to on-screen layout and memory concerns) and asked to rush out a translation with limited checking and rarely any changes. The teams even forced the translators to write European languages in double-byte Japanese fonts, replacing European letters with obscure kanji. This wasn't because they were lazy; they just didn't know how to handle European letters.

FFVII was probably the turning point, as it showed the company that foreign languages could sell huge numbers. I joined just as FFVII was being released in the US and we were gearing up to make a PC port in the US as well.

Over the years, we have increased the in-house localization staff in Tokyo from just two to over thirty members. We also have established offices in Europe and Asia and have become our own publishing house in North America.

We have worked hard to gain the teams' trust and cooperation. This has allowed us to take small steps to innovate, starting from (*gasp!*) proportional fonts, support for European letters, and menu layout changes, all the way to original content for foreign versions. Lately I have worked on a system to allow singular/plural item names with proper articles--that means no more "Got 2x potion!" Now the system can automatically change grammar to match the substituted item names to display "Received an ax"/"Received two axes," or "Obtained a potion," etc. This might sound simple, but teaching Japanese programmers (who don't have articles or singular/plural in their language) how to accomplish this along with even more obscure European language grammar has been an ongoing challenge.

Now almost all the development teams assume they will be making foreign versions of their games, and localization is called in at earlier and earlier stages. Some games, like FINAL FANTASY XI, were created in English and Japanese simultaneously. Others, like The Bouncer, were even created in English first.

Our company has seen that effort pay off as the international fan base has boomed--and our overseas sales now make up over a third of our profits. On the other hand, players have seen the quality of our titles improve in leaps and bounds to the point where our storytelling is on the par with literature in their own languages. Sometimes we receive backhanded compliments as people lament the loss of the silly mistranslations that were the norm in the Eighties. The better our work, the less people notice that the game is a translation from Japanese. It should feel like it was originally created in the player's own language.

Q. What do you think is the most challenging part of working in localization?

A. There are many challenges that aren't so obvious.

So, if ever you notice something strange in one of our games, instead of criticizing that one little slip-up, look back over the quality of the rest of the game and give a thought to the localization staff who worked so hard to make the game as best as they could given the restrictions they had at the time.

Q. What was your favorite (and least favorite) game to localize, and why?

A. Oh, gee... Tough question.

I have been lucky in that I've loved all the games I have been asked to work on. I was always a Square and Enix fan, so it's been a privilege to serve with the development teams and an honor to work on their games. So I have no favorite or least favorite game per se.

That said, I can tell you which project was the hardest going for me personally. It was "Xenogears." The game was ambitious even for Japan. It was the first major title I had to manage and translate myself. Because of its controversial content and the linguistic and conceptual challenges it presented, the original translators assigned to it quit or asked to be assigned to other titles. When it went over schedule, I ended up having to not only direct, but translate and program as well. (Heck, I even burned the master disks!) The team basically left it in my hands as they went on to their next game. I worked around the clock, sleeping in the office for months to bring it to a shippable state. (At the same time, I had trouble with my own religion when the elders heard about the content of the Japanese version.) As a translator, I wanted to respect the game's creators and keep the content as close as possible to the original. Even the non-controversial parts were hard to translate-- all those scientific concepts and philosophies. I look back and wonder how we ever finished it. I guess my naivety at the time was a blessing in disguise. If I knew then what I know now, it would have been a totally different game.

At the same time, Xenogears was a catalyst for changes as it showed what changes were needed. After that, we added assistants and editors to improve the quality. It was also the first game where the localization members sat in with the development team and worked with them. After that, I thought the Xenogears team would consider me a failure and wouldn't want to work with me again. Instead, it seems I gained their respect, as they asked me back for both Chrono Cross and FFXI, and they have been more cooperative and enthusiastic with each project we localize.

