I'm an electronics engineer, and might be able to add some insight here. I obviously don't know everything about how the battery and charging circitry is set up on the Nook, but I've designed several Lithium Ion based systems, and can make some general comments.
On the Nook, as someone has already stated, the eink screen only uses power when there is a refresh or a page turn. That's why it actaully displays something when it's sleeping. I'm also assuming that the microprocessor inside is sleeping most of the time, and only wakes up when you hit the button.
Comparatively, the Wifi, 3G, LCD, and touch screen capabilities, use HUGE amounts of power compared to the microprocessor and the eink screen. Even notive your nook get warm when you use Wifi or 3G for an extemnded period. There's a reason for that, it's using lots and lots of power. If you're just reading and turning the page, you arne't using much power at all.
All this is why the estimations for battery life are so weird to some people. If you keep turning on the color LCD for example, you're drawing an awful lot of power, at least compared to the power required to turn the page every so often. If you accidently activate the LCD while running a battery life test, you've just added a huge amount of inconsistency to your test, and you might as well start over.
Lithium Ion cells lose overall capacity based on time, but mostly charge cycles. Most cells are rated for 500 full charge cycles, after which the capacity of the cell with be aound 80% of a new one. This is why if you keep a cell phone for years, and charge it every day, you'll eventually need to replace the battery.
Lithium Ion charge and discharge curves are tricky. At the high end (fully charged) the voltage level of the cell is at its highest. When the cell is nearly depleted, the voltage level is lower. Lithium Ions have a discahrge curve that starts high, drops to the middle and stays nearly flat in voltage for most of the discharge (say 70% of the capacity), then drops sharply at the bottom end. Some might notice that the battery levels seem to drop much faster when it is at a low level. This is just a simple voltage-to-current effect. The higher the voltage of the battery (at a high charge level), the less current required. At a low charge level, the voltage is lower, thus more current is required to maintain the same amount of power usage, and therefore, the level drops more quickly.
The level indicator, mostly likely, relies on what is commonly called a 'fuel gauge' or a 'Coulomb counter'. It's sole job in life is to watch how much current is going into or out of the lithium cell. Some fuel gauges do a beter job than others, the Nook's seems to do fairly well really. The think about fuel gagues is that they must 'learn' about the cell.
Remember earlier that I said the capacity of the cell changes over its lifespan? Well, fuel gagues must account for this, and other factors that effect the performance of the lithium cells. For a fuel gauge to 'learn', it needs to watch a full charge and discharge cycle for the battery. This is why some recommend that from time to time to discharge the battery completely, and charge completely again. This will retrain the fuel gauge, and give the higher level software on a the Nook a better idea of what its battery is doing.
Software, and maybe some hardware, automatically turns off the nook when the lithium cells get to a certain low level. It's true that if a lithium cell is discharged too low, that is cannot be recharged again. However, unless the Nook has a HUGE design flaw, this cant' actaully happen with normal usage, even if you run it down until it turns off, and keep trying to turn it on again. The lithium cells will have protection hardware that will protect them from both under voltage (over discharge, which destroyes the cells ability to recharge), and over voltage (over charging, which degrades the cells ability to hold a charge, and will eventually lead to a fire if the voltage is too high). You probably don't need to worry about that last fire bit, but I've seen it before... I have a good video...anway...
Someone mentioned that its possible that the Nook jsut won't work if you discharge too low. As I said before, unless someone messed up when they designed the protection circuitry, this isna't really a possibility.
)
