In <20050601151226(_dot_)GA30578(_at_)primefactor(_dot_)com> Mark Shewmaker
<mark(_at_)primefactor(_dot_)com> writes:
On Wed, Jun 01, 2005 at 08:38:00AM -0500, wayne wrote:
In <20050531180225(_dot_)GA7253(_at_)primefactor(_dot_)com> Mark Shewmaker
<mark(_at_)primefactor(_dot_)com> writes:
Everywhere else in the I-D, the term "identity" is used.
But didn't you just change the document to use identity elsewhere too?
No, I don't think so. The term "identity" has been used since
draft-lentczner-spf-00. There really wasn't a consistent naming
scheme in draft-mengwong-spf-*, neither "scope" nor "identity" is
used.
I haven't thought of a good set of words to use yet, [...]
At this stage of the game, I'm very reluctant to introduce new
concepts.
Even for optional info for a trace header--just so a naming convention
would already exist? (I admit this is a couple steps past adding
"scope=" or "identity=".)
IMHO, yes, even for something like this. The problem is the "new"
part. It leads to too many unanticipated results. For example, your
suggested syntax doesn't mesh with the current ABNF, so either the
ABNF would need to change, or the syntax would need to change.
Maybe it is just me, but I've been burnt way too many times by last
minute changes right before software ships, and I think a spec is even
a worse place to be making last minute changes.
-wayne