On the other hand, although it has been the toughest game since Xenogears, I'd say FFXI is currently my favorite. I am proud of what we have achieved. Despite sharing the exact same servers as the Japanese, I think the game itself feels more natural than any other game we have localized. I have worked on FFXI's localization for four years now--and the first two were by myself. I helped name people, places, and things as well as plan out the backgrounds for the whole world of Vana'diel along with the development team for both the Japanese and English versions of the game. It's the biggest, longest, and most challenging game we've ever worked on. Despite all the technological and logistical hurdles, it has been a greater international success than we imagined. So in that sense, it is my favorite title right now.

Q. It can be argued that the FF series, being an international juggernaut, tends to be more "internationally minded" than other, independent games. In other words, very little actual Japanese custom and idiom would be put into it in the first place. Games such as Star Ocean: Second Story, which are more "obviously Japanese" in many scenes, would represent a different sort of task. Have you ever worked on such an RPG, and how did you tackle it?

A. Phew! Thanks for saying they seem international to you. They all seem so Japanese to us! (Laugh)

The FF series didn't just become international-minded. We've had to work for it, building up good relations with the team so that we can check for content issues during the early planning and development stages. You should have seen some of the things we had them change. For example, the blitzball poses and religious gestures from FFX originally had characters putting their fist up over their other arm as if giving someone "the finger"! We had to request retakes at motion capture studios. Fullmetal Alchemist also had the main character counting to two on his fingers, but the way he held his fingers up looked very much like an "up yours" (reversed V-sign) in British culture. But we changed it at the storyboard stage well before it was animated.

To avoid wasted efforts, we have made internal homepages with our localization staff posing for motions that should be avoided by the dev teams. Also, the teams have learnt to run things-both motion and graphics-wise-by us before it is too late to change them. As a result, you don't see too much bowing and other Japanese motioning in recent titles by dev teams from Square's side. We also fix up a lot of the "Japanese English" that appears in the Japanese titles, as we know a lot of non-Japanese people will also play them.

If we can work around these in the translation process, then we try to respect the original team's wishes. (For example, when people gossip about someone, that person is said to sneeze in Japanese culture. In FFXI, a San d'Orian prince sneezes when he is gossiped about. We kept it in as a quirky scenario of the Elvaan culture in that game world.)

Sometimes we end up doing two takes of the scene. Look at Koh's gestures when disguised as a baddie in the elevator in The Bouncer. We did one scene that made sense for Westerners, and another take for the Japanese audience. Just switch language modes to see the difference.

In the online game FFXI, we allow Japanese festivals and celebrations to coexist with Western ones. So while Japanese players can now enjoy egg hunts and Halloween, North American players will experience cherry blossom viewing, Hina Matsuri (doll festivals) and Obon (Old Saint's Day masked as a summer dance fiesta where everyone gets dressed up in kimono and perform line dances). As localizers and game creators, we just have to explain these in a way so that they suit the game's world.

Speaking to my fellow staff, Brian Gray - who worked on FFX-2 - says he thinks that title is still very much Japanese when you think about it. Here's what he had to say:

Between Yuna's wrist-flapping run, Brother's anime-esque antics, and in-game J-pop musical numbers, FFX-2 is probably the most non-Western looking Final Fantasy approach to date. Everything, from the outfits the girls wear to the character animations to the downright bizarre subplots, is over the top in the Japanese version. To keep things balanced, we decided to write equally unpredictable and light-hearted English dialogue. In the end, American gamers might scratch their heads and say, "Is this for real?" during some of the more bizarre situations Yuna lands herself in, but the English dialogue never gets so heavy-handed that it forces the gamer to take what they're watching too seriously.

On the other hand, the reason Star Ocean seems more Japanese to you is probably because, being licensed by the Enix side of our company to an outsourced company, we don't have as much input into its development until localization actually starts. (By then, it is often too late to change things because of cost and time constraints.) But as our two companies' approaches to game development become standardized, I think you'll start to see the internationalization of the other titles occur more naturally.

Q. How often do you introduce cultural references into the English text of the game? "Macarena Temple" in FFX comes to mind, but are there others we might have missed? Do you think we'll be seeing more in the future?

