ietf-822
[Top] [All Lists]

Re: Fwd: I-D ACTION:draft-klyne-msghdr-registry-02.txt

2002-02-05 08:00:25

But, at the end of the day,
what the people out there are looking for is something that gets their
headers into use, at least on a trial basis.

I've generally found that when someone produces a well-written 
proposal to do something that is obviously useful, and they've
taken the time to clearly address the potential negative 
consequences of their proposal, that it's not too difficult to
get that proposal published as some kind of RFC - either 
Experimental, Informational, or Proposed Standard can generally
be produced in a matter of weeks.  A working group is generally
not needed for these cases (of course, that's up to the ADs to
decide).

The process generally bogs down either because the proposal isn't 
complete or well-written, or when the proposal fails to address 
valid (and usually predictable) concerns.   Sometimes these are
inherently difficult to address, and sometimes addressing these 
concerns involves resolving a conflict between the valid interests 
of multiple constituencies.  But when the process bogs down
it is generally a good indication that the proposed protocol or
extension should *not* be widely deployed in its current form.

People do sometimes argue endlessly about minutiae, but that's 
usually because there are more fundamental issues at stake.
Once the fundamental conflicts are addressed, the minutiae can be
sorted out. 

For two specific headers that have been mentioned here: X-Priority 
and X-Face, I don't see any problem with publishing their definitions 
as at least Informational RFCs.  There are certainly other precedents for 
documenting protocol extensions in RFCs even when those extensions
violated the rules in effect at the time - e.g. unapproved use of TCP
port numbers has been documented, as has unapproved use of PPP 
extensions, and variant/incorrect use of ASN.1 tags.

But I can't help but wonder if there aren't other, more 
controversial proposals that people want to deploy without
community review, and that's why they're trying to get IETF
to say that we don't need review of mail protocol extensions
before deploying them.

Keith

<Prev in Thread] Current Thread [Next in Thread>