著者: Steinar Bang 日付: To: exim-users 題目: Re: [Exim] 8bitmime?
>>>>> Phil Pennock <Phil.Pennock@???>:
> On 2002-03-21 at 07:55 +0100, Steinar Bang wrote: >> Back in 1993, when I first started using it, I got yelled at for
>> sending unreadable messages, because at the time "just-send-8" was
>> the common practice here in Norway. > Fine. One country had, at one time, an accepted viewpoint which was
> not that in the standards or that followed by the rest of the world.
IIRC it wasn't just Norway. This was common practice in the rest of
the Nordic countries.
Please note: nobody argued that this was a scalable solution, but it
was a solution that made things work at that point in time.
Also note that I'm not against an MTA doing 8bit to q-p conversion if
it determines that the receipient isn't 8bit clean, and it has an 8bit
message to send. I'm against bit-stripping if the receipient doesn't
explicitly say that it is 8bit clean.
I'm also against bit-stripping on the receipient side (which is where
you come in, I guess). My gut instinct is to let the 8th bit live as
long as possible.
[snip!] >> Eg. in Netscape 4.x this setting affects both mail and news, and you
>> *will* get yelled at if you use q-p on Norwegian USENET groups. > How are USENET policies relevant to what options Exim should
> advertise when speaking SMTP?
A limitation in a combined MUA/newsreader cause people to make a setup
choice that sends messages with C-T-E 8bit to receipients in eg. UK or
Holland.
These 8bit messages should be handled as RFC compliant as possible.