A. Our policy is that cultural references are okay if they don't ruin the game's "world feel." They also have to be funny for the people who recognize them, without seeming out of place to people unfamiliar with what is being referenced. Or as Colin Williamson, one of our editors, puts it:

It shouldn't be glaringly obvious - subtle is better - and it must be 100% transparent to players who won't "get" the reference. And don't forget that a game's script can be locked down up to six months before it appears on shelves, and what may seem funny this week may be passé' by the time the game is released.

Usually we try to restrict them to be hidden features for people who look for them. Brian mentions:

In FINAL FANTASY X-2 there are quite a few FINAL FANTASY and other gamer in-jokes hidden throughout the game, but most are tucked away in impossibly hard subquests or seldom-viewed NPC dialogue. The "Macarena Temple" reference from FFX is famous because it's right smack in the middle of a mandatory event sequence, but I think generally we try to save the cultural references for gamers who stray off the beaten path.

Actually, I can still remember when Alex was writing that line. I'm always a sucker for silly humor, and I'm a pop culture junkie. Of course, I laughed and agreed to it at the time, so long as the "Aye!" was done in a way that it wouldn't seem really strange to people who didn't know the song (or years later, when everyone has forgotten the old one-hit wonders).

That said, I'm always making bizarre references in all of my games. From puns on product names like "Bartweiser Beer" and "Margearitas" to catch-phrases like "Guns don't kill people...people kill people!" in Xenogears. FFXI, having so much dialogue delivered by wacky Tarutaru, sports character names like "Detective Poiroto-Boiroto" and "Miss Marpelpel" (from the "Agatha Crystalie Whodunit Series"!), and a group of kiddie mystery-solvers known as the Star Onion Brigade (S.O.B.'s for short) throwing out lines from Scooby-Doo cartoons. (Also, we try to make the quest names read like punny newspaper headlines, with "Silencer of the Lambs," "Inspector's Gadget!" and "Vampire Hunter D-minus" being some of my sillier ones.)

As you can see, I give the legal department nightmares. I also tend to go over the top with my humor and need the editors, my fellow translators, and the QA department to keep me back in touch with reality. (The same goes for all our staff. For instance, we removed references to "Doublemint" ads from Chrono Cross because the twin-related jokes would only make sense to American players and not the rest of the English-speaking world. But we had Nikki sing a real rock song's lyrics at one point, as we knew it was a classic song most people would remember.)

For further reference, Colin, who had edited Sword of Mana, says:

I try to shy away from cultural references in favor of subtle references to other Square-Enix games. In Sword of Mana, there are about a dozen gags that pay homage to the original English FINAL FANTASY ADVENTURE (Seiken Densetsu) - we tracked down a copy of the original script, and tried to work in as many references without making them too obvious (or weird). I haven't seen anyone mention them, so maybe they blended in a bit too well (laugh).

As far as other titles go, FINAL FANTASY XI has a few certain "spoony" running jokes if you look hard enough, and several of the more light-hearted upcoming titles have 'em by the boatload…if you know where to look and are relatively in tune with gaming culture!

So the answer is: yes, yes, and yes. We often make cultural references, we're sure you've never caught all of them, and you can expect many more to come!

Q. Are you ever worried about the reaction from fans about you "messing up" or otherwise straying from the vision of the Japanese version of their beloved games? Have you ever regretted making particular decisions in localization after the fact due to this sort of thing?

A. As Peter Jackson said about his Lord of the Rings movies, he didn't worry about people saying he ruined the books, as the books will always be there. The same goes for our games...the originals still exist for people who speak Japanese.

Some people may disagree about our interpretation of a game, but we are not out to make a literal translation. Our job as localization staff is to make the original work feel as natural and accessible as possible for the widest audience in our target language. Naturally, the only people who can say that we "messed up" a translation are bilingual people...but being bilingual gives them a different insight that most other people don't have. It might sound high and mighty to bash us for changing certain content, but there are many reasons why we can't do a literal translation at times.

