Keith Moore <moore(_at_)cs(_dot_)utk(_dot_)edu> writes:
Too many WGs have made the mistake in thinking that their consensus
would be acceptable to the wider community, or could be made acceptable
without major changes. More WGs need to find out what is acceptable to
the wider community long before they go to Last Call.
On top of that, the current consensus, such as it is within USEFOR, is a
consensus of exhaustion, and somewhat marginal even at that. It's been
years, we still don't have any sort of base document, and arguments like
this one have driven many people off of the mailing list entirely.
Yes, this is unfortunately not an uncommon problem. It is all the more
unfortunate because all those stray chickens tend to come home to roost during
last call. Or worse, even later.
I've personally just ran out of energy to argue about the right way to do
non-ASCII newsgroup names. There's some huge amount of complexity that's
going to get shoved somewhere, and there are fewer gateways than there are
news clients or news servers, so I suppose it makes some sense to shove
the complexity there. It seems unlikely to me that the current approach
is going to actually work that well, but I ran out of anything better to
offer that's not unacceptable to some other group of people.
Newgroup names are an interesting case because they aren't something email
cares about. While an 8bit newsgroup name in some news-specific header
field may not survive a trip into and out of email, at least it doesn't
trash any of the email-specific pparts of the message.
An argument to have this one thing be in 8bit would still be a very tough sell,
but far far easier than arguing that it is OK to have 8bit fields that are
shared between email and netnews.
I never understood what was wrong with RFC 2047 encoding, at least for
everything except the Newsgroups header (which poses some special
challenges), but many other people on USEFOR seem to feel that it would be
a catastrophically bad decision, and as I only speak English, I don't feel
particularly well-qualified to comment.
There was plenty of input from non-English speakers during the design of MIME.
Indeed, the original motivation that started the MIME work was to
"internationalize email". We ended up going a bit further than that...
The majority of folks on USEFOR also seem to me to have bought
wholeheartedly into the concept of Unicode and the belief that Unicode
will displace all other character sets in the near future, that all
software will soon expect any untagged 8-bit data to be in Unicode, and
that handling Unicode is all that's necessary to move forward into the
non-ASCII world. I part company some on that point; these days, Unicode
looks to me a lot more like IPv6, namely a better technical concept that
may eventually get quietly deployed but that isn't going to change the
world any time soon.
I hate to have to say it, but I'm with you on this as well. I believe Unicode
and more specifically UTF-8 is the best answer we have in hand at present and
as such constitute a preferred solution of sorts, but this doesn't mean it has
the ability to magically displace stuff that's already out there. On the
contrary, use of all sorts of other charsets continues unabated, and speaking
as someone who deals with this stuff regularly as part of my day job, attempts
to actually deploy UTF-8 as a solution are unsuccessful on a fairly regular
basis. The necessary will just isn't there.
My only hopes for significant displacement of other charsets arise from the
develope of UTF-8 only solutions in other protocol spaces. These uses put
pressure on older protocols to trade up. But I have to admit the results of
such pressure are taking a long time to show up.
Ned