This side of it, I am aware of. 
I see that doubts have been cast on the thesis that touching a cold bulb will reduce its life. I’ll be very interested to see the results of any subsequent experiment.
The pressure may be less than atmospheric, but it’s definately not a vacuum. If you put the bulb into a vacuum chamber it will explode ;). Having some gas in there helps prevent evaporation of the filament.
I read once that the filament is initially coated with some substance that is supposed to react with any residual oxygen the first time it is turned on. I’m having trouble varifying that at the moment.
Correct sir. Bulbs are filled with argon, nitrogen, or a similar fill. At what pressure I don’t know. Personally I think part of the reason they go “pop” when they break is that’s what thin glass bulbs do.
It is worth noting that the filament itself is made of material that will react with any residual oxygen the first time it is turned on. Can’t see why they’d need an extra coating, of course that doesn’t mean it’s impossible.
My guess is that a cheaper coating burning up means a longer filament lifetime. They could also have less stringent purity restrictions on the filler gas if they can just use up the oxygen with the coating. This is all I’ve found:
This site claims krypton to be the elemental gas. The thick plottens. 