if there are multiple character sets that can properly encode my data, how to choose one?
Is there any difference in disk or memory usage? When joining tables do they need the same char set?
Choosing character set usually based on the environment, each set has pros and cons. for example if your system has mixed character I strongly recommend
utf8 but even utf8 has many types. for example:
utf8_general_ci is faster than
utf8_unicode_ci but less accurate for sorting. and if you want to use specific language Czech for example then you need to make the character set
utf8_czech_ci to get full language function.
I found the following URLs regarding encoding hope they will help: http://dev.mysql.com/doc/refman/5.0/en/charset-charsets.html http://dev.mysql.com/doc/refman/5.0/en/charset-unicode-sets.html http://code.openark.org/blog/mysql/mysqls-character-sets-and-collations-demystified
the following URL contains some examples about disk usage for character set: