coverting numbers to strings

François Pinard pinard at iro.umontreal.ca
Sun May 28 09:18:57 EDT 2000


nospam.newton at gmx.li (Philip 'Yes, that's my address' Newton) writes:

> The sending end -- he's sending out the posts UTF-7-encoded, which is
> even worse to read than quoted-printable in my opinion.

If you have MIME properly installed, quoted-printable is neither good or bad,
it just works.  The same goes for UTF-7, yet I agree with you that support
is relatively newer, even if the RFC has been around for many years by now.

> [...] the article wasn't even flagged as being UTF-7-encoded!

This is bad, indeed.

> Z 3 Penguin, please turn this off.  Not everyone can read UTF-7.

Or at least, make sure your MIME headers properly describe your contents.
We could surely dive into deadly debates about UTF-7.  Take it as progress.

MIME is not perfect (despite some fanatics say it is), but it is usable
if properly implemented.  Some MIME implementations stink as implementors
do not read the RFC, and Microsoft purposely deviates from them, as they
always do, a way to break competition.  Hopefully, there are some good MIME
implementations, others will improve, these things will sort themselves
over time.

The same goes for UTF-7 and UTF-8 email: this is a way towards Unicode,
which is not perfect either (despite some fanatics really believe it is), but
might be mildly usable in most countries.  Yet, Unicode implementations are
usually improper.  Worse, Microsoft is behind Unicode, all prone to raising
confusion wherever it breaks better approaches that competition offers.
So Unicode itself is a bit improper.  Things will surely improve over time,
yet it is less evident, today, that everything will sort out in the long run.

Despite doubting Unicode, this step towards a better universality should
be taken, even if it implies various suffering on the way.  There is a bit
of unavoidable suffering behind any progress, that we should accept as a
kind of contribution to that progress, and to be part of our own history!

-- 
François Pinard   http://www.iro.umontreal.ca/~pinard






More information about the Python-list mailing list