Brian also had a similar comment:
It's kind of hard to say what's "messing up" and what isn't. There are certainly gamers out there who respect that our games are Japanese products. These gamers would probably prefer we translate everything literally and leave the interpretation up to them. But I think the majority of gamers out there just want to play the games and have fun, without wading through all kinds of linguistic barriers just to figure out what's going on. That's why localization is more than just translation. If something seems like it would sound clunky or confusing when translated literally, we change it in the hopes of achieving something more natural. I don't personally regret these changes. Localization doesn't "destroy" the original Japanese product, which will always be there. It simply is a compromise between faithfulness to the original and accessibility to the largest group of gamers possible.

As a simple example of things that would not translate, take Chocobo Racing. In the Japanese version's "Hungry Land" stage ("Gingerbread Land" in English), where the FF characters dress up as Japanese folktale heroes Momotaro (lit. "Peach Boy") and Kiji ("Pheasant") and say a few silly lines, we turned them into "Hansel and Gretel" for the English version. Purists could argue that we should have kept it as "Hungry Land" and kept the bizarre reference to Momotaro, but when your target audience is children, it's obviously better to use something more familiar to them (and it fit the setting even more than ol' Peach Boy did!).

Colin also added some interesting points:
We're well aware that people will take issue with certain changes. Said changes can be made for the sake of clarity, ESRB ratings, and the fact that some stuff simply makes zero sense in English. References that hardcore Japanophiles may find gut-bustingly funny will make non-Japanophiles (the vast majority of the buying public) scratch their heads in confusion. If "messing up" means changing inherently Japanese content to something similarly attractive, understandable, or funny to a casual English-speaking game player, then we'll take that bullet. Localization should not be straight, no-frills translation.

The teams are also very much in tune with our work. It's not just a matter of dropping off a few megs of text files at the programmer's cubicle and picking up a finished ROM an hour later. The team keeps tabs on us and takes an active role in localization, and if we're breaking their vision, they'll really let us have it. (Laugh) If you see a change that seems really drastic, it's probably been deemed kosher by the director. The people in charge understand the culture gap and will change things to the best of their ability.

Actually, as a side point, sometimes the English versions are better than the Japanese versions, as we can remove bugs that were found after the mastering of the Japanese (such as those reported by Japanese players). Also, some staff members like to add or change things for the foreign versions. In Chrono Cross, for example, the scriptwriter Kato-san agreed that some of his explanations of background events and people's connections weren't very clear, so we added extra text for the English version. For instance, the relationship between Korcha and his "adopted" sister Mel wasn't explained clearly, so Japanese people were creeped out by the fact that Mel had a crush on her brother. Also, the reasons why there were two moons in Chrono Cross (when Chrono Trigger's world had only one), and whether Belthasar planned the Time Crash or not, were not explained very well either. So, Kato-san and I rewrote parts for the North American version.

These close relationships with the dev staff who sit around us on each project are really valuable. They do look over our English (and other languages) and comment on things. Some staff are adamant that we do not change certain things, while others trust us to the point where we don't have to inform them of changes (even though I still do out of courtesy and respect for their work).

I can't think of any regrets I've had about changes I have made personally, as I always did the best I could at the time. Sometimes I wish the team allowed me to change one minor thing or another, as it annoyed the perfectionist in me. But I respect the team's wishes, and as a director myself, I realize that time and money also need to be factored in to things.

Q. What are your thoughts on online fan translations such as the one done for FF3? Have you ever seen one, and if so, what did you think of its quality, considering that the translators have no input from the original programmers and often have to "reverse engineer" compression schemes etc.?

A. It's hard to answer this question because of the legal issues involved.

Of course we do view these, as we are interested in seeing how people render our work. Sometimes I am amazed by the lengths people go to. As you said, the text in the old games was compressed, so programmers had to uncompress and recompress the text in order to change it. They also have the same memory restrictions we had, so we are often impressed by the efforts. Technically, it is an amazing feat. (The most amazing one I saw was a Russian version of FFVIII that came out just weeks after the PAL release. Not only was the font replaced with a Russian font, but the translation was quite well done given the memory and time restrictions. Unfortunately, they were illegally selling this, so they had to be stopped.)

