Who said that we'd have to replace the current single character tags
with other single character tags?

There were a couple of threads last November about changing o= to use
words and splitting out the semantics around the various semantic axes
that are covered by the value (sending mail, signature presence, 3rd
party signatures, redirect).

One message that I wrote had a table, showing what a couple possible
syntaxes might be for o=
(http://mipassoc.org/pipermail/ietf-dkim/2005q4/001307.html, slightly
modified here):

                o=-signature,-3ps       signature=never,3ps=never
?/WEAK          o=~signature,-3ps       signature=optional,3ps=never
~/NEUTRAL       o=~signature,+3ps       signature=optional,3ps=allowed
!/EXCLUSIVE     o=+signature,-3ps       signature=always,3ps=never
-/STRONG        o=+signature,+3ps       signature=always,3ps=allowed
./NEVER         o=nomail                nomail
^/USER          o=user                  checkuser

William Leibzon also had suggestions along the same lines
(http://mipassoc.org/pipermail/ietf-dkim/2005q4/001306.html):

>  1. Signature required/optional:
>    sig=MUST/SHOULD/NEVER/USER (sig=STRONG/NEUTRAL/NEVER)
>  2. 3rd parties allowed/not
>    3ps=ALLOW/DENY/USER
>
> (Or if you like o=STRONG/3PS | o=NEUTRAL/NO3PS | o=USER/USER)

Hector Santos had several suggestions for additional values in the 3ps
space. (http://mipassoc.org/pipermail/ietf-dkim/2005q4/001318.html):

> signature=always, 3PS=IGNORE  -- Keep Original, don't strip, resign
> signature=always, 3PS=APPEND  -- Append, don't strip or replace.
> signature=always, 3PS=RESIGN  -- strip and replace.

        Tony Hansen
        [EMAIL PROTECTED]

Michael Thomas wrote:
> For two reasons:
> 
> 1) With my developers hat on, I couldn't really care the least:
>    if you have to look them up, you probably need to look up the
>    other single character tags too. This is just a matter of being
>    familiar with the spec, and h, z, b, m etc are all equally
>    opaque IMO.
> 
> 2) I'm guessing that we will utterly fail to have a single word that
>    describes the rich semeantics of the policy attached to whatever
>    symbol we choose. Witness this latest brouhaha with "exclusive"
>    which is not even part of the current draft. An abstract symbol
>    which has no baggage of its own and is, in fact, just a pointer
>    to the normative text seems like a better way to avoid
>    misinterpretation.

_______________________________________________
NOTE WELL: This list operates according to 
http://dkim.org/ietf-list-rules.html

Reply via email to