Matt Gushee wrote:

So the upshot of this is that, though jis-x-0212 glyphs make up a very small proportion of the Japanese text that is printed (I'd guess 1-2 percent), a large proportion of documents (40-50 percent, maybe) require one or more glyphs from that set. So that's another 8000 glyphs, if you want to do it right.

I don't know how much we need JIS X 0212 support. I always thought that it wasn't much used at all, mainly because Microsoft doesn't seem to support it with their Shift-JIS encoding. Also, it used to be hard to find fonts containing those glyphs, or to find an editor capable of editing them. But I have to admit, I don't know that much about Japanese to know how much those glyphs are needed. Like Matt said, maybe for names and technical documents, the need for JIS X 0212 glyphs is higher. The use of Japanese/Chinese characters is also a really interesting topic. Many people used to believe that the invention and widespread use of the computer would mean the end of those characters, but in reality, it is the opposite. Thanks to advanced Input Method Editors, rare and 'obscure' characters get used even more, since they can easily be accessed. So JIS X 0212 support in ConTeXt would cool to have. It's also quite interesting to see how JIS X 0212 gets encoded. It's part of EUC-JP CodeSet 3. If an implementation of Japanese in ConTeXt supports EUC-JP, and needs to support JIS X 0212, we need to have a ConTeXt switching mechanism to change from a two-byte encoding (EUC-JP CodeSet 0,1 and 2 for JIS X 0201 and 0208 support) to a three-byte encoding (EUC-JP CodeSet 3 for JIS X 0212). This could be an interesting challenge to implement, but I don't see a reason why it shouldn't be possible. Naturally, UTF-8 supports everything we'll ever going to need! :-) My best, Tim