"How Difficult Can That Be?" - The Work of Computer and Video Game Localization

Frank Dietz, Ph. D.
ATA-certified English-German translator, specialist in technical translation, software localization and video games

1. Perception and Reality

Years ago, when I worked as an in-house German translator at the computer game company Origin Systems, I reviewed a number of applications for an job opening the department had advertised. The cover letter of one of these applications started as follows: “I have never translated a computer game, but how difficult can that be?” I have always remembered that naïve statement (and, by the way, this applicant later scored very low on the test we sent out) as symptomatic of the common misconceptions concerning video and computer game localization. This attitude can be summed up as: Games are something for children, therefore they are simple, and localizing games must be easy. That is erroneous has been clearly shown by the unfortunate effects of badly localized games, which can range from awkward (think of the infamous phrase "all your base are belong to us" in the European version of the Japanese game Zero Wing) to simply unplayable (for a discussion of translation-induced “plot stoppers”, which prevent players of localized versions of finishing the game see Dietz 2006, 125).

Besides these condescending attitudes towards games, there has been a kind of culture gap between professional translators and gaming aficionados. This gap will gradually disappear, as games have been moving out of the hardcore user demographic and into the mainstream. However, most translators still more or less stumble into the field of games localization by accident, and the following discussion aims to help beginning and aspiring game localizers find their way into this field.

2. Computer Skills

I had to laugh when I saw an article about translation in the German news magazine Der Spiegel in 2002. The accompanying illustration showed an elderly man sitting in a library with a box full of index cards in front of him. Even then, this image was incredibly outdated. Translators have been using computers, the internet and CD-ROM dictionaries for years now, and those interested in computer/video game localization have to be particularly computer-literate.
There are several reasons for this:

  • You have to be familiar with the specific hardware and software terminology in your source and target languages. If terminology such as “analog controller”, “SLI mode” or “anti-aliasing” is alien to you, you might have problems translating games-related documents.
  • You have to be able to play the game. This might be difficult for video game localization projects, where special development/testing versions of the console may be needed, and even on the PC side you probably will have to deal with special copy-protection technologies in order to play a yet unreleased game (game development companies are extremely worried about game copies “leaking” and then being pirated). However, you should spend some time playing the game, if at all possible. Many mistakes I have seen in game localizations were obviously caused by someone not being able to visualize what would be happening on the screen. If playing the game (or at least watching a QA tester play the game) is not possible, you should try to get as much information (screen shots, videos, plot summaries, walkthroughs) as possible in order to understand the world of the game.
  • You have to be able to deal with hardware and software conflicts. Unfinished games can be rather unstable, and might even crash your computer. You should ideally be able to update your video and sound card drivers, report error messages, even deal with any necessary hardware upgrades, such as installing more RAM or a better video card.
  • You have to be able to research quickly. That, of course, means using the WWW. Forget about finding gaming-related information in conventional dictionaries. However, there is a wealth of gaming-related sites out there – websites of print magazines, online magazines, discussion sites, cheat sites, walkthroughs, etc.

Game localization is unique in the sense that it may require both the skills of a technical and a literary translator. If you are localizing a fantasy role-playing game, for instance, you will have to render tales of elves, fair maidens and mythical treasures, but you will also deal with advice about video card chipsets, versions of DirectX and sound card incompatibilities.

3. Knowledge of Gaming

Localizers should also be familiar with the gaming scene and jargon in the target culture. Are there, for instance, terms that would not be translated? Are there standard translations for certain game types (such as “first-person shooter” or “real-time strategy”)?

Hard-core gamers tend to be quite passionate about their fields of interests and will be quite vocal on internet discussion sites if they consider a localization to be amateurish. Therefore, translators should familiarize themselves with the “gaming scene” in their target culture. This might appear difficult, particularly for older translators, but researching the gaming-related websites mentioned above and subscribing to gaming magazines in the source and target languages should go a long way towards this goal.


