But aren't we forgetting something here? Would UTF be readable in
Japan? Japan has a large and established email community. We can't
simply ignore them.
Erik
I did not advocate getting rid of current codesets used for mail. If
I did, I would advocate getting rid of both 7-bit ASCII and the
Japanese 2022 stuff. I am just saying that we should stop inventing
new ones in favor of one which works for all languages. As with most
things it will take time to migrate fully. But at least we can set a
sensible direction for the future.
Walt
Thanks for the reply, Walt.
Let's take the discussion further.
OK, so you seem to agree that we should not attempt to "get rid of"
the character encodings currently used for email. And you seem to want
us to migrate to an encoding that "works for all languages".
There are several questions that we may wish to ask ourselves. Is it
really necessary to have an encoding for "all" languages? Currently,
most of the email in the world (in terms of volume) is in either one
language (e.g. ASCII), two languages (e.g. the Japanese subset of ISO
2022), or several languages (e.g. Latin-1). As far as I know, there is
no large volume of email that uses "many" languages simultaneously.
There does not seem to be so much demand for truly multilingual email.
We have seen some people demanding a small enhancement. For example,
there are people in Japan that wish to use Hangul (Korean) together
with the current Japanese+English.
For these people, the easiest implementation (by far) is to simply
extend the current ISO 2022 subset. The Japanese currently use ESC$B
(that's 3 bytes, escape, $, B) for Japanese, and ESC(B for English. So
all they need to do is add a few lines of source code to the existing
software, to make it understand the sequence for Korean (i.e. ESC$(C).
Fonts are already available for Korean in that encoding, so there is
no need for complex mappings between the transmission code and the
rendering (display) layer.
On the other hand, if these people used an encoding based on Unicode
(i.e. level 2 of the 2-octet form of 10646), then users of older
software will not be able to read *any* part of the message. So they
would only send such messages to people that already have the right
software. The problem is that it is much more work to modify software
for Unicode than an ISO 2022 extension. In fact, there already exists
a version of Emacs for English+Japanese+Korean, using 2022. One of the
nice things about this is that it does not necessitate changes to the
email and netnews software. Do you seriously think that Unicode
proponents can change this trend? Or would you say that Unicode would
be used for *other* purposes? Perhaps in *other* countries? If so,
which purposes? And which countries?
Regards,
Erik