As such, to an expert in any one of those fields, it's very hard to digest because looking at it from the perspective of one field might suggest it is doomed for failure. But experts tend to look at things from the perspective of their own expertise.

For laymen who might know a little bit about each of those domains, it feels like "Yeah.. this could work..". A simple analogy might be about how hardcore web developers always feel about PHP. They always say it's a pile of garbage and it's bound to self destruct and swallow the earth in the process. However, PHP trudges on with some very large projects and sites under it's belt.. and novice developers will almost never notice a problem with it. Experts can be wrong.

So yes, bitcoin has it's issues if you analyze each aspect of it. But maybe if you zoom out and take an overview it might start to make sense how all the parts get together and even with flaws with each sub-aspect it just might work.

Wait, I think we are missing his point. Last sentence of the article:

"So let’s talk both about whether BitCoin is a bubble and whether it’s a good thing — in part to make sure that we don’t confuse these questions with each other."

I think this article does not go into the 'will it work?' debate. It alerts us instead to another debate, 'to what degree do we want it?'

It is an addition to what Charlie Stross was saying. He is just adding "guys, why are your comments replying to a different question?"

At least, I think that is the discussion he wants to start. But his talk about the ceiling and floor in value in the middle of the article actually belongs to the OTHER discussion, the one he does not want to start but spends half of the text going into.

It is like trying to talk about empirics to consider for deciding national policy, while everyone is shouting their usual political opinions.

So either you're misunderstanding Krugman's argument or I'm misunderstanding yours. I think you're confusing his normative and positive arguments (which are entirely separate!). You have used an argument against his normative argument ("Bitcoin is evil because it's wasteful.") but then used it to challenge his positive argument ("Bitcoin won't work out because it has no value floor."). That a crypto currency might end up helping society does not make it a good value store. Why would it?

For the time being, let's ignore those and concentrate on the primary purpose of Bitcoin's infrastructure, to facilitate decentralized value transactions between arbitrary parties. It is the infrastructure's fitness for this purpose that I see as giving it its floor; this may seem a slightly circular argument, but by enabling what it does it will always have the value of accomplishing that goal, even if it's trading for vast fractions of a cent (like Dogecoin).

Adding the network effects of its installed base, this probably places its floor somewhere well above Dogecoin's current value, but certainly far below what it's currently trading at. And indeed, it could very well undergo further corrections until it reaches a level of stability appropriate for a value store.

> I have had and am continuing to have a dialogue with smart technologists who are very high on BitCoin — but when I try to get them to explain to me why BitCoin is a reliable store of value, they always seem to come back with explanations about how it’s a terrific medium of exchange. Even if I buy this (which I don’t, entirely), it doesn’t solve my problem. And I haven’t been able to get my correspondents to recognize that these are different questions.

In other words, "to facilitate decentralized value transactions between arbitrary parties" does not actually provide a floor and thus give the coin intrinsic [1] value. Being useful is different from being valuable, at least insomuch as storage is concerned. I think Krugman has this one right.

One point made elsewhere in this discussion is that while Krugman might be right about the floor, he may be wrong about the floor being necessary to make BTC a stable, suitable store of value. Specifically, he might be overestimating the degree to which the dollar has a meaningful price floor, and the degree to which such a floor is actually the reason the dollar is stable and dependable. In other words, perhaps it maintains its value through adequately shared fiction, and perhaps BTC could do that too. I don't really know enough to have a strong opinion on that issue, but I think that's the structure of it. It's also worth pointing out the Krugman has the conventional view here.

Another question to ask of Krugman is why being a store of value is even a goal; what does the "success" in "successful currency" mean? Does it mean that BTC will be the unit of account, with prices and salaries denominated in it? That's certainly the case Krugman is taking on (hence the normative claims about central banks), and the case many here on HN arguing for. Or can it succeed as just a medium of exchange? I can trade my dollars for bitcoins and send them to someone, who will convert them to euros on receipt. Then who cares how much they're worth?

[1] For certain values of "intrinsic". The semantics aren't very clean here.

Being a store of value is a goal if you want people to hold a currency. For example, when I was an exchange student in Ecuador long ago, the Ecuadorian sucre was not seen as a good store of value. Better-off people would, as much as possible, not hold sucres; they'd buy dollars. Poorer or less connected people couldn't do that as easily, so they just got screwed. Eventually, the currency collapsed entirely and now they just use the dollar: http://en.wikipedia.org/wiki/Ecuadorian_sucre

I don't think BitCoin can survive purely as a medium of exchange. There's the obvious reason: why pay two sets of transaction costs when you can pay just one? But I think the bigger problem is that there's a period when someone is holding BitCoins. If those are a stable store of value, then you're ok with that. But if not, you're in the land of currency risk, and the only people who like currency risk are currency traders; everybody else hates it.

But to play devil's advocate because it's fun: to the degree to which it is useful to hold onto the currency you also use as a medium of exchange, then perhaps that does provide a value floor. In other words--assuming it is successful as a means of exchange--the fact that you have to pay a transaction cost to get out of a BTC position means you'd rather hold onto BTCs so you can use them later at no penalty. So then they would have intrinsic value, based on your aversion to trading them for something else; anything that makes you want to hold BTC other than their market price counts. The question there because which is the chicken and which the egg: being a successful means of exchange or being a good store of value? And of course, whether that transaction fee provides a meaningful enough floor.

