Message-Id: <v0300781dae353c1c1971@resnick1.isdn.uiuc.edu>
In-Reply-To: <9608121856.AA29264@leftbank.com>
References: <v03007818ae3522720fc8@resnick1.isdn.uiuc.edu> from "Pete
 Resnick" at Aug 12, 96 01:19:15 pm
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Mon, 12 Aug 1996 15:08:15 -0500
To: cos@leftbank.com
From: Pete Resnick <presnick@qualcomm.com>
Subject: Re: Characters and octets and 822bis
Cc: drums@cs.utk.edu

On 8/12/96 at 1:56 PM -0500, Ofer Inbar wrote:

>If I send email with a body composed of 7-bit characters which make no
>sense when interpreted as US-ASCII, but do when interpreted as some
>other encoding, am I violating RFC822?

No, absolutely not. That is perfectly legal, assuming that what you mean by
"make no sense" is "wouldn't make sense to a reader if they were displayed
as simply US-ASCII character glyphs".

>Isn't this what MIME can be used to do?

Right. But the key for 822 is that, unless otherwise labeled by some
mechanism like MIME, the data in the body *is* to be interpreted as
US-ASCII characters in the range 0 through 127. How to interpret data in
the body that is outside of that range, or how to interpret the characters
that are within that range as having meanings other than as US-ASCII
characters, is beyond the scope of 822.

>Did you actually mean US-ASCII, or are you using it as
>shorthand for <00>..<7F>?  I guess there are different levels of
>"interpretation" here, and at one level you could claim that any
>message body composed of <00>..<7F> is US-ASCII, but that's confusing.

Well, yes, I am using US-ASCII to a certain extent as a shorthand for
"characters in the range 0 through 127" (I'm trying to stay away from hex
since it implies octets, which I would rather avoid), but as I said above,
any interpretation of characters in a message other than a US-ASCII
interpretation is beyond the scope of the 822 document.

Does this make sense?

pr

--
Pete Resnick <mailto:presnick@qualcomm.com>
QUALCOMM Incorporated
Work: (217)337-6377 / Fax: (217)337-1980