To: "Daniel W. Connolly" <connolly@hal.com>
Subject: Re: <URL:...> considered harmful
In-Reply-To: Your message of "Tue, 13 Sep 1994 00:09:09 EST."
<9409130509.AA06650@austin2.hal.com>
Date: Tue, 13 Sep 1994 10:58:16 -0500
Message-Id: <26565.779471896@silky.cs.indiana.edu>
From: Marc VanHeyningen <mvanheyn@cs.indiana.edu>
Dan says:
> In message <16569.779430042@hound.cs.indiana.edu>, Marc VanHeyningen writes:
> >>Is the following
> >>a URL or a message id:
> >>
> >> <URL:my-scheme://foo.bar/abc@foo.com>
> >>
> >>It satisfies the syntax of both, since the path syntax of my-scheme
> >>might allow '@' characters.
The above is simply false; colons are not permitted in Message-IDs since
they are specials in RFC 822 (unless they're quoted, which is not the
case in this proposal.) Or did you mean some other spec for Message-IDs?
> >So, the problem is possible spurious URL identification.
>
> And spontaneous message ID identification. By choosing <URL:...>,
> we are breaking existing code that picks message id's out of
> plain text. I see no motivation to do this.
There are already tons of things (like email addresses, pseudo-SGML
constructs, BNF grammars, and people simply delimiting things with
angle brackets because they <like> it) that may confuse this code.
(To the extent that it exists; most code like this looks for ids in
In-Reply-To: and References: headers, not plaintext.)