Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!uwmcsd1!ig!agate!ucbvax!UDEL.EDU!Mills
From: Mills@UDEL.EDU
Newsgroups: comp.protocols.tcp-ip
Subject: Re:  Network Time
Message-ID: <8807061117.aa19795@Huey.UDEL.EDU>
Date: 6 Jul 88 15:17:00 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 19

Miles,

Your network is hardly unique. MCI currently synchronizes about a dozen
plesiochronous "islands" in their digital telephone network using LORAN-C,
Sprint plans to do the same and AT&T plans eventually to use the Global
Positioning Satellite (GPS). LORAN-C provides timing to within some tens
of nanoseconds (for things like digital telephone synchronization) and
in principle can provide UTC verification to wihtin a couple of
milliiseconds. It does not provide a UTC timecode, which must be
disambiguated by other means. There are no LORAN-C receivers known to
me that are designed expressly for time service, as against position
service, although some may be capable of rho-rho navigation, which
requires a much more precise local clock than the usual hyperbolic
navigation.

Most folk are expecting GPS to become the system of choice. I do not
know if its designers are sensitive to the time-distribution issue.

Dave