[Cryptography] BitCoin bug reported
patrick at rayservers.net
Sun Feb 16 11:18:45 EST 2014
Ben Laurie wrote, On 02/16/2014 06:08 AM:
>> You wrote that "the upper limit on the value of a bitcoin is set by the
>> cost of electricity to mine it." That's just not true. The value of a
>> bitcoin can far exceed the cost of the electricity needed to mine it.
> Wat? If I'm buying a bitcoin, why would I ever pay more than it would
> cost me to mine one?
You have a point. I was forgetting that just about anyone can mine a
bitcoin. It's not like mining a pound of copper, which requires a very
specific set of skills, property, and capital equipment, not widely
available to everyone. When people pay for a pound of copper, they are
paying far *less* than it would cost them to mine it themselves.
I'm not completely off the mark though. There may be plenty of people
who would gladly pay $600 for a bitcoin, when the cost of electricity to
mine it themselves might only be $300. That's because (1) they don't
know *how* to mine a bitcoin, or (2) they can't be bothered. Because of
specialization, they are more adept at earning dollars than mining
bitcoins -- but they want bitcoins.
>> You wrote that "the value increases as the difficulty of mining
>> increases." That's just not true. The value of a bitcoin can fall as
>> the difficulty of mining increases.
>> The cost to produce something and the value of the thing produced are
>> two entirely separate quantities, and not necessarily correlated.
> This is almost content-free, but what content it has seems wrong.
Well, the general case is obvious. I could spend a hundred hours
welding together an elaborate assembly of random metal parts, at great
cost to myself, and the value of the resulting thing would likely be no
greater than scrap value. Similarly, I could spend ten hours of hard
labor breaking out all the sheet rock in my house, and the resulting
mess would have negative value.
So I hereby reaffirm my second point. The difficulty of mining a
bitcoin might rise to infinity, while the value of that bitcoin falls to
> The upper value of a bitcoin is bounded by the cost to produce it. The
> cost of producing a bitcoin is the current difficulty * the cost to
> run the latest rig * the number of people-running-latest-rig
> equivalents who are currently mining. Two of these quantities can get
> (almost) arbitrarily small. So, the value of a bitcoin can get
> arbitrarily small - but not because it is an entirely separate
Yes, that's one way for the value of a bitcoin to get arbitrarily small
-- for the cost of running rigs to drop, and the number of people
running those rigs to rise. The presumption has been that the
difficulty of mining would rise to outpace those other factors, so the
value would not plummet. After all, we *know* that the difficulty will
eventually reach infinity, when the fabled 21 million mark is finally
But there's another way for the value of a bitcoin to get arbitrarily
small, and that's the "Emperor has no Clothes" effect -- when people
start saying "Wait, it's just an effing useless hash, and I can't fill
my teeth with it, make jewelry out of it, or make non-corrosive undersea
electrical connectors with it!" That is when Carl Menger's Subjective
Theory of Value kicks in with a vengeance. That day may never come,
because people currently value bitcoin just *because* it is something
which allows abstract numeric units to slide from one owner to another,
with no central clearing platform, and that inherent property itself is
enough, not the ability to do anything else with it.
More information about the cryptography