Getting a bit OT here, but what the heck.

The Unicode standard has always guaranteed a 1:1 two-way mapping with existing character sets, and it contains a number of duplicate characters (in special compatibility regions) to cope with precisely the problem you mention.

No, the perceived problem with Unicode, when it was first introduced, was that certain characters are traditionally written slightly differently in China and in Japan, and Unicode unified each set of equivalent characters to have a single codepoint. IIRC the Japanese mistakenly assumed that this meant Unicode was trying to force them to use Chinese character shapes, which would obviously have signalled the end of the world.

Once an anti-Unicode movement had come into existence, they subsequently came up with various other kinds of anti-Unicode FUD; for example, one popular criticism is that Unicode does not necessarily include certain incredibly rare characters used only for writing ancient personal names (not that these were included in the then-standard JIS encodings either), while another, slightly more valid complaint was that UTF-8 required three bytes for every Japanese character rather than two bytes in Shift_JIS (of course, UTF-16 solves that).

Anyway, the upshot of all this was that many came to believe that Unicode was inherently unsuitable for encoding Japanese text, hence Ruby's failure to support it initially. Thankfully the resistance seems to be dying, perhaps aided by the fact that the Japanese language still exists even though Windows has been using Unicode internally for many years now...