True Tales from Localization Hell

This article first appeared on USgamer, a partner publication of VG247. Some content, such as this article, has been migrated to VG247 for posterity after USgamer's closure - but it has not been edited or further vetted by the VG247 team.

Jeremy Blaustein

In 2001, Blaustein would face his greatest challenge as a localizer: Dragon Warrior VII, which would be the series' grand return to America after nearly a decade of absence. This English-language version, however, wouldn't be free of the problems that plagued its Japanese counterpart, which languished in production for nearly five years before its 2000 release. With an unprecedented amount of raw Japanese text in front of him, Blaustein found himself facing creative and technical challenges he never encountered before—and without our modern tools and processes to aid him.

USgamer: Were you aware of the history of Dragon Quest VII? Because I believe it had been in development for maybe five years, which was kind of strange for a game of that era.

JB: No. Simply put, I didn’t know anything, nor had I played any Dragon Quest or Dragon Warrior [games]. So I had no knowledge of anything that was going on—beyond [individual] scenes. I did not know the series. I was just approached by the American contact point, because I had done Valkyrie Profile, and I had done voice over direction, translation and the VO. And they were very happy with it.

We didn’t even have the word "localization," and the whole idea of localization was an afterthought... For example, R&D teams, as you know or can imagine, have a development schedule. They work on a game. When a game’s done, they put everything down and they take a vacation, and then they get onto their next project. But there’s no time in there for, "OK, let’s deal with the localization." The Japanese developers [are] only thinking about their own market.

So, and then when [the Japanese developer] started [working] again, the teams would be mixed up. Different people would be assigned to different projects, completely busy with them, and their bosses wouldn’t want you bothering them about a previous game, because now they’re on a new project. So, yeah, no answers, no help, no organization, no idea what’s going on. And from my perspective, it’s just, "OK, here’s the text, it’s coming in in these enormous batches."

USg: Yeah, I was going to ask if you knew what a gargantuan game this was shaping up to be, because it felt like they were just adding content to it to make up for the fact that it was delayed for so long.

JB: I became aware of the size of it early on, and yeah, I panicked, frankly. I thought, well, my God, how could I possibly deal with this? I approached a friend of mine who was a programmer. I was living in western Massachusetts, and I knew this guy, he’d gone to Princeton and he was a programmer. Very smart guy, very weird guy, built his own house. And, he saw it as a technical problem. Basically, I said to him, "I just don’t know how I can literally manage this, because there’s so much text and so many files that even to take the time to understand the structure of the files [seems impossible]” which is a very important thing to understand, as you can imagine.

Some files are going to be menu things, some files are going to be spells, and some will be items, and some will be dialogue, and some will be divided into chapters. There are different ways the text files will be organized, but without first having a firm grasp of what this organization of these massive numbers of files are, you can’t figure out a smart way to approach it.

What we wound up doing was creating our own translation tool. They have a computer-assisted translation tools, now, that everyone uses, Trados or memoQ, these kinds of things. But, you can imagine if you have five, or ten, or fifteen, or even twenty translators all doing something, to make sure that they’re all doing things the same way and correctly... [A]nd even just the process of delivering and receiving the assets and managing them, maintaining them, it’s an unmanageable headache at that scale. The scale is just ridiculous.

So, anyway, he began to build this platform, and we formed a company called Wordbox. Wordbox was the name of our software, as well, this tool that we built, and it put everything online, so all the translators could work at the same time seeing the assets online. We built in glossary functions, so we built a glossary as we go. All these tools that are now accepted as a regular part of the computer-assisted translation tool process. You’ll see all these things, we basically built those without really knowing—Trados existed at the time, but it wasn’t online.