For my first blog post on Gamasutra, I will share with you a small article I wrote recently about game localization. A lot of my friends who know about my job as a video game translator have asked me, in various ways, why it takes so long for game to be localized for the French market.
More generally, it's common to see gamers complaining about the time difference between game releases in different territories - sometimes up to several months. Even if game localization takes much less time than it used to, there are still cases of long delays between local releases.
It is important to understand the game localization process as a whole. Even in this era of dematerialized content (which means less time spent on packaging, manufacturing, etc.), there is not much you can do to speed up the translation/QA testing stages.
Does game translation really take that long?
As a professional French translator, I have worked on RPGs containing several million source words. However, even an experienced translator won't be able to handle much more than 3,000 words a day. In some cases, a single translator would need a few years to translate a large RPG on adventure game!
Of course, splitting the work between translators is an obvious solution to reduce that time, but it also creates issues that are time-consuming in themselves: translators have different writing styles, so consistency issues may arise, making the proofreading and linguistic testing process longer.
In all cases, testing takes an awful lot of time: to see the translated strings in context, even with an automated script, you will need to sit in front of a screen for hours and hours. When mistakes are found, developers have to implement the fixes and send the game back to testing, and there is no shortcut here - you can't have unlimited developers doing this work.
With a well-optimized localization process, developers can get their games ready for new markets in a reasonable time. But for text-rich games, there aren't too many workarounds, especially if you decide to localize the game after its initial release.
And game strings are just the tip of the iceberg
The translation of game strings is just a small part of the process, especially for big budget titles. Commercials for example need to be localized. It is not rare to see completely different commercial for a same game from a territory to another. It takes a lot of time and people to come up with ads tailored for the specifics of each market.
I could also mention websites, press releases, local events and so on, but I'm sure you got the idea.
Isn't there anything we can do about it?
In order to release a game worldwide at the same time, organization is key: you would need to prepare the game texts during the design stage and get the localization team on board as early as possible: ideally translation should start before programming does! Linguistic testing should then be undergone at the same time as functional testing.
The problem, currently, is that localization is overlooked by most developers: in general, texts are prepared in the late stages of developments, when it's already too late. The best way to go about this is to educate developers about game localization and have them plan localization as early as possible.
Source: Yoshino Trad