I'm currently working on a prototype for a consumer facing application which hopefully will have global applicativity.<p>I've often heard the adage that it is never too soon to worry about localisation and timezones, but am wondering whether when trying to develop a MVP whether this holds.<p>I know there are a number of founders here, so would like ask: "When did you start worrying about localisation? Would you have approached localisation differently if you were to start over again?"
I start localizing after the product feels finished, or only translate parts that <i>are</i> finished and not likely to change soon.
So by my rules, a prototype (that is finished) could show working localisation with minimal effort. (Just using an array with strings instead of a full blown localisation/formatting solution.)<p>If my product should take timezones into account, its data formats and business code are designed to account for that from the start.
Localization is a hard problem. Too many factors to consider. If you are building a MVP, I would advise not spending too much time on it. You can certainly leave some placeholders in the application if it fits but building a full blown localized product is 2-3 times harder (in my experience).<p>You also said "which hopefully will have global applicativity."<p>So you are not quite there yet where you know for sure that it will be global. What if it doesn't or takes you a while to first validate the market in general ?<p>My advice: Build the MVP quickly, validate the product and then think about localizing if you get enough traction.