Re: FWD FYI: is ISO-2022-CN actually used?

[Markus S. <mar...@jt...>]

Fred.Batty@Sun.COM wrote:
> Our Messaging Server supports the charset and, although we don't have any
> data on what percentage of emails are encoded in that charset, in order to
> maintain compatibility we need to continue supporting it as long as the
> charset continues to be in use, even if by a small community of users.

First, while I did propose in a separate email to remove the _generic_ ISO-2022 converter from ICU 
2.8, I am currently only exploring whether the ISO-2022-CN charset is used at all.

I am not proposing to remove this converter from ICU 2.8, although I may propose to remove it in a 
future ICU release if the feedback to this question here and on the unicode mailing list indicates 
that ISO-2022-CN is not used. (So far, the unicode list feedback confirms this.)

In fact, I am currently working to fix a number of bugs in the ISO-2022-CN converter. The 
maintenance involved with this and the size of the associated conversion tables lead me to ask the 
question about usage.

This should match what you say above: "as long as the charset continues to be in use, even if by a 
small community of users" - surely if the charset is not used at all except for isolated experiments 
you would not need to support it yourself.

> ... If for example an old some Chinese university mailing list
> system sends email in that charset to users of our Messaging Server, and
> ICU did not support it, then the emails would appear as garbage to these
> users. ...

Please note that ISO-2022-CN is quite "new" - RFC 1922 dates from 1996, when Chinese users already 
used GB2312/EUC-CN, GBK and Big5, and when UTF-8 was becoming supported.

In a way, ISO-2022-CN was created at a time when the need for it was evaporating.

Best regards,
markus

-- 
Opinions expressed here may not reflect my company's positions unless otherwise noted.