On 4/1/2010 11:05 AM, Fred Baker wrote:
So - does RFC 5841 update RFC 3514, or obsolete it?
Probably not. RFC 3514 is actually a protocol. RFC 5841 is not.
A protocol needs to specify deterministic behavior by participants at both ends
of the exchange. Otherwise there cannot be interoperability.
The new entry defines publishing labels. That is, its focus is on conveying
mood rather than worrying about the use of mood. At the least, it might have
suggested that a bored packet label should be subject to a flood of packets in
response, just to make the issuer's day more interesting?
This perhaps sounds like sour grapes to the fun of the April 1 series. Let me
assure you that it is.
Some years back, I submitted an IP-over-email specification and Postel rejected
it. He noted that the email service would presumably be operating over IP and
that having IP at two levels created a potential addressing anomaly that the
specification needed to resolve. (Later IP-over-IP tunneling work probably
proved him wrong, but he was certainly right that it was an issue, absent
empirical data.)
I protested that for goodness' sake this was an April 1 specification. He
responded that specifications need to work...
FWIW, the good news is that the labeling is based on DSM-IV, but the bad news is
that DSM-IV is designed to aid in completing insurance forms, rather than really
being tailored for treatment-related guidance in dealing with humans or packets...
Given the insurance orientation, perhaps the real purpose of the labels is for
bilateral SLAs?
d/
--
Dave Crocker
Brandenburg InternetWorking
bbiw.net
_______________________________________________
Ietf mailing list
Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf