Koha cannot describe 650, 651 etc tags in the form “$a SUBJECT $z PLACE $x TOPIC” (i.e., z subfield first, before x subfield, e.g., “650 -0 a Women $z China $x Social conditions”). This must be rectified. Otherwise Koha's MARC support is no better than CBS's.

Similarly, Koha cannot describe 650, 651 etc tags in the form “$a SUBJECT $x TOPIC $y TIME x ANOTHER_TOPIC” (same subfield occuring more than once in the same tag, and ordering is significant). This also must be rectified. Otherwise Koha's MARC support is no better than CBS's.

Dewey should be varchar, not double. This fixes a Dewey problem, but should also immediately make it possible to use LC call numbers in Koha.

A second set of borrower's names, i.e., allow both English and Chinese names to be entered and searched in the database. Currently, Koha can only gracefully handle names in English (or other Western alphabets).

Detect the OPAC language according to the browser settings - i.e., multilingual interface choosable by the user

The library should be able to specify a maximum fine. In both the TPL and the CCC library, there is a maximum amount of fines that can be charged; Koha does not seem to currently support this.