> > It is quite well possible to typeset a Japanese textbook for the Chinese
> > language with a single Unicode font and no language tags. Some typography
> > perfectionists will be unhappy if you do that, but everybody will be able
> > to read everything correctly without difficulty.
>
> On what basis can you insist such a thing? Are you a native
> Chinese, Korean, or Japanese speaker? Or, are you studying
> CJK languages as a foreign language?
I am very familiar with zh_CN, zh_TW and ja_JP, which may be my problem:
I can perhaps not imagine the viewpoint of a Japanese reader who has never
read anything except monbusho standardised Japanese -- no Kanbun, no
old Japanese books, no calligraphy. There may of course be more such
people than I realize.
> No, it is against the fact. For example, average Japanese people
> cannot read Chinese character of U+76F4. "will be"?
That is still a very small difference. But it can of course strike
someone completely used to variation as strange. Certainly not as
unreadable in context (e.g. "shourai" "jiang1lai2"), especially not a c+j
context in which various shapes appear. If I was to typeset a Beginners'
Chinese textbook for the most unsophisticated possible Japanese reader
under XTerm without language tags, I'd use a nice Unifont that sticks to
the Japanese glyph variant where differences exist.
> If you succeed to change Japanese education system and then wait for
> 80 years until all living Japanese people will have been educated by
> your system. After that, we can have common unified CJK fonts.
My point was that one need not go too far in trying to be perfect on these
issues in environments such as XTerm. But as it turned out, there was no
need to discuss about this at all, since perfection can be achieved using
the language tag.
It is of course a bit unfortunate that we have to load different fonts for
all the tags. At least in a system installation environment that is
cumbersome. If all the variants were resolved, as you seem to be
proposing to the IRG people, one single font could be used for perfect
xterm CJK. One can now of course try to introduce variant into the fonts
themselves, but that again seems rather dirty. Or one can recognize that
there is some minor unresolved limitation left over and either hop that it
can be removed by some future Monbusho reform or just accept it. Anyway
it may not be so difficult to accept that it is currently impossible to
typeset a "Beginning Chinese for Japanese people" textbook on an xterm
from within a Linux installation disk with limited memory.
Sorry to have participated in this flamewar and to have made it more
complicated. Maybe some of the thoughts about the priority of unifonts in
XTerm are relevant nevertheless.
-phm
-
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/lists/
