So I have thinking: if I have a game (which I am developing but it's taking a long time), and it's in English. And I want to release it in an another country (i.e. Japan), but I don't have the power to translate it in Japanese. I can try in learning it in another language. Do you localize your games by yourself or a team? Let me know.
I helped localize the English for this game here on Itch. The original language was Chinese (which I don't speak any of). But that experience gave me a lot of insight into the best way to approach localization:
1) First of all, keep in mind that there is a huge difference between localization and translation. Translation is basically the attempt at a 1:1 conversion of words between languages. And quite honestly, if you're on a tight (or no) budget, Google Translate will get you 80% of the way there.
Think of Translation as the beta version of having your game in another language.
Localization is what comes AFTER translation and focuses more on making sure that the translated text captures the spirit of a scene in the game. Localization is like the final version of having your game in another language. Google Translate will not help you at all with localization, because localization is very subjective and scenes can be interpreted a variety of ways.
2) If you want to have your game in different languages, you want to find your translators and localizers sooner rather than later, but you also don' t want them to start working until the text of the game is pretty much set in stone. Don't be like the guys I worked with who were trying to find someone to localize the English of their entire RPG 2 months before release.
3) You might need a "Team" of translators, one for each language, but you shouldn't need more than one translator for each language. If you have 2 different people both translating English to Japanese, the style of translator A might conflict with the style of translator B. Just have one good translator per language to keep it consistent.
4) For the best localization quality, you want the final localization done by someone who has the target language as their native language.
5) This one is very important! Keep save files at all major points of your game and/or upload Youtube videos of your own playthroughs (without skipping dialogue and narrative). Share these with your translators and localizers. Localization is only as good as how the localizer is able to interpret what is going on in a scene. For example, if a character says "I am mad", that can easily be translated into another language. But if the sprite on the screen is beet-red and has smoke coming out of their ears, then a good localizer might change it to "I am mad!!!" to better capture the spirit of that scene.
Hope this helps!
First of all, keep in mind that there is a huge difference between localization and translation. Translation is basically the attempt at a 1:1 conversion of words between languages. And quite honestly, if you're on a tight (or no) budget, Google Translate will get you 80% of the way there
Your text in Japanese by translatorpage
まず、ローカリゼーションと翻訳には大きな違いがあることを覚えておいてください。 翻訳は基本的に、言語間の単語の1:1変換の試みです。そして正直なところ、予算が限られている(またはまったくない)場合、Google翻訳は80%の道のりを手に入れます
translated back by different translatorpage
First, keep in mind that localization and translation have a big difference. Translation is basically an attempt to convert 1: 1 of words between languages. And honestly, if your budget is limited (or not), Google Translation gets 80%of the way.
For no-budget, no trusted volunteer project, automated translation is very good at translation of meaning. Not just word by word translation. You can check by translating idioms like "That was the last straw." Part of what you mean with localisation is just good translation. Translating that idom literally like this is the last of that dry grass is not translating the meaning. The meaning is, that an action pushed something over its limit.
Localisation is going beyond good translation. It is rewriting a thing like it would have been expressed by a native. That can also be a temporal thing, if you adapt a theme for modern times, like that series Sherlock did. They did not just refilm some Holmes stories, they made the story believable in modern times, including smart phones and the jokes about them being a gay couple.
The diff in translation and localisation is very apparant in names. To an english native speaker a name like Raven or Stormwind or Steward is something different than to a non-english native. The first is a black bird. You literally are named like the bird. The city has a descriptive name and localised it should be something with strong air currents. And while steward is a name now, in LotR that one guy had that position and a good translation should localize it to the respective title.
Most stuff does not need localisation. Only good translation. Just look at Superman. Or Captain America. Localising them would not go well. (Though people use these terms not very precise most of the time ;-)
In games you have the added problem of your own invented terms and concepts. If you write clear in your own language, automated translation of meaning is good. But if you need knowledge about the story there will be even more things a native speaker should go over. Like that example with mad. It could also mean crazy, instead of angry.
That being said, I think the first step is to actually make the game translateable. If the text is embedded deeply or procedural generated, that is a problem. Some engines are even auto-translateable by third party tools, on the user end.
