It's kind of not much of a debate. If next to no one does it, it's because it's not profitable.
Basically, the market for english speaking people in the world is around what, 1.5 billion people ? For Germany, the number of people that don't speak english, and are physically capable of buying the game is what, 25% of the population ? So 20 million ? (yeah, 3 year-old will not buy video games ad mostly don't speak english, and I know for a fact than more than half of the German population speaks enough English, and some of the people that don't speak foreign languages ain't really the most open-minded people out-there... I'd assume a significant part of them don't want alien stuff).
We're therefore speaking of an increase in potential customers in the 1.6%. Ok... let's be crazy and assume 2 percent. Let's go even more crazy and assume that a game would make 10 dollars per sale after taxes, sales, whatever, and sells 100 000 units (both are VERY high numbers as far as indy video games are concerned) over its complete life-span, with 1 expansion. That makes it 20 000 dollars "gained", and a 2 year shelf life. I'm roughly basing numbers on TLF, by the way.
Now, prices.
- base translation would be in the 5000 dollar range. To be redone for each major update. I'm taking into account some overhead costs like proof-reading and the like.
- each minor update would require proof-reading - let's assume 200 dollars each.
- I'd assume that everythin ain't perfect and that some typo-related bug reports happen, roughly once month. That's again around 200 dollars per month (dev work + translator).
Taking into some averages, I'd suppose 20 minor updates during the life-time of both the base game and the expansion. That means, 10000 for "base translations", 2 times 4000 for the minor updates, 2 * 12 * 200 for typos, which amounts to 22800 dollars.
That's a net loss. That's what would have costed, more or less, TLF, which, as far as game goes, is very successful. And, that's not taking into consideration marketing to other countries in other language, the risk associated with translations failing, the fact that games are created with a "culture" background and that it's not because it's succesful in one place that it would appeal to a foreign market.
It's also not taking into consideration that translating for the sake of translation is dumb. If a developper is sane, he'd want the translation not to bring additional sales to pay for the translation only, but to bring additional profit. Otherwise there is little point, because the extra work is just going to make the game more complex to do, increase the budget, and therefore increase the risk and losses if the game fails. At last, please take into consideration that the number's I'm using are based on costs the companies I've been working for during the last ten years as a dev, so... there is potential for improvement, but I don't think I'm far from the mark.
Now, you can tweak numbers as much as you want, but as long as the tranlsation cost isn't more than 30 or 40% of the expected additional sales profit, I personally wouldn't do it.
