The ISO 8859-1 character set and its <language>_<country>
collations are technically best. You should try the resulting
sort order of ES_ES with your data and decide whether you agree.

Punctuation characters are ignored by the <language>_<country>
sort order which is both standard conforming and useful, but
surprises some people. Check for yourself.

More information than you may ever want to know at:
http://www.cs.tut.fi/~jkorpela/chars.html
http://czyborra.com/charsets/iso8859.html

All UNICODE characters in PDF charts:
http://www.unicode.org/charts/

Interactive character reference:
http://www.zvon.org/other/charSearch/PHP/search.php
start here if you don't know where to start:
http://www.zvon.org/other/charSearch/PHP/search.php?searchType=101&script=LATIN&id=25


Crossreferences of other character sets to UNICODE
ftp://ftp.unicode.org/Public/MAPPINGS
ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859
ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS

Regards,
Peter Jacobi
Hamburg, Germany