The game industry has evolved into a giant in the past ten years, and now more than 50 million game players around the world enjoy different types of games on their mobile phones, computers and game consoles. Facebook, game consoles like PlayStation and Xbox as well as every PC host many offline and online games and applications. Gamers often choose to play the same game on various platforms and as every single platform doesn’t use the same specifications (RAM, CPU or graphic card), game companies create nearly the same game with limited game playing features.
Game localization, globalization and internationalization are long and toilsome processes. Translators and language service providers need to deal with different issues: context, terminology, cultural effects and visual translation. These issues differ in every language family. Language families, directions and fonts influence the difficulty of the localization process, which includes character corruption and length, screen limitation and so on. So, what are specific game localization issues in Turkish?
Context
Polysemous words — words with more than one definition — play a critical role in game localization. Many words in many languages have multiple senses, meanings and contexts. In game localization, a simple word can hold two or three meanings in any target language. The biggest dilemma here is the nonglobalized game content. The game context is generally written in English, published in several languages and distributed to the whole world.
For example, how can the word short be problematic in game localization? In Turkish, it will be problematic if game content writers and engineers don’t assign two different segment IDs for the word in a soccer-related game: short (in Turkish “şort”) as in the portion of the soccer uniform worn on the legs, and short (in Turkish “kısa”) as in the length of soccer field grass. But often, there is only one string for this word and that means we have to just pick one of these shorts at random. Such adjectives are usually troublesome for translators using computer-aided translation tools as these tools register repeated segments to translation memories.
Simple polysemous words may actually be more difficult to localize than complex words. For example, home and away are among the first words we learn in English, and their translations are simple even for beginners. However, the words’ usage depends on where they appear. If the translator takes the words in the most literal sense and doesn’t check the context, game and platform, they will translate these two words with the first meaning that comes to mind, like “Uzakta” for away and “Ana Sayfa” for home. However, if it is a mobile sports game these have to be translated “Deplasman” or “Dış Saha” for away and “Ev Sahibi” or “İç Saha” for home. Without context, game localization will be a pain in the neck for translators, editors and management staff like project managers and game testers.
Cultural differences and varieties in games lead translators and localizers on a new localization journey. Every single country, region or race has unique values. So adopting these values to video games has a secret key: internationalization. Games with global signs, images and letters smooth the localization process for localization teams and companies. However, political signs or figures, culturally inappropriate images and so on may be hard to localize, especially if these are sensitive for Turkish or global gamers. For example, if the game contains politically or historically polemical signs and phrases (such as Nazi symbols), it may be a struggle to localize these into some languages and cultures.
On a positive side, culturally and globally accepted signs and figures will be effective in games sales. Correctly localized games make local people invest in games. The Uncharted series, for example, nearly traveled the whole world and picked well-known cities like Istanbul and London. This attracted the attention of local gamers and made the game well known for potential buyers.
Linguistic problems
Internationalization problems are not limited to context and visual problems. Turkish is written in the Latin alphabet and it’s a Ural-Altaic language so it has different grammar rules than common gaming languages like English, German and Spanish. Although Turkish has Latin letters it also has unique consonants and vowels: Ç, ç, Ğ, ğ, İ, i, I, ı, Ö, ö, Ş, ş, Ü and ü. Even if Turkish shares the same language family with Japanese and Korean — game industry giants — they have unique letters. Ural-Altaic languages have nearly the same grammar rules but totally different alphabets. This difference leads to different localization problems like length restrictions, character corruption, text truncation and so on.
Turkish Characters. The characters Ç, ç, Ğ, ğ, İ, ı, Ö, ö, Ş, ş, Ü and ü sound uniquely Turkish. When the game internationalization process is underway, engineers must pay special attention to the consonants Ç, ç, Ğ, ğ, Ş, ş and vowels İ, i, I, ı, Ö, ö, Ü, ü. These letters may cause character corruption if the game localization engineer doesn’t prepare the game using UTF-8 language code.
Many mobile games are launched with restricted language support, and even if they support UTF-8 in general, they can’t support sub-language specifications like Turkish. In that case, “ı” is shown as “ý” and other special Turkish characters may be shown with a question mark. Using the correct language code for every language is vital for internationalization and localization.
