[leafnode-list] Re: Base64 coding in list messages
Peter J Ross
peadar.ruadh at gmail.com
Tue Oct 21 22:38:45 CEST 2008
On Tuesday 21 October 2008 20:35:50 Christian Ebert wrote:
> * Whiskers on Tuesday, October 21, 2008 at 20:17:04 +0100
>
> > On Tue, 21 Oct 2008 19:34:26 +0100 Peter J Ross <peadar.ruadh at gmail.com>
> >
> > wrote:
> >> I use UTF-8 whenever non-ASCII characters are required (because I'm
> >> living in the 21st Century),
> >
> > Same here (although up till a year ago I was still using the various ISO
> > codings as my console-based software couldn't handle UTF-8).
> >
> >> but I'm not sending raw 8bit data. I don't
> >> understand why messages sent as quoted-printable should be re-encoded as
> >> Base64.
> >
> > That's what I'm hoping to discover. Base64 uses about twice the
> > 'bandwidth' of 'quoted-printable', for one thing. The only email
> > software I've come across that uses Base64, is the email program on my
> > Nokia mobile phone (which is a pretty poor excuse for an email program).
>
> Might be Mailman, if it behaves similar to the Python email
> module. There the default content-transfer-encoding for utf-8 is
> base64.
Why? What's wrong with QP?
> Perhaps to ensure data integrity just in case a charset
> declaration is, ahem, neglected somewhere. iso-8859-1,
> windows-1252 and friends are mapped to QP.
>
> And: utf uses slightly more bandwidth than iso-8859-1 ... <duck>
If UTF-8 were more widely used, conversion wouldn't be needed, so bandwidth
would be saved. :-)
--
PJR :-)
More information about the leafnode-list
mailing list