Dangers of Crowdsourced translation

Hey, you! Yes, you. Are you a proud game developer who considers translating their game into another language? If so, I prepared a few tips and warnings for you, in order that you can avoid low-quality translations and dissatisfied fans.

Have you ever heard of crowdsourced translation (CT)? In case you haven’t, here’s a basic concept: you gather some volunteers, i.e. completely random people, and give them strings to translate. Those volunteers often work for free because:

  • they simply want to have your game translated and play in their native language;
  • they are aspiring translators and simply want to gain experience;
  • you promised them to put their name in Credits and/or grant them free copies of your game, etc.


In the end, players can finally enjoy your game (not everyone is fluent in English and apart from that, it’s always nice to have a localised product) and you get a translation done for free. Everyone is happy! Sounds too good to be truth, doesn’t it? Precisely.

My experience with crowdsourced translation? Rather bad, I wouldn’t like to go through this again. Few years ago, Abbey Games (a team responsible for a great game named Reus) decided to translate their game into various languages. They asked their fans to come aboard, promising new content for their game. I started translating strings into Polish, because I wanted to gain experience and to unlock new content. Most of the context was rather clear to me (I had played over 20 hours at that time), so the only problem I had was finding proper names of some plants and animals in Polish, and dealing with constants, which wasn’t that bad. What was bad were the people. There had been no organised groups, no professional translators to handle the situation properly (at least none of whom I knew). Practically nobody cared to keep things consistent, a term like Village or Giant had like 6 different translations. It was a huge mess and it saddens me that nobody managed to finish this translation.

I wasn’t the only one who has had negative experience with crowdsourced translation. Let me quote some friends of mine, professional translators whom I asked about their opinion on CT.

Thomas Faust

avatar[regarding Reus translation] Long story short: it stalled at 56% for a long while because all the translators jumped ship. That’s actually the worst thing that can happen – and it happens often. Up to this day, the German Reus translation is flawed. No good game deserves this fate.
[regarding Cross of the Dutchman] We have 3 forms of address here that apply: „you” can be translated as „du”, „Sie”, and the old-fashioned „Ihr”. Now the translation kept switching between those words – sometimes three subsequent sentences in a dialogue had all three in them. Why? Because nobody playtested the thing and every translator had a different style of translating. So it’s a mess. Surprise.

Michał Tosza

av22Yes, I do have some experience. Editing was a terrible nightmare. There are many people who don’t have much experience with translating (which isn’t really surprising) and yet they don’t give a heck about terminology, styleguides, tips and instructions — nothing. An editor often needs to do everything from the scratch. If the employer has enough common sense, he will pay his EDITOR. Otherwise, he will have a tough time.

Certainly, crowdsourced translation may have some advantages:

  • Costs — If you’re low on money and really cannot afford hiring a professional, you might try doing CT instead;
  • Accessibility — Using crowdsourced services with a neat UI can make it easier for your volunteers, while allowing them to download the file and translate it with CAT tools can make their work faster;
  • Community — Asking directly your fanbase and offering various bonuses (new content, including translators’ avatars in game etc.) for their help perhaps will make you seem to be more open and kind to your fans…? This is a really far-fetched advantage, I know.

There are some serious disadvantages that you need to take into consideration, though, such as:

  • Costs — While it might be a cheaper solution at first, think what would happen if your translators messed up everything. Unless you invest in a proofreader or have a professional translator working as a Project Manager, you cannot be sure if the text is translated correctly. In the end, a bad CT translation may cost you not only money (bad translation may discourage people from playing your game), but also your fans who will be dissatisfied with low-quality translation;
  • Anonymity — You don’t know who exactly will work on your translation. Perhaps those people have some experience or perhaps they have absolutely no clue about translating… you don’t know. Apart from that, each person has different style, different personality, making it difficult to cooperate in larger groups;
  • Organisation — Or rather lack of thereof. Lack of communication between team members might be disastrous to your project. If translators do not cooperate they will either give up, or present you with an ungrammatical, illogical, incoherent and illegible text;
  • Motivation — As soon as something goes wrong, people will lose their motivation to work. After all, it is a voluntary work, isn’t it? Why should they care? If your current team leaves and then new members come, they will have to go through the translated text and see how correct it is. Of course, nobody will bother to do that, so they’ll start translating new text without checking what previous members did what results in a huge mess.

Alright, if you still insist on doing CT here are few ideas how to improve the general quality of work and translation:

  • Hire a professional editor — Just as Michał Tosza said earlier, make sure to hire a professional who will check quality of the translated text. It would be a good idea to have few PMs to take care of the volunteers, monitor the progress; just your usual PM stuff
  • Do some testing — Before you make that translation public, make sure you had some players test it. Sending them instructions how to properly mark errors, or even better: giving them a special app or using the same/another CT service, would help them deal with their task
  • Be direct and communicative — With your volunteers: try to keep your translators motivated, praise them for progress, put some sneak peeks of new content you will include in the game; but also with your fans: ask them for their opinion from time to time, whether they like this translation or not, it might be helpful!

That being said, I would still recommend you looking for professional translators and for professional GROUPS, as they are usually organised and know how to handle such projects efficiently.

I hope this entry gave you some insight on how translators feel about CT. If you have some experience with CT, have your own story to tell, feel free to share it in the comment section bellow. I’m really curious what you, a developer, think.

Jedna uwaga do wpisu “Dangers of Crowdsourced translation


Wprowadź swoje dane lub kliknij jedną z tych ikon, aby się zalogować:

Logo WordPress.com

Komentujesz korzystając z konta WordPress.com. Wyloguj /  Zmień )

Zdjęcie na Google

Komentujesz korzystając z konta Google. Wyloguj /  Zmień )

Zdjęcie z Twittera

Komentujesz korzystając z konta Twitter. Wyloguj /  Zmień )

Zdjęcie na Facebooku

Komentujesz korzystając z konta Facebook. Wyloguj /  Zmień )

Połączenie z %s