I've long known that my 1998 Firefly's (Geo Metro) ECU has a few special lines of code in it for the alternator.
EG:
1) When the ECU detects no output from the alternator (when disconnected, or if it fails), the daytime running lights don't come on. Presumably this feature was added to help the car go further - to safety, or a repair shop - before the battery dies.
2) The alternator field isn't energized immediately after starting the engine. Presumably this is for easier starting, and possibly for a smoother (unloaded) idle before first driving away.
I discovered today though that this "feature" makes it possible to inadvertently kill the battery if you never cross the RPM threshold the ECU monitors where it switches on the alternator: 1900 RPM (at normal operating temp - not sure if it's a different RPM at other temps).
I haven't tested the other conditions, but I think that when the engine is below operating temp, there's also a "timer" in the code, that will switch the alt on if the driver just starts the engine to let it warm up.
So I was featherfooting a warm engine out of town this morning, shortshifting, and apparently I managed to get up to top gear and onto a secondary highway without going over 1900 RPM.
The ECU never told the alternator to fire up.
The thing that alerted me to the fact that something was "wrong" wasn't the voltage (didn't have it up on the ScanGauge gauges screen at the time), but I noticed the instantaneous MPG was quite a bit higher than I would have expected for the conditions.
A few clicks on the SG and I found battery voltage sitting at 11.7 as I was cruising down the road.
Just an observation / potential warning.
Likely a rare set of circumstances, but I learned today it's a possibility.