Version:            (using KDE 4.4.0)
Installed from:    openSUSE RPMs

This is likely going to be a rather time-consuming task, but would increase the appeal of Kontact immensely for Japanese, Chinese as well as Korean users (possibly others as well?):

It would be very useful for Japanese, Chinese and Korean speakers, or users of
the respective languages, to have an additional "Reading of name" fields in
Kontact's address book.

While it is sufficient for most languages to have single fields for last name,
first name, company name, etc, Japanese, Chinese and Korean names distinguish
between the (a) characters the name is written with (usually Chinese logograms,
called Hanzi in Chinese, Kanji in Japanese, or Hanja in Korean), and (b) the
reading = pronunciation of the name (for which different scripts are used,
including latin characters, Japanese hiragana or katagana, and Korean hangul
characters).

Users of these languages would, therefore, benefit greatly if additional fields
for the 'reading of name' were created in Kontacts address book. Which fields
receive such a field should be in line with localized operating systems for
PDAs / mobile phones in those countries in order to facilitate synchronization.
The would probably include (based on Japanese Palm-OS):
     1. last name
     2. first name
     3. company name

It should be address that internally, the extra fields are usually stored as
part of the name fields, but separated with special characters.

If address book entries make use of such extra fields, sorting of names ought
to be based on the 'reading of name' fields, not on the name fields themselves.

Since such extra fields might distract users of other languages, a nice
solution could be to have a check box on the address book entry whose clicking
would open up / add the additional fields (while leaving it unchecked would
hide the fields).

(I'm refiling this feature request for KDE 4.4. Filed it previously for an earlier version as Bug 197800)