Prepositions. Turkish is an agglutinative language and this may lead to challenges associated with suffix usage. EA Games launched the annual FIFA series, and in FIFA15 there is an unsolved localization issue with suffixes. The localization team and translators left a preposition unlocalized, which led to a critical problem. There were two variables, {1} and {2}, and these are interchangeable with every single soccer player’s name in game data (almost 10,000 of them). The phrase to be localized was “{1} passed to {2}.” This may seem simple to translate, but for Turkish it’s not. If you don’t know how to handle prepositions and suffixes in Turkish, you may encounter many issues like these.
“{1} passed to {2}” was translated as “{1} {2}’ye pas Verdi.” The “to” preposition turns into a suffix when localized into Turkish, but by translating “to” as “-ye” in Turkish, the localization team omitted other potential translations. This translation may be correct if the last letter of the player’s name in variable {2} is i, e or ü. In that case, the use of the “-ye” suffix will be correct. However, if the last letter of the player’s name in variable {2} ends with any of the vowels a, u, ı, o or ö or any consonant, this suffix will be incorrect. For example, if {2} is “Ronaldo,” the translation will be “Ronaldo’ye,” and we can’t use the “-ye” preposition after o in Turkish. As it turns out, there are two simple solutions for this problem. The first one is passive voice usage and the second one is reciprocal voice usage. If the localization team used reciprocal voice, the translation will be “{1} {2} ile paslaştı.” So the problem is solved with a simple rewriting trick. Passive voice usage may not help here, but in other games it can be useful. In a combat scene or war game, “{1} killed {2}” can easily be localized using the passive voice: “{2} {1} tarafından öldürüldü” or “{2} öldü. Öldüren: {1}.” There are simple and easy shortcuts such as this in game localization and any untested game will most likely have these kinds of localization problems.
Dot, comma and percentage. Punctuation marks have different usages in every single language and they are crucial in localization. In game internationalization, punctuation marks must be coded for all languages in order to be localized. The most common internationalization errors in punctuation, such as dot and comma usage in decimals and percentages, have to be checked before launching.
In Turkish as in many other languages, the comma is used for decimals rather than for thousands. Often this simple difference is skipped or neglected in internationalization and testing, and thus many games localized into Turkish have these kinds of errors.
The percentage sign in Turkish is written “%100” with no space between the sign and number. In other languages this is written differently.
Negatives. How can two very different source texts be translated into Turkish the same way? It sounds weird but it may happen when the context is a black box for the translation team.
For example, “Don’t shoot” and “Shooting” may both be translated as “Ateş Etme,” and if the game isn’t tested a game player may encounter an incorrect translation in their game experience. There is an occasional solution for this situation: using an exclamation mark. This way, negative or imperative voices can be differentiated. So it is also crucial for Turkish game creators to differentiate these two phrases in Turkish as “Ateş Etme!” and “Ateş Etme.”
There is another workaround for this problem: creative translation or transcreation. If game translators and editors experience this problem they may transcreate a new translation to solve it. This is possible by translating a phrase using a synonym. For example, “Don’t shoot” may become “Stop shooting.” Creating such new phrases will help every localization expert in each language to cope with the same target inconsistency errors.
Truncation. Long phrases and sentences are always an obstacle in game localization, as there is usually a length restriction for every different platform. Character limitation obstacles push translators and editors to be more creative and understandable in the target language. Even if game producers prepare their games using internationalization standards, there are some words that may not be truncated or interchanged with others.
Configuration is generally truncated in games as Config and in Turkish it may be translated as “yapılandırma” or “konfigürasyon” depending on the context. The config word needs to be truncated in Turkish, too. However, if we use the first six letters of “yapılandırma” it will be “yapıla” and this won’t give the same feel and meaning in Turkish. If we use “konfiguration” in the Turkish translation it will be “konfig,” which also does not have the same feeling in Turkish.
The shortened version of any long phrase or word in Turkish may be truncated by omitting vowels or a part of the word or phrase. However, the key of truncation is a readable and understandable target. If the truncated word can be understood by a native, then it will be useful.
Game localization is a long and continuous process for Turkish, as for every language, and needs to be tested in every phase. Every translator needs to play the game before translating, if possible. By playing the games they could eliminate major errors that may occur if they don’t understand the context. Moreover, if the testing team plays the games before translation, the team will guide translators to highlight the tricky parts of the localization and translation process. For the sake of continuous and smooth delivery, screenshots, references and gaming experiences are vital. Don’t leave translators and editors unguided in the sandy deserts of localization.
