The MUI extension and the <language-code> subdirectories (like 'en-us') are a convention introduced in Vista (see for example
http://msdn.microsoft.com/en-us/goglobal/cc442492.aspx#MUI), with some support built into Windows (GetFileMUIPath and other APIs). However, the technique itself of isolating the "localized" strings into separate per-language resource-only DLLs has been used forever.
One can match strings between languages based on their numeric resource ID. For an example, looking into the en-US\ULIB.DLL.MUI message table (which holds the strings for a number of file utilities including XCOPY) using
http://www.angusj.com/resourcehacker/Code: Select all
C:\etc>reshacker.exe -extract %systemRoot%\system32\en-us\ulib.dll.mui, ulib.rc, messagetable,,
the generated ulib.rc file contains a line:
Code: Select all
22063, "Does %1 specify a file name\nor directory name on the target\n(F = file, D = directory)? %0"
The corresponding line with ID 22063 in the German de-DE\ULIB.DLL.MUI would contain the equivalent German string.
Liviu