At 08:26 AM 9/19/01 -0700, ned(_dot_)freed(_at_)mrochek(_dot_)com wrote:
How does one reliably track down all of the defined RFC [2]822
headers? That is, when proposing a new header, how is one to be sure that
it is not already defined somewhere?
Simple: Search all the RFCs and internet-drafts. This is hardly rocket
science.
But as a practical matter the real issue isn't standardized use, or even
proposed use in a standard. It is the fact that anyone can use any field name
at any time without any specification whatsoever. This is much more difficult
to gauge. Personally, I do it by keeping track of all the fields I encounter
that seem reasonable -- 511 so far.
OK, I oversimplified my motivation for considering this.
There are two reasons that I have wanted there to be a header registry:
- to have a way to find if there is an existing header that serves some
given purpose. RFC2076 and subsequent drafts serve this purpose pretty
well, but there's no mechanism for catching new headers as they are defined.
- to have a way to transfer RFC822 labelling and semantics to different
environments. Currently, there is no easy reference to say "here are the
header names and this is what they mean". My immediate interest here is
message metadata for message archive/retrieval type of systems, using an
XML-based format (e.g. draft-klyne-message-rfc822-xml-02.txt). I can
imagine other possibilities.
Seems to me like a lot of work that won't really address the problem.
I think Keith's suggestion is a manageable approach.
#g
------------------------------------------------------------
Graham Klyne Baltimore Technologies
Strategic Research Content Security Group
<Graham(_dot_)Klyne(_at_)Baltimore(_dot_)com> <http://www.mimesweeper.com>
<http://www.baltimore.com>
------------------------------------------------------------