Other fan sites do unofficial translations-whether inside ROM images or just as text- but don't realize that their work may be seen as "weakening our brand" or "ruining our characters." Besides pirating taking away sales of the official translations (thus causing our localization efforts to not pay off), if their unofficial translations are weak, people may conclude that the original games were rubbish, affecting the sales of the final game. So, best intentions aside, the distribution of fan translations cannot be viewed lightly.

We have had people send us their own translations of our games when applying for jobs here. This does indeed show an impressive level of dedication. However, I can remember one or two such people accidentally insult us by saying, "Because your original translation was crap, I created my own version of your game. So you can see for yourselves why you should employ me!" This not only showed a lack of knowledge of the other factors that go on at a company, but they were being rude to the people they were hoping would employ them...which isn't the best tactic. (Laugh)

Q. Do you think voice work has changed what localization needs to do? (For example, some of the lines from FF6, while classic, would sound terrible if spoken.) Do you think the team should be going further than it had in FFX? For example, eliminating unprovoked grunts like "Uh? Wha'? Ah!" that sound awkward to an English-speaker, or actually reprogramming scenes to sync the speaker to the words, similar to what was done in Kingdom Hearts?

A. Absolutely. There are many more factors involved now with voices. (And it is much more restrictive...as Brian attests to in his answer below.)

But written dialogue sounding strange when voiced is not a problem exclusive to video games. Many books that read well don't flow when acted out, and even when you look at the dialogue on old TV shows (say out of the 70s) the dialogue can seem dated and weird. Add in the memory and space restrictions we had in those old games and it's no wonder they sound odd when read aloud.

I think Xenogears was the first Square game to have voice-overs localized, and boy, have we learned a lot since then. People often complain that the lips don't match, but they don't realize that Japanese games and cartoons very rarely lip-sync. (Japanese cartoons like Doraemon just have two or three mouth flap cels that never match what is being said, and no one cares. I guess the Japanese are so used to watching dubbed movies here that they don't think such details matter much.) So if the original Japanese lip movements don't match the dialogue, there is little chance of us fixing it in English (although the guys who did FFX and FFX-2 sure tried where they could).

Fortunately, some dev teams have realized that lip-syncing does matter for the Western audience. For instance, we decided to record The Bouncer's voices in English first, using two cameras placed in front and to the side of each actor to record the lip movements and expressions. After the English lip movements were rendered, we then dubbed the Japanese version to match the English. Tokita-san's team deserves full credit for taking such a revolutionary step within the company. Other recent projects I have worked on, like FFXI's opening movie, followed this lead.

The Kingdom Hearts team did a fair effort of redoing the lip-syncing for English and European language releases. But it is labor-intensive and takes the team's time away from creating their next game. Until we work out a better system to automate the process, it's more out of love for the foreign versions of our own game than out of necessity (as it doesn't make much economic sense either).

So we are always working on improving things, and I'm sure you will start to see the bad lip-syncing and unnatural reactions improve with time as all the teams become accustomed to working with the relatively new media and the localization of it as well.

Brian Gray (one of the translators of FFX-2) also chipped in the following answer:
Yes. Before, localizing dialogue was pretty much just a matter of replacing Japanese text with English text. Character motion on-screen tended to not be very complex, so you could write just about anything and it would probably match what's going on visually. Lately, though, with voiced dialogue and more cinematic event sequences, we have to think about other things. What are the characters doing? Who are they facing? How far away are they from whoever they're talking to? How long a pause is there between the lines? All of these affect how believable a given translation will be. This isn't just a matter of localization, either. The Japanese development team also needs to leave enough flexibility in their design so that we can fix unnatural pauses, grunts, or gestures. Overall, we're getting better at coordinating this. Kingdom Hearts was a positive step forward, and Final Fantasy XII and our other coming releases will look even better.

