On Sat, Jun 15, 2002 at 02:59:08PM +0200, Tony L. Svanstrom wrote: | On Sat, 15 Jun 2002 the voices made Michael Moncur write: | | > When a new release comes out I like to be anal-retentive and go through the | > GA second-guessing its scores. This is my report for 2.30. | | > - This works well for me but users in some countries may want to change it | > score SUBJ_FULL_OF_8BITS 4.298 | | A sign that: | a) RFC != IRL. | b) SA is focused on Englishspeaking users. | c) That I shouldn't decode headers before them reaching SA. | d) All of the above.
#a isn't going to change as long as you encourage it. Positive
reinforcement vs. Negative Reinforcement.
Not only that, but those who break the rules get in trouble. Not all
MTAs can or will happily pass along improperly tagged 8-bit data.
Anyone here use demon as their ISP? In addition to mmdf fully
adhereing to RFC 821 (SMTP is a 7-bit transport you know), postifx and
exim have the ability to enforce RFC 2047 too, and a number of sites
do.
#b is not true. SA provides lots of configuration as to which
languages and locales you legitimately receive so that it doesn't
wreak havoc on your native-language messages just because I can't read
them and only see them in spam.
Also, RFC 2047 allows non-english data in headers with the proper
encoding. SA fully supports RFC 2047. Just because many
people/software producers like to break the rules when the feel like
it doesn't mean it is Right. It IS nonsensical to have 8bit
characters in message headers because the recipient's software has NO
way of knowing which charset to pull the glyphs from. That is why
RFC2047 was created; to support the non-english users in email and
allow software to decently handle it. If _you_ like to break the
rules too, you can do so by modifying your version of SA.
-D
--
"He is no fool who gives up what he cannot keep to gain what he cannot lose."
--Jim Elliot
http://dman.ddts.net/~dman/
msg06394/pgp00000.pgp
Description: PGP signature
