A common assumption about internationalization is that every user fits into a single locale like “English, United States” or “French, France.” It’s a hangover from the PC days when just getting the computer to display the right squiggly bits was a big deal. One byte equaled one character, no exceptions, and you could only load one language’s alphabet at a time. This was fine because it was better than nothing, and because users spent most of their time with documents they or their coworkers produced themselves. Today users deal with data from everywhere, in multiple languages and locales, all the time. The locale I prefer is only loosely correlated with the locales I expect applications to process.