Chris
I am sitting here wondering what is going on. To quote from your session
summary,
"It became apparent that the proposed protocol was not
going to be implemented in a widespread manner any time soon. The WG
agreed that widespread adoption will probably only occur if we stick with
the observed syslog protocol"
This seems to say that the WG is the meeting and that what happens on the list
is of lesser account. Your statement is probably correct but is so important
that I think you MUST confirm this on the list, asking people whether or not
they are going to implement the protocol as currently defined, inviting them to
reply offlist (if they want to) and then telling us the results of this.
Shutting down a WG on the basis of what persons unknown may or may not have said
of which there is no record does not instill faith in the IETF process.
The later minutes record as an issue
"Will it be implemented? -- Angle brackets w/PRI inside or new "
but give me no idea as to who if any gave what if any as answer to this
question.
(I continue to be surprised by IETF implementation reports that list just how
many organisations, often ones I have not knowingly heard of, have implemented a
particular protocol, which makes the work of the IETF worthwhile even when the
700lb gorillas are not taking part).
The minutes further confuse me when they say
Question to mailing list: should a new doc be generated to include angle
bracket,
new time stamp, and SD-ID
I see no such question to the list nor do I understand what might be asked. I
take angle bracket to be a reference to re-instating PRI in order to give
greater backward compatability but what is "new time stamp" and so what about
"SD-ID"? What are you on about?
Rainer helps when he flags internationalisation (something the meeting seems not
to have noticed), HOSTNAME and timestamp as the key issues but for me, places
too much emphasis on backward compatability. Great to have, especially with an
easy migration path to pastures new, but sometimes we do have to take a leap
that leaves some behind; the issue is, how many? Back to implementations.
Tom Petch
_______________________________________________
Syslog mailing list
Syslog@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/syslog