And Colin, our in-house editor at Tokyo HQ, had this to say:
Voice work changes the ebb and flow of things enormously. Generally speaking, you get a lot more freedom in your translation if you know that your dialogue is only going to be displayed as text; your main limitation is what you can display on the screen. With voice, you generally have to match the exact length of the original Japanese line, and in many cases, have to match the lip flaps of the on-screen characters. I'd much prefer writing spoken dialogue for a character whose mouth is obscured by a mask or armor!

Fortunately, we're seeing more cases where the team will re-animate for the English dialogue, but that does consume both time and money - just imagine how many hours of cutscenes there are in the typical voiced RPG.

And as for the monosyllabic grunts, it's an inherent limitation of the system. For most in-game scenes, voice lines are played sequentially with a half-second pause before and after each line (or a request for a keypress). That in itself prevents any rapid-fire exchanges or dialogue overlap - and it makes the editor or ADR person's job all the more difficult. We have to ensure everything is going to flow together, even with those "uncomfortable silences." Of course, we could remove the grunts entirely, but that would mean having to re-script, remove lines, re-animate things... Not easy in an event- and timing-critical game environment. You're bound to break something along the way!

Q.Is there a policy regarding translations from previous FFs (for example, the changing of Gurgu Volcano from FF1 into Mount Gulug in FF9)? Is there a desire at all to keep continuity alive in that sense?

A. Naturally, we strive to maintain consistency. However, over the years each FF game has come to be translated by different people, so naturally mistakes or deliberate changes have occurred. Whenever any mistakes or discrepancies are found, we try to correct them in the next game they appear in. (e.g. "Vicks & Wedge" was actually meant to have been "Biggs & Wedge"...a cultural reference to pilots in Star Wars. Unfortunately, the original translators missed the reference and weren't able to check with the authors. We changed it once we realized...but the discovery only came after we built a good relation with the dev team and were able to discuss such details with them.)

Other changes come from the progression of the game system. Originally we were given strict length restrictions, so the spell names in FF1 were four or five letters long. Later we were able to spell "Thunder" or "Lightning" instead of LIT1. From FFVIII onwards, we changed the spell naming convention to match the made-up Japanese spell names, with -ra and -ga endings, etc. We thought it improved the game's own original world-feel. For FFXI, we had to further adjust both the Japanese and English spell naming system so that both languages could use the same words. I spent weeks making different versions of the names with the director, Ishii-san, until we found a happy medium for both languages. During those discussions, I also found out we had translated the spell name DIA in Japanese (HARM in the original English FF1) wrong. DIA actually was an abbreviation of "Dispel Undead" (the "u" in "undead" is written as "a" in Japanese). "Diu" or "Diun" didn't work well in English, but seeing as "DIA" connotates "daylight" in Latin-based languages and undead are usually harmed by sunlight, I decided to use "Dia" in FFXI, and had the translators fix it in the remake of FF1 onwards as well.

But the classics of all time are the summon monsters. The dev teams even change their group name each game. So we've had them called "Summons," "Espers," "Guardian Forces," "Eidolons," "Aeons," and "Avatars." It's almost become a custom in localization to think up a new name for these each time. I just don't want to be around when we do FINAL FANTASY XXVII! *Laugh* Overall, we try to have our translation staff play the previous games in any series they are on. Wherever we can maintain naming conventions, we do. However, in the case that it no longer suits the new game, or in the case of mistakes that we discover later on down the track, then we also give the translators the freedom to fix whatever they deem necessary.

Seeing as your question is about the original FF, I asked the translator for the upcoming FFI&II GBA remake, Joe Reeder, to give us his thoughts:

There's no pre-determined policy about maintaining continuity in translation. Of course, wherever possible we try to use the established translations, unless there is some compelling reason to do otherwise. A good example might be Samantha Soul/Soul of Thamasa. Samantha Soul appeared as an item in FFVIII and an armor in FFX. However, when we began working on FFX-2, it was clear that it had been mistranslated in earlier games (it was actually supposed to be referring to Thamasa, a city in FFVI), so we adjusted the translation accordingly.

