Applications designed on Win32, depending on the development language or IDE, use either UNICODE or ANSI. ANSI applications, like programs designed with Delphi and C++ Builder, use code page encoded strings and a non-Unicode version of WIN32 API. Why do you have to know this information?
There are several different code pages. English and most other Western languages use code page 1252. Japanese uses 932, Russian uses 1251, and so forth. To render an ANSI string correctly, the application must have the right system code page. This page is the default code page of the system. In Windows ME, 98, and 95 you cannot change the system code page.
Usually, you will see the problem when trying to run your application. If strings in the menu items and components do not show correctly and display gibberish, the system code page is not right. This effect is called Mojibake.
The following screenshot shows a Japanese application that is run on a computer where the Japanese system code page is active.
Everything is fine here.
However, if the same application is running on a computer that has the Western code page, the application appears like this. As you can see, hopefully without the help of your Japanese colleagues, the strings are mojibake ("character changing" = gibberish).