From: Jean-Marc Desperrier (jean-marc.desperrier@certplus.com)
Date: Wed Jul 03 2002 - 12:32:16 CDT
Charles Lindsey wrote:
>No, I think it remains true that there are newsreaders out there that do
>not understand RFC 2047. Heck, all those 152,100 articles that Andrew
>found must have come from user agents that would not or could not
>_generate_ RFC 2047, in which case I doubt they were able to _decode_ it
>either.
>
Quite true, but this is another rat's nest.
They will not be happier with UTF-8, than with RFC2047.
From what I've seen for exemple, tw.bbs.* seems to be nothing but a
gateway between bbs system, and either of the two proposal will surely
gener quite an uproar by them, because I don't believe the bbs system
will be updated to support either.
They could update the gateway though.
>And why do all those articles with naked ISO-8859-1 headers appear
>in the French newsgroups if it is not because of the existence of
>newsreaders that are incapable of generating ot decoding RFC 2047, or
>both?
>
Because it was the recommended usage to use naked ISO-8859-1 headers in
the french hierarchy, but this is moving now.
First because most newsreaders commonly used recognise RFC 2047, but
another reason for the change is surprising.
It's just that google groups refuses to recognise subject in naked
ISO-8859-1, and when people respond to messages using google groups, all
accentuated characters turns to '?'.
The bad consequence of this, lots of subject garbled with ?, suffice to
see several of the "dinosaurs" of fr.* who have being fighting RFC 2047
all along, know favor it over naked ISO-8859-1.
Very surprising turn-out.
And an argument in favor of incompatible change, as long as they are
reasonnably acceptable way to adapt to the incompatible change.