4. Subject Matter Expertise

This is an issue that is sometimes neglected in the discussion of game localization. The thematic content of games can range from extremely simple (as in an arcade game) to highly complex (think, for instance, of a flight simulator with extensive reference materials, such as the titles in the Jane's Combat Simulation series, which often came with 250-page manuals).

The required subject matter expertise can therefore range from literary (fantasy role-playing games, science fiction games) to extremely technical (sports and military simulations). While I would happily undertake a the localization of a World War II flight simulation or a naval strategy game, I would, for instance, avoid a baseball simulation, as I just don’t know enough about that sport.

A lack of such subject matter knowledge can have embarrassing results, as, for example, in the translation of the manual for the helicopter sim AH-64 Longbow into Swedish (ridiculed at the time on flight-sim newsgroups) which mistranslated the pilot jargon “Winchester ammo” (= we are low on ammunition) and stated that the attack helicopter (which is equipped with a 30-mm-cannon, Hellfire missiles and 5-inch-rockets) had only “shotgun shells” left. (Google groups archive for comp.sys.ibm.pc.games.flight-sim. Date: 1996/06/15).

The upshot of all of this is that translating games is not child’s play. A translator must become thoroughly familiar with the special subject matter of each game, be that medieval alchemy, skateboarding or avionics, as well as with the terminology and conventions of each gaming genre.

5. The Localization Process

Over the years, the scope of localization and the recognition of the importance of international markets has increased alongside with the growth of the computer and video game industry. While 20 years ago, games might not have been localized at all (or only partially, e.g. with only the printed manual and installation guide translated), complete localization and the simultaneous (or near-simultaneous) launch of several language versions have now become much more prevalent.

The reason lies in the high costs of game development. Major titles no longer can be cobbled together by a few people working in the proverbial garage, but are produced by large teams of programmers, designers, graphic artists, sound specialists and others who work often for years and require multi-million dollar budgets. At the same time, market competition is fierce, and the average “shelf life” of a game is extremely brief – after a few months, it will be sold at a reduced price, and in a year or two, you may find it in the bargain bin. All of these factors exert enormous pressure on game developers to serve multiple markets simultaneously through a so-called “sim ship” (i. e. the different language versions are all to be released simultaneously), in order to recover the huge development costs as quickly as possible.
Game localization is performed in various ways. Some companies may use in-house staff, others hire a freelance translator (or virtual teams of translators, see below), or use translation agencies. In many cases, it is actually not the game developers, but the distributors (or their foreign subsidiaries) who take care of localization. Unfortunately, this often means that the translators will receive the material to be translated, but not the game itself.

What does this mean for the translator working on a game localization project? Unless an entire development team has had previous experience with localizing a game, the process is likely to be error-prone and difficult.

At the root of many of the problems connected with game localization lies the fact that a simultaneous (or almost simultaneous) release of several language versions requires parallel development. Programmers, designers, sound technicians and graphic artists who are usually under enormous stress during the beta and final stages of game development will also have to devote some of their time to creating (and fixing) foreign-language versions. On the translator’s side the parallel development aspect means working with a text that is, despite all assurances to the contrary, still unfinished, and sometimes requires frantic re-writing and re-translating during the last few days before the game reaches the market.

That the localization of games is often a difficult process comes as no surprise, considering the industry’s general reputation for poor planning and management efficiency. As Simon Larsen puts it bluntly in his report Playing the Game: Managing Computer Game Development: “Game development projects are generally badly planed and badly managed. This results in delayed productions and exceeded budgets” (Larsen, 2002). While I consider this judgment somewhat too harsh, I still would have to add: “And localization usually comes as an afterthought and is sometimes managed by a person who has no experience in this field.”

6. Virtual Teamwork

Games are often large projects with extremely tight deadlines. Therefore, large game localization projects may be divided among several translators who form a “virtual team”. In this case, the virtual team takes over certain project management functions, in addition to the actual translation work. One person, for instance, might be the “translation memory manager” responsible for updating and distributing TM files on a daily basis. Another team member might take on the role of “query manager”, collecting questions about the texts and forwarding them to the development team.

