From: J.B. Moreno (planb@newsreaders.com)
Date: Wed Mar 05 2003 - 15:08:19 CST
On 3/5/03 2:44 PM, Dirk Nimmich at <nimmich@muenster.de> wrote:
> J.B. Moreno wrote:
>> Clearly the "just send 8 bit" crowd are a large portion of those
>> who aren't using English.
>
> The question is why the crowd does just send 8 bit. My assumption is
> that the main reason for this is that most software does it by
> default though it would be able to use the mechanisms described in
> RFC 2047. (Sending 8 bit chars in the body is ok as long as the
> transfer encoding and the charset are declared in the header.) Your
> housewives in Korea don't know how to change that default and would
> happily "use" RFC 2047 in headers if that was the default.
I'm not one to argue that defaults are very important, but they generally
get set for a reason -- and I think that in this case the reason was the
howls of outrage from those that couldn't decode it.
Sure there are likely to be vastly fewer howls if 2047 was used now, but
given that the /existing/ situation is to just send 8 bits, I think that the
protests that /will/ happen will prevent it from being used.
We've already had one backlash from a lousy introduction of MIME, do we
really need to have another.
And punycode or 2047 -- the "default" charset will need to be UTF8 (posting
to a newsgroup such as "alt.how-to-make-a-euro¢" where the ¢ has different
byte values in different charsets, just won't work). So all we've gained by
using 2047 is to avoid some problems where mail intersects news, at the cost
of courting rejection.
Whether it is an encoding or not, raw UTF8 will be seen as "just another
charset", but one that (a) has all of the characters they are concerned with
and (b) has the blessing of the internet standards groups. It's my
belief/conclusion that these two facts will result in little resistance when
individual housewives are told to switch to UTF8.
-- J.B. Moreno