I'm not convinced of that either, and it's especially weird to have the thing enforcing the floor to be falling, but it's fun to think about.

I think that experiment has been done with paper currencies where people stop trusting the backing. It would still have the same cost-to-get-out value (indeed, I'd think that value would go up as the currency fails). I know the Ecuadorian sucre went from something like 500 to the dollar to 25,000 at the end, so it would seem that the cost-to-get-out value is small even with traditional currencies; given BTC's theoretically low transaction costs, I presume it would be even smaller.

You're absolutely correct. People forget that bitcoin functioned just fine as a black market currency back before it became a rite of passage to opine on its future. Barring prohibition by governments, the cryptocoin ecosystem is here to stay and will likely continue to grow.

It seems to me that the shared fiction of the dollar is based on the shared fiction of America. The fiction of America creates a real American government that has a strong interest in maintaining the fiction of the dollar.

From what I can tell, Bitcoin does have a similar pair of shared fictions, the idea and the currency. But I'm not seeing the thing that America has between the two fictions: a powerful organization with a strong interest in keeping the money useful as money.

If the infrastructure goes on to do great things for society then it has (and has always had) intrinsic value and thus the price of BTC rebounds (or more likely never collapses in the first place).

Of course, you can say the infrastructure makes the coin more useful as a medium of exchange, but that's a totally separate point, as Krugman points out.

My personal view is that none of these things--gold, dollars, or bitcoin--has enough intrinsic value to make a useful value floor, and that as value stores they all just rely on a shared convention. So I'm not sure Krugman is right, and perhaps bitcoins will work out without that floor feature, and perhaps without ever really being used as a store of value. But this argument about infrastructure just doesn't work.

It is both the mining equipment (at any point in time) and the blockchain which make possible the use of BTC as both a storage of value and a medium of exchange. As long as there exists both a globally-distributed secure ledger and a network capable of maintaining it, then the individual accounts within its purview obviously have whatever values it records them as having.

I think there's a basic disconnect here about what we mean by "store of value". The simplest way to put it is that something stores value well if I can buy a predictable amount of stuff with it in the future. That the ledger does a good job of establishing that I have N bitcoins is different from telling me how much those bitcoins are actually worth.

Of course, we can easily imagine scenarios where Bitcoin loses all value but I think a sufficient imagination can produce scenarios for the other stores of value listed on that page as well. So, this is obviously a bit subjective territory (i.e., different people can assign different probabilities to each of the circumstantial propositions) but there is a strong argument from consistency for the designation of Bitcoin as a 'store of value'.

I personally think that mathematically-interesting (and rare) numbers do have an intrinsic value (even if only as a novelty or perhaps antique/collectible).

But I don't see where you've established that Bitcoin has such a floor, as I said above. I don't think we're getting anywhere, though, so I'm going to break off.

I understand that you do not think that those other entities have floors either (and I, having a sufficient imagination, tend to agree with that). However, in that case, the term 'store of value' is non-sensical since, given a sufficient imagination, no entities can ever satisfy the 'floor' requirement.

My understanding could be wrong, as my understanding of Bitcoin would definitely be at the "noob" level :)

Brinks' entire CIT vehicle business is based upon just a subset of this function, and we can agree their value is not zero.

So as I see it Bitcoin is already obsolete; there are new coins that achieve the transaction capabilities for cheaper - they can be used in more potential applications, thus they're more interesting speculative instruments.

The specific benefit of proof of work mining comes in insuring more people have a chance to enter during initial distribution, but it still massively favors the early entrants and especially the people who have the hardware environment to mine efficiently.

So gold was used extensively 2000-3000 years ago. Exactly what was its intrinsic utility then? Plating astronaut helmets and high-conductivity, corrosion-resistant electronics?

The intrinsic utility of gold was that it was eminently verifable. It's got a characteristic ductility and it was a yellow metal. You could easily verify if it was 100%, 80% etc. In other words, to the ancients, the intrinsic utility of gold was "its trust model". Sound familiar?

And then, promptly, they also repeatedly ran their economies into the ground by conquering and mining tons more gold, and inflating their currency into worthlessness.

And this happened over and over - the Spanish did it, the Chinese did it, the French did it. Every gold economy before the invention of actual economic thought destroyed itself with gold.

While I don't think this analogy works without flaw, I do think that it draws a powerful connection between something that is widely regarded as a store of wealth, and something that is vying to compete in the same space.

Well, usually, the intrinsic utility of gold and silver was that they could be used as jewelry and/or used as instruments of worship. Most bullion money has evolved from substances considered divine.

Remember the Golden Calf?

You mightn't like my tastes but I don't like your gold necklace either.

Glass is not easy to work into jewelry on its own (or at least wasn't 2000-3000 years ago). Plastic-coated aluminium didn't exist 2000-3000 years ago.

What I think you're saying is that the block chain and it's supporting infrastructure and the functