Lawrence Greenfield wrote:
Newsgroup names are a very hard problem. Note that the way Usenet works
right now requires these get passed gracefully through e-mail: in
moderated newsgroups posts are e-mailed to the moderator.
Not at all difficult when one realizes that the newsgroup "name" isn't
really a name at all; it's just a goup of tags that identify a place
in a hierarchy. Many of the existing newsgroup component tags are
derived from abbreviations, contractions, or acronyms.
I'm not sure I understand the NEED for the "canonical" name to be pure
UTF-8 (that is, encoded Unicode) instead of some other encoding.
Exactly -- there is no real "need", any more than there is a "need" for
ISO 639-2 language tags to be anything other than what they are; comprised
of a set of characters which are amenable to being used in (among other
places) email header fields. Nobody -- at least nobody worth paying any
attention to -- is throwing tantrums because the language tag "ja" isn't
comprised of some utf-8 (or JIS) sequence. For display / UI purposes
there is no reason why client software couldn't display the appropriate
Kanji in a menu, etc. where the language tag "ja" is recognized, and it
is precisely to the user agent (clent) software where such matters should
be constrained. Ditto for charset tags, header field names, various
keywords that appear in various headers, etc.
Indeed, the existing unencoded, non-Unicode tags are quite convenient
for news administrators, as typical filesystem tools will work fine with
the tags on a wide variety of systems. That is not necessarily the case
for utf-8 encoded tags, and the Usefor WG would be well advised to consider
the IMAP experience (elided here for brevity).