I'm having difficulty running "accent insensitive" searches with
utf8_general_ci. This query confirms that it's possible:

SELECT _utf8'José Canseco' COLLATE utf8_general_ci LIKE _utf8'%jose%'
COLLATE utf8_general_ci

(Returns "1")

I've also confirmed that utf8_general_ci is the character set of the
DB field I'm searching against:

SELECT DISTINCT COLLATION(name) FROM players

(Returns "utf8_general_ci")

So in theory, all I need to do is enforce the character set on the
search string to ensure utf8_general_ci:

SELECT * FROM players WHERE name LIKE _utf8'%jose%' COLLATE
utf8_general_ci

This doesn't return "José Canseco", but does return "Jose Canseco" if
I change the name. I've tried playing about with the query, but can't
get it to work. Any ideas would be much appreciated. I'm stumped!

Thanks.