Message-Id: <199607231057.GAA04472@fliptop.pilsnet.sunet.se>
To: Keith Moore <moore@cs.utk.edu>
cc: ietf-drums <drums@cs.utk.edu>
Subject: Re: Recommending use of 8-bit transport
In-reply-to: Your message of "Tue, 23 Jul 1996 00:37:48 EDT."
             <199607230437.AAA01501@ig.cs.utk.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
Date: Tue, 23 Jul 1996 12:57:43 +0200
From: Lars-Johan Liman <liman@sunet.se>

moore@cs.utk.edu:
> The 8BITMIME spec does not require this, and I propose that we don't
> try to walk down that path again.

OK, not here.

> Instead, I suggest:

> a) if a site cares about receiving non-ASCII mail, it should make
> ...

I think I can agree with this.

> b) if a recipient forwards mail from his mailbox to somewhere else,
>    that recipient is responsible for arranging either for an 8-bit
>    clean path, or conversion of the message to 7-bit.

> My assumption is that we'll get (nearly) universal, 8-bit transparency 
> more quickly by allowing 8BITMIME implementations to simply be 8-bit clean, 
> than we will if we require them to do 8-to-7 conversion.  

At first this made me feel uncomfortable. To me, the C-T-E system and
8BITMIME was created to overcome the old 7bit limitation, and to give
implementors the framework to do automatic conversion, but maybe we
have reached the point in time when 8bit transportation is in such
majority that we can slowly, slowly start to deprecate 7bit
transportation. A first step of doing this would be to adopt your
point of view. Earlier, 7bit was the default, and if you wanted 8bit
you had to do something manually. From there we went to "automatic
conversion", where we are now. The next logical step is "if you want
7bit, you have to do it by hand", and then finally (in years to come)
rule out 7bit entirely. Yes, maybe the time has come to start taking
the next step ...

> I also believe that we'll have far less damage to mail messages if
> we don't do conversion very often.

Agreed to 100%! I would like to quote Stef, who at the Toronto IETF
said something that struck me as one of the most fundamental truths of
the electronic mail world:

  "I have never seen a message that was improved
   by passing it to a gateway."

				/Liman