Take the 2-minute tour ×
Programmers Stack Exchange is a question and answer site for professional programmers interested in conceptual questions about software development. It's 100% free, no registration required.

I recently converted an iphone app to support different languages, and was wondering which ones I should include, and which ones aren't worth the trouble.

What I mean by that is best illustrated by an example. In Ireland, the Irish language is spoken by very few of the people there. It could be considered a dying language. Almost everybody speaks English (if not everybody). So in this example, I don't think it's worth the trouble to support.

In addition, the number of people using modern technology may be limited as well. For example, most people in Cambodia would not be likely to purchase software, and therefore the benefits of localization are reduced.

What languages do you support when localizing?

share|improve this question
As an aside, good localisation is expensive. You need to hire people who are proficient in the technical jargon of a language. If you do it on the cheap, then you risk insulting the mothers of a whole nation. And it is not just the app, you also need to come up with new advertising and marketing in the assorted languages. Make sure that the cost is less than the expected profit of getting in to new markets. –  smithco Feb 23 '11 at 5:15
I agree, but it depends on the software being developed too. Smaller applications could use simple translations (like Google Translate) and inform the users of that fact when the app starts, then ask them to help with corrections. I think most people would understand the translations aren't perfect at that point, and would reduce the number of potentially 'insulted' people. –  Javy Feb 23 '11 at 5:59
@Javoid: i disagree on 'simple translations', a bad translation is worse than none (and machine translations are still some of the worst). even the barely literate can easily memorize foreign language commands used on a simple application, but the wrong terms carry wrong messages, and (worse) pollutes the language. every time i hear (in Spanish) 'accesar' (bad translation from 'access', the correct Spanish word is 'acceder') i have to fight the urge to hit somebody. –  Javier Feb 23 '11 at 6:26
I've seen a lot of Android apps apparently using Google Translate or something for the German translation, and trust me, it's often hard to guess what was originally meant. Often the results are funny and sometimes completely useless. In some cases, it's even extremely bad. E.g., let's assume you have a "Space Invaders" clone and the instruction text is: "shoot at the alien". Google translates that to "schießen Sie auf die Ausländer", which means "shoot at the foreigners" (immigrants) and makes you look like a fascist. The problem is: You can't assess how good the translation is. –  user281377 Feb 23 '11 at 11:37
I'd like to know the general answer to this as well. I have a couple apps that seem to sell evenly across almost every non-English speaking region, but in tiny amounts compared to U.S. sales. Which 3 language localizations would likely get me the most bang for the (good quality translation) buck, if that's what's in the budget? –  hotpaw2 Feb 24 '11 at 2:48
show 1 more comment

closed as primarily opinion-based by MichaelT, GlenH7, gnat, Simon, Bart van Ingen Schenau Jan 2 at 17:59

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise.If this question can be reworded to fit the rules in the help center, please edit the question.

7 Answers

up vote 3 down vote accepted

I'd agree with David in using data to decide, but in some cases you need to be proactive.

To get adopted in some countries may require you to provide a localised version and you won't get many (if any) users until you do provide that version.

For example, you might find that your app isn't offered for sale in country A because there's a law in place that requires it to be offered in both official languages of that country.

share|improve this answer
add comment

I have Google Analytics on my web app and use it to find the top 20 countries hitting it. Supporting those languages requires about 11-12 locales I think. You could apply the same idea from your web site analytics if it's not a web app.

share|improve this answer
add comment

It totally depends on what your customer base is.

To take an extreme example: The app I currently work on is targeted specifically at users in the public sector in Germany. Since it is highly specific to Germany and German laws and regulations, it is very unlikely that it will ever be used outside Germany. So there's no localization, everything is in German.

If on the other hand you write an application that is useful to a global audience, you'll have to find out which localizations make sense. Then I would just consider each new language a feature request/user story. Just like with any feature, you must weigh usefulness against cost to decide if it is worth implementing.

share|improve this answer
add comment

I'm assuming, the original unlocalized version is English.

  • For Europe first priority is usually FIGS (French, Italian, German, Spanish). Personally I'd also consider Russian there;
  • With Spanish and Portugese you also cover most of Americas;
  • For Asia it's CJK (Chinese, Japanese, Korean);

But really, that depends mostly on who your target audience is.

BTW. don't just look at raw numbers. For example there is tremendous number of Hindi native speakers, but English is one of the official languages in India.

share|improve this answer
add comment

Have you considered designing your application such that language doesn't matter so much? For example, do you need to be fully fluent in English to understand how to play "Angry Birds"? What about Ocarina? or Camera Plus?

While it isn't easy to develop a language-neutral UI, many apps manage it through good UI design using internationally-recognizable symbols and this helps keep localization efforts/cost down in the long run in terms of translation.

As an alternative, a video tutorial showing you how to use the app can help non-English speakers to get started.

Also for you to consider is how much effort you are going to spend marketing to a particular demographic. If your app has appeal to Japanese-speaking markets, it may be worth your investment to pro-actively translate it and tweak it for cultural references and idioms.

There have been many classic FAILs when companies try to internationally market products in a naive manner. One example that springs to mind is the Chevrolet 'Nova' car, which means 'Doesn't go' (no va) in Spanish territories. Hence they changed that to 'Corsa', but the damage was already done. Here's a digest of the funnier ones.

share|improve this answer
Would up vote this if I could - using images instead of text where applicable is always a great idea. –  Javy Feb 23 '11 at 15:10
Even if the app doesn't use any text in any language, the App store description for the app will still require a good text description. –  hotpaw2 Feb 24 '11 at 13:37
add comment

It completely depends on the nature of your app, of course. The app simply may have more appeal to users in certain countries than others. So mine your customer data (if such info is available to you) to get an idea of where your app is popular, and use that info to select the languages you want to create translations for.

One option to consider is providing some method for users to contribute translations. For example, provide a word list on your website and let people know they can help you translate.

Edit: If you dont have any data to mine, and just need to guess, how about looking at the list of wikipedia articles by language . Its one data point, but might give you a good idea of who's speakers are most web connected.

share|improve this answer
I usually go by demand, and by offers to help translate. I start with two language options and a button/link that says "contribute a translation" and takes people to the website to download the requisite files. Fortunately I know people who speak a variety of languages so I can usually get someone to verify the contributed translations (which is much easier than doing them in the first place). You could email random users in likely countries if you don't know anyone who reads the language. –  Мסž Feb 23 '11 at 5:26
I like the contribute button idea. Very nice. –  Javy Feb 23 '11 at 5:55
add comment

Most applications I've ever worked on were so specific to a particular locale that no localisation or internationalisation was required. Weirdly, my current project is explicitly ordered to make it possible to internationalise a web application with minimal effort. At the momemt it supports only German, English, and French because those are the languages for the only markets it's currently deployed in. For the future, who knows what will be needed. But the goal is to enable localisation of the application by simply adding a few rows in a database and a few files with screen texts and messages, rolling out the system into a new market without having to ever even restart the application servers. Spanish and Italian are probably going to be added in the not too distant future, personally I don't foresee many others for the time being.

share|improve this answer
add comment

Not the answer you're looking for? Browse other questions tagged or ask your own question.