Testing ntpd and/or timing from gpsd

[Eric S. Raymond]

This discussion is going in a useful direction.  It needs to turn into
actionable advice and code.

I've opened up the wiki feature on the NTPsec repository. It's here:

https://gitlab.com/NTPsec/ntpsec/wikis/home

I've started a page on local performance monitoring:

https://gitlab.com/NTPsec/ntpsec/wikis/local-perfomance-monitoring

Because I anticipate that the content will eventually be merged into
the NTPsec main documentation, the format is asciidoc rather than the
default Markdown.

Dan, Hal, Clark Wierda: please consider yourselves the working group
on local performance monitoring (others may join, of course).  The
initial text of the page lists mission objectives.

Dan's remark that "If there is only one number reported, I'd say it
should be something along the lines of: "after ntpsec stabilized, 99%
of all local clock offsets were within +/- X us" (or 99.9% for a
tighter tolerance)" seems like the beginning of a good discussion
of figures of merit. A nice provocative querstion to continue
with is how we should weight the *dispersion* of clock offsets.

Discuss on the mailing list if you wish. Commit conclusions on the
wiki. Don't be afraid to commit a conclusion you're not yet certain
of, but when you do that indicate your confidence level.

Give me something I can use, gentlemen.  There are large decisions to
be made and not enough hard data to make them with confidence.  The
little machines on the Official Windowsill of Mad Science can give us the
raw numbers we need - now we need to figure out how to turn that into
information.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>