These outright translation errors occurred much more often in earlier games (often due to the extremely tight character limitations of the day), and we intend to correct these where possible in the upcoming FF1&2 Advance and other titles so players can more fully experience the text as it was intended in the original Japanese.

Cases where there is no strictly "correct" translation, such as Gurgu Volcano, are completely at the discretion of the translator; different translators have different feelings on whether such changes are appropriate. Personally, I prefer to see these remain the same.

Q. What's the funniest thing you've ever had to translate?

A. Each game has its crazy sections and we've all had to translate a lot of weird stuff over the years.

Koji says:
A rapping leech (don't ask...)

Colin says:
I noticed that the journal entries that Li'l Cactus keeps in Sword of Mana were a little dry--so one day, I took the initiative to rewrite all fifty of them in rhyme. Some of the puns I used in the limericks are enough to make a grown man cry in agony. Oh--I had done this late in the translation schedule, and the European translators were ready to have my head for making such a major (and time-consuming) change.

Phil says:
Doctor Shantotto...who is always insulting adventurers. (Windurst in FFXI) (Actually, her characterization was started by me, but Phil took it and ran with it. She also rhymes her sentences as well.)

As for me...
Chrono Cross had a lot of bizarre moments...like a rock concert musical done in Shakespearian prose aboard a ship, not to mention playing around with all those accents and speech impediments for the forty or so PCs. FFXI has the whole of Windurst, which I deliberately made over-the-top (as the Japanese was silly enough to begin with). I just love writing the Star Onion Brigade, as well as other Tarutaru who speak in Dr. Seuss-like rhymes.
Oh, and Mog, "Gobin Hood," and the other cast of characters from Chocobo Racing was a great deal of fun to play around with too.

Q.From your experience, what skills/educational backgrounds have you found most useful in your work at Square-Enix?

A.Programming ability, interest and experience in the gaming industry, and knowledge of Japan and other countries in general have really helped. I was originally studying Economics in the hopes of entering politics or becoming an ambassador, but my secret desire has always been videogames, so I took Computer Science and Japanese as minors. Looking ahead at the next three years of Economics, I realized it was just going deeper into conflicting theories and it was way too boring, so (unbeknownst to my family) I switched to a double major of Computer Science and Japanese. In my fourth year at Sydney University, I came back to Japan to study at Sydney's sister school, Hosei University--and to try to break into the Japanese game industry. Making the entry into the field was really hard, as most companies (including Square) would only consider me for translation positions, while I was aiming for the role of game programmer. I ended up taking a job at a lesser-known company to gain the experience to step up to a bigger company.

Without the solid experience of living in Japan, and language skills, I wouldn't have made it at all. As a gaijin, you really have to win the Japanese over to get anywhere.

Also, localization isn't just about North America. We deal with European and Asian languages on a daily basis as well. So we have to have a good knowledge of these cultures and a bit of understanding of their languages, too.

When translating, we need to have a wide knowledge of as many fields as possible, as you never know when mythology or religious ideas, scientific terms, or science fiction concepts will come up. Also, a solid grounding of pop culture in Japan and the West is a must as well. Finally, a good sense of humor will not only let you translate, but make the workload bearable.

Q. What sort of education/experience would best help people looking to work at a large video game company like Square-Enix?

A. It depends on which area you want to enter. Even in localization, there are many areas:

Koji Fox, one of my fellow translators on FFXI, suggests the following for translators:

I'd also add that you need to be able to write well in your native language. Remember, our localized games are competing with novels, movies, and other games made in the target country. In other words, it must be good enough to sell the game in that country.

