Well, that's how Spanner's consistency works. CockroachDB uses a hybrid vclock/wall clock solution that's possibly going to be tuneable to just clocks. https://jepsen.io/analyses/cockroachdb-beta-20160829

Every GPS chip in every GPS-enabled device includes the clock — by design, it can't work without it. So perhaps it is possible to quickly hack something from a cheap Android phone placed somewhere near the window?

Using GPS as your time source is a fabulously good way to trash your entire database during a leap second. See this paper for a discussion on how essentially everybody but Google gets this wrong: http://crin.eng.uts.edu.au/~darryl/Publications/LeapSecond_c...

GPS time does not have leap seconds: http://www.oc.nps.edu/oc2902w/gps/timsys.html

> GPS Time is a uniformly counting time scale ... The word "uniformly" is used above to indicate that there are no "leap seconds" in this time system.

> The GPS message contains information that allows a receiver to convert GPS Time into Universal Time

In fact the article you linked corroborates this, because they used GPS time to show how inaccurate NTP servers are.

That was sams idea I had. I found that SparkFun even had a guide on it:

https://www.sparkfun.com/pages/GPS_Guide

Better resources: http://www.satsignal.eu/ntp/Raspberry-Pi-NTP.html

https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts (I don't understand a lot discussed there, but many interesting links and info about current deals to be found)

There also are differences between the quality of the timing output of different GPS modules, with some being optimized for timing applications (as opposed to location finding)

Appreciate it as I am wanting timing-focused ones.