> The key enabler of these properties is a new TrueTime > API and its implementation. The API directly exposes > clock uncertainty, and the guarantees on Spanner’s timestamps > depend on the bounds that the implementation provides. > If the uncertainty is large, Spanner slows down to > wait out that uncertainty.
...so really, GPS and atomic clocks aren't for defining "now" they're for providing a keeping the window of uncertainty around "now" to a small value. The value mentioned in the paper for this is "generally less than 10ms", which is pretty good for a large-scale distributed system, but massive in terms of precision timekeeping. It is, after all, an uncertainty that's three orders of magnitude larger than the GPS glitch that started this discussion...
Of course, they make use of GPS, but also add their own atomic clocks.