When conducting interviews for translators, we look for previous experience in translating both individually and in groups. Also, translating games is so different from anything else, previous work on games is a great benefit. (I know it can be a chicken-and-egg type problem trying to gain that initial experience, but even if you have to start at the bottom doing something you hate at first, at least it is a start.) We don't actually have any set requirements for education, as some of the best people are self-taught. The more information and samples you give us, the easier it is for us to judge your abilities. Most of all, personality plays a great role. We have turned down some truly skilled translators simply because none of us could ever imagine working with that person. Our work is a team effort here, so we don't want to ruin the working environment.

For Localization Director or Coordinator positions, I'd also suggest:

For Localization Programmers:
At the bottom level, you should be able to program simple tools for converting data and line-checking. You should also have a good knowledge of European letter and Asian character encoding methods and how to implement these and convert between the two. You should also be able to explain the ins and outs of NTSC and PAL TV conversions. At a higher level, you should be able to program for the platforms and PC quite well. While you won't be expected to program a game from scratch, you will be expected to be able to take the team's source code, understand it, and make all necessary changes for up to seven language markets.

For a Localization Assistant's position:
My loyal assistant on FFXI, Seikoh Hokama, says:
Basic knowledge regarding computers and software. We use different kinds of software, from applications for computer graphics to program compilers to business software. Also, knowledge of HTML and CGI/Perl (common gateway interface) scripts is helpful sometimes.

You will start off with a lot of simple typing, counting, and cutting-and-pasting-type manual work. You will probably generalize into all three fields above, being able to help Localization Directors, translators, and programmers do their jobs. Eventually our assistants are expected to become good enough in one field to be promoted into taking that job. (So far, most have gone on to become Coordinators, but two or so are aiming for translator positions right now.)

For an Editor position:
Tokyo HQ's in-house editor, Colin Williamson, says:

Experience? I'm not an English major (I wound up with a bachelor's in Asian Studies), but I'd been contributing to print magazines from high school through college--I imagine that helped a little bit. Before joining Square I worked as a planner, FMV designer, and texture artist at another game company, where I cut my teeth on the horrors of development. Editing all day long seems like a vacation in comparison!

The ideal editor has exceptional English skills, can catch the tiniest of grammatical errors in a heartbeat, and works fast enough to to keep up with multiple Bible-sized scripts that are continually dumped on his or her head. Some game scripts may need little work, while others require full-on rewrites. The editor's biggest responsibility is script continuity; the game must be clear and succinct, and should have one "voice."

Punching up dry scripts with snappy dialogue and one-liners is also important--as long as the tone of the game allows it. I'd imagine familiarity with comics writing would be tremendously helpful as well--I grew up reading Carl Barks's "Uncle Scrooge" comics from the 1950's, and those influence my writing style more than anything else.

Oh, one last thing--you need tact! Remember, the editor is tearing down and rebuilding a lot of the translators' work, and proper justification ("I think it flows better if we lead in with this line") is preferred over uncensored criticism ("Richard, this line sucks!").

(BTW, did you spot the typographical error in my above reply?)

Q. Out of curiosity, how often outside of your work do you play video games?

A. Most staff members in localization are hardcore gamers. I generally spend most my non-working hours playing games, so I am worse than most people here. (But I usually multi-task...playing games while watching TV, listening to music, or...um...playing other games at the same time. *Laugh*) I'd average at least two or three hours each night, and a huge chunk of my weekends, playing all the major releases (to know what the competition is up to, as well as being a fanboy myself) as well as some of the quirkier games most people would never play. I think most of the people around me are the same.

What surprises me is how much we FFXI people get addicted to our own title. FFXI's translators, Az, Koji, and Phil, still play for several hours almost every night...even after working on the game for over 10 hours in the office. (I've had to wean myself off of it.) Most of the dev staff around us also play most nights of the week. It's kind of hard to get angry at the people who turn up to work late saying they had been up all night playing FFXI. *Laugh* Occupational hazard, I guess.

Q. Thank you very much for taking the time and effort to talk with us!

A. It has been a pleasure to do this Q&A session with you. I look forward to seeing the final results and hearing people's feedback. No one ever becomes famous for translating, but at least people might get some insight into what we go through in the localization business.