I did the FBN assignment and the reason I chose UTF-8 as my encoding when I rewrote the deprecated code was that was the encoding in the existing database file.
I know there have been different versions of the FBN assignment, but I got one where I was already given the database in binary format - I didn't have to write a data converter class like some people did. I gather most (if not all) of the new format assignments do require you to write your own data converter to produce a binary format file.
Anyhow, the binary file I got was in UTF-8, so it was a no-brainer to set my encoding to UTF-8 when replacing the deprecated code. That's why I said I didn't even mention it or try to justify my "choice".
I assume that if you talking about fixing deprecated code you are doing some version of FBN too (I think you only get interfaces for the new exam, no implementation) so you are probably in the same position I was.
Andrew - do you ever sleep? You always seem to manage to be first out of the blocks with a reply to all the questions!