ietf
[Top] [All Lists]

Re: ITC copped out on UTC again

2012-01-22 20:27:44

"Eliot" == Eliot Lear <lear(_at_)cisco(_dot_)com> writes:
    >> Can you tell me which protocols use future timestamps in an
    >> moving form (not stored at rest in a certificate in a DANE RR,
    >> for instance), which care about discrepancies of less than 1
    >> minute?

    Eliot> iCal, for one, which can be used for recurring events that
    Eliot> have nothing to do with computers.  Also relevant, would be

Forgive me for being dense, but I don't understand how this is relevant.

iCal, as far as I can understand, stores start/end dates in human form.
For instance, from RF2445:
     BEGIN:VTIMEZONE
     TZID:US-Eastern
     LAST-MODIFIED:19870101T000000Z
     BEGIN:STANDARD
     DTSTART:19971026T020000
     RDATE:19971026T020000
     TZOFFSETFROM:-0400
     TZOFFSETTO:-0500
     TZNAME:EST
     END:STANDARD
     BEGIN:DAYLIGHT
     DTSTART:19971026T020000

I don't see seconds-since-epoch anywhere here, I see years since
guy-was-nailed-to-cross here.  As such, that number is assumed to
already be adjusted for leap-seconds, so it's a non-issue.

What I'm looking for is a protocol where a hostA says to hostB,
  "lets meet behind the red-router for a 4.2s tryst 27.563456years from
   now"  

it's important that the period of time until the meeting be large, and
the precision of the meeting be small enough so that the uncertainty due
to leap-seconds exceed the precision of the meeting time.

We do not know how many/when leap seconds will be inserted in the next
hundred years, but based upon one every 2-3 years, it would be around
one hundred years before the accumulated uncertainty due to whether it
was one every 2 years or one every 3 years, begins to exceed 1 minute.

I've heard statements about financial systems, and I've obtained the FIX
specifications from http://fixprotocol.org. (The copyright permits
redistribution, but fixprotocol.org makes you register to get the .zip
file... )  and as far as I can see in my naivety, few timestamps in this
protocol are likely to exceed 1 year, and none of them would seem to
need a precision of less than 1 minute.

-- 
]       He who is tired of Weird Al is tired of life!           |  firewalls  [
]   Michael Richardson, Sandelman Software Works, Ottawa, ON    |net architect[
] mcr(_at_)sandelman(_dot_)ottawa(_dot_)on(_dot_)ca 
http://www.sandelman.ottawa.on.ca/ |device driver[
   Kyoto Plus: watch the video <http://www.youtube.com/watch?v=kzx1ycLXQSE>
                       then sign the petition. 





Attachment: pgprpPg2uZW0P.pgp
Description: PGP signature

_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf
<Prev in Thread] Current Thread [Next in Thread>