Re: [gpsd-dev] Clarifications needed for the time-service HOWTO

[Gary E. Miller]

Yo Eric!

On Thu, 24 Oct 2013 08:08:09 -0400
"Eric S. Raymond" <address@hidden> wrote:

> Miroslav Lichvar <address@hidden>:
> > I'm just suggesting to remove "(what the inventor of NTP whimsically
> > calls a "falseticker")" from that text. A falseticker is a source
> > that doesn't agree with other sources, large jitter doesn't imply
> > that.
> > 
> > The last sentence in the paragraph should possibly be removed too.
> > I don't think there is a fudge option to compensate for an asymmetry
> > in the network delay. Large symmetric delays are not a problem.
> 
> Done.

Reading the new text that seems a fair compromise.  OTOH, you say
'two reasons' then mention three.

I've been digging in RFC 1305 for the nitty gritty and while it mentions
delay, jitter and wander as error sources it explicity avoids descibing
the procedure for declaring a falsechimer.  If it scared the RFC editors
off we should also avoid it.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
        address@hidden  Tel:+1(541)382-8588

signature.asc
Description: PGP signature