ietf-nntp Time

[Charles Lindsey]

In <20010321113907.A98701 at demon.net> "Clive D.W. Feather" <clive at demon.net> writes:


>I would like to propose a few extra words about time. I'm not sure where
>the best place to put them is; perhaps with the DATE command.

No, I think this is all too complicated. There is an existing sentence un
DATE which covers this issue. The most that is needed is a modest
extension to that. Then leave the gory details up to the implementors.

>    X.X Time

>    As described in section 9.1, each article has an arrival timestamp.
>    Each newsgroup also has a creation timestamp. These timestamps are
>    used by the NEWNEWS and NEWGROUP commands to construct their reponses.

>    The DATE command MUST return the same timestamp as is used for
>    determining article arrival and group creation times. This clock
>    SHOULD be monotonic, and adjustments SHOULD be made by running it
>    fast or slow compared to "real" time.

The first sentence could form a suitable "modest extension". But the
second is none of our business. It should not be part of this standard to
tell implementors of operating systems how to conduct themselves. (nd in
fact adjusting the time suddenly is actually safe in one of the
directions).

...

>    In order to allow for minor errors, clients MAY want to adjust the
>    timestamp back by two or three minutes before using it in NEWNEWS.

That is a sensible precaution, and covers all that is needed. If a
server gets its DATE response and its history file creation out of sync
by more than a minute, it is probably in such serious trouble that
everything it does is suspect. Remember that absolute correctness of the
time is irrelevant.

-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133   Web: http://www.cs.man.ac.uk/~chl
Email: chl at clw.cs.man.ac.uk      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5