I have a csv file which, if I import using the Wizard, the File Origin: is set as '932: Japanese (Shift-JIS)'. This causes, as one example, the character Ï (i umlaut) to change to a ?
However, if I simply double click the file to open it, the character is preserved and displays correctly.

So my questions are...

a) Why is the file, seemingly, treated differently by Excel depending on how it is opened?

b) How is the File Origin: determined by Excel?
(I believe I have previously read that there is nothing in a csv file that can contain details of how it was encoded, and therefore Excel has to essentially scan the file to make a best 'guess'.

c) Is there a way to set and preserve a default value for the File Origin:?

d) I have found mentions of this problem via Google, hinting at a registry change, but they all seem to refer to a much earlier version of Excel and Windows.
Would/could this still apply to Excel 2007/Windows 7 (64-bit)?

e) What is the best File Origin to use for general English name/address data? I have tended to always use 'Windows (ANSI)', but 'MS-DOS (PC-8)' seems to be selected quite often.

Thanks.