wayne <wayne(_at_)midwestcs(_dot_)com> wrote:
The one major objection I had to what was said in the LMAP document is
all the vague references to "changing semantics". I don't see how any
of the LMAP proposals seriously change any semantics,
They don't. The text in the original discussion paper said that no
semantics were changed. I will update the next version of the
document to explain why. Have two parties: originator "O" and
recipient "R". They each can either use LMAP "+", or not "-". Let's
work through these combinations:
O- R- : no one uses LMAP: no change to anything
O+ R- : originator publishes data no one looks at: no change to anything
O- R+ : Recipient looks for LMAP data, which doesn't exist.
The recipient SHOULD inter-operate with non-implementors:
no change to anything. If the recipient doesn't
interoperate, that's his choice.
O+ R+ : both agree to use the new "changed" semantics: no problem.
The ONLY problematic situation is the third one. The LMAP
discussion document makes this clear, and describes in detail the
choices available, and their costs and benefits.
and this phrase seems to have been latched onto and blown out of
proportion by various people. (Mostly people I don't recognize
being involved discussions on SPAM-L, NANAE, ASRG, SPF-discuss,
etc.)
Who don't want to change anything, or who are unable to work through
a simple 4-step analysis of the effects of the proposal.
Alan DeKok.