In order for this to work, members of the virtual team must be active communicators, using e-mail. Instant messaging or VoIP phone to exchange information on a constant basis. If you check your e-mail only once or twice a day (or still use a dial-up connection), this might not be right for you. Team members also have to be reach compromises on terminology questions in order not to slow down the group's work. If you are used to working alone, a virtual team might take some adapting, but it will be worth while, as you gain access to game localization projects that would have been beyond your own capacity.

7. Outlook

Game localization can be a complicated and chaotic endeavor. The structure of the gaming industry, the often rather parochial outlook of development teams, and the enormous pressure to ship several language versions of a game at once (particularly in publicly traded game companies) all create obstacles for the localization process. There are a number of steps, though, that both translators and members of development teams can take to reduce friction and make the process more effective:

  • There should be early and frequent communication between translators and developers in order to avoid interface design dilemmas, file format issues or cultural insensitivities.
  • Translators should also have one contact person in the development team (sometimes jokingly referred to as the “translation czar/czarina”) who can distribute queries to the appropriate team members and ensure that all relevant materials are routed to the translators.
  • Source code tracking software (such as Visual SourceSafe) can not only be employed to track revisions of code, but also to ensure that changes in the English text are flagged and communicated to the translators.
  • Translators should receive basic design documents early on, so that they can gather reference material suitable for the particular type of game.
  • The use of translation memory tools (which is particularly important considering that many successful games have several sequels and add-ons) should be increased. This could involve an industry-standard program like Trados or SDLX, or a proprietary product developed in-house, such as Ion Storm’s LÖGAN.

Most importantly, translators should have a chance and be expected to play the games they are localizing. “Blind” localizations are unfortunately still all too common, partly due to developers’ concerns about software piracy (though protection utilities such as SafeDisc reduce that risk), partly because of lack of interest on the translators’ side, who may not be aware of the complexities of game localization.
The (decidedly) second-best solution would be providing translators with a wealth of background information, such as screenshots, design documents and game walkthroughs and later have the game tested by native speakers of the respective target language.

References

Bates, Bob. 2001. Game Design: The Art and Business of Creating Games. Rocklin, CA: Premier Press.

Bemis, Greg. 2002. “Game Localization: What does it take to bring foreign games to the US?” TechTV, January 18. http://www.techtv.com/extendedplay/videofeatures/story/0,24330,3368070,00.html.

Chandler, Heather. 2005. The Game Localization Handbook. Hingham, MA: Charles River Media.

Dietz, Frank. 1999. “Beyond PacMan: Translating for the Computer Game Industry.” ATA Chronicle 28 (9): 57.

Dietz, Frank. 2003. “A Translator’s Perspective on Games Localization.” Multilingual Computing & Technology 14 (5): 21-25.

Dietz, Frank. 2006. “Issues in Localizing Computer Games” , Perspectives on Localization, ed. Keiran Dunne, pp. 121-134. Amsterdam/Philadelphia: John Benjamins.

Esselink, Bert. 2000. A Practical Guide to Localization. Amsterdam/Philadelphia: John Benjamins.

King, Brad, and John Borland. 2003. Dungeons and Dreamers: The Rise of Computer Game Culture from Geek to Chic. New York: McGrawHill.

Larsen, Simon. 2002. Playing the Game: Managing Computer Game Development. International Edition, Version 1.1. Blackwood Interactive. http://blackwood.dk/PDF/PlayingTheGame-IE.pdf

Trainor, Helen. 2003. “Games Localization: Production and Testing.” Multilingual Computing & Technology 14 (5): 17-20.

Online Sources:

Gamasutra: http://www.gamasutra.com/

Game Research: The Art, Business and Science of Computer Games: http://www.game-research.com/

Game Studies – The International Journal of Computer Game Research: http://www.gamestudies.org/

Computer Game Localization Glossary: http://www.frankdietz.com/softgloss.htm

 

Novembre 2007