Re: document comments

[Peter Svanberg <psv@xxxxxxxxxxx>]

Markku Savela <savela@tel4.tel.vtt.fi> writes:
> On the other hand, as you say, we here in Nordic countries will
> continue to use national variant for quite while yet. Instead of
> letting the fuzziness of ISO 646+variants to continue, we could
> perphaps register some of the national variants, like some specific
> Swedish version as "CHARSET=ISO-646-SE"?

...and Ran Atkinson <atkinson@tengwar.itd.nrl.navy.mil> writes:

> However, RFC-822 is extremely explicit in stating ANSI X3.4 _NOT_
> ISO-646 or anything else.  By explicitly mentioning ISO 646 we encourage
> its use which is undesirable.  This is an old old item for this
> group and for MIME and there was an explicit decision last spring
> to deprecate the improper uses of ISO-646 in favor of MIME.  One
> _can_ use MIME to pass ISO-646 by having a local convention of
> "X-ISO-646-foo" to denote the particular national usage as a Charset
> identifier.
> 
> Passing ISO-646 without an explicit charset header per MIME is not
> currently interoperable and will tend to decrease interoperability
> the more it is used.

...and others with about the same message.

---Concerning MIME-labelled non-ASCII letters

MIME also states:
    It is the opinion of the IETF working group that a large number of
    character sets is NOT a good thing.
and
    The omission of the ISO 646 character set [as a registered character
    set name] is deliberate in  this regard.

but I agree that Charset=ASCII should be ASCII! How should the
646-variants be labelled then? Officially register them (I don't know
how many there would be a need for) or encourage non-official interim
X-type names?

---Concerning non-MIME-labelled non-ASCII letters

Unless we use "gateways" to filter messages, these suggestions anyhow
implies making all our UAs (slightly) MIME-aware. I don't think that
will happen - for some type of sites - much faster than Latin 1 and
full MIME awareness. So, non-MIME-aware UAs will continue to send out
non-MIME letters - in the Nordic countries as in the USA. The
difference is that the implicit character encoding in (some of) the
letters coming from our UAs is not ASCII.

(In this sense MIME is not backwards compatible for us - we must
change things to get the same functionality as we had before. I
suppose this fact is an as unavoidable consequence of our
non-822-conformance as our non-822-conformance was of our languages
non-conformance to English ;-).)

It was not my intention to start a big discussion about ISO 646 etc.
I just wanted to clarify these problems. This also partly forestalls a
just started internal discussion and planning for how to initiate MIME
in our countries. We will return (on the MIME list) when we have
suggestions to discuss.
---
Peter Svanberg, NADA, KTH		    Email: psv@nada.kth.se
Dept of Num An & CS,
Royal Inst of Tech			    Phone: +46 8 790 71 46
S-100 44  Stockholm, SWEDEN		    Fax:   +46 8 790 09 30