It's common, especially amongst the Austrian types on this forum, to use inflation in it's original (monetary only) meaning, while others (typically the Keynesians and monetarists) use it in it's more modern price inflation meaning. If we're using the modern meanings of inflation and deflation, then Bitcoin is very strongly deflationary at the moment. If we use the original terms, then it's strongly inflationary, but demand is increasing even faster than supply.
It happens that any non-monetary "definiton" of inflation fails to be a definition and practically works as either a red herring or strawman in any discussion of economics. Should somebody finally be able to construct a non-monetary definition that actually works as such it may be taken into account (at some point, by someone). As it is, the keynesian pseudo-inflation is pretty much only interesting to politicos.
I agree completely. I count myself amongst those "Austrian types."
Eventually, Bitcoin will switch to being properly deflationary, but given that a) it will be a very gradual slope, or at the very least very long plateaus between deflationary events (lost or destroyed private keys) and b) lost and hoarded coins are functionally identical, and unless a private key is known to be lost, it's unwise to treat it as actually gone, I don't see it being a problem. When/if it does become a problem, Bitcoin2 is a simple code fork away.As was pointed out at the beginning of the thread, there is the base currency inflation/deflation, and there is the inflation/deflation due to financial instruments (the M3, etc.). As bitcoins mature, protocol defined inflation will be under 1% within a couple decades, maybe poster above is correct that 1% are lost every year, but I suspect that the total inflation/deflation will swing in larger amounts up and down based on the financial instruments and credit built upon the bitcoin system, totally dwarfing the M1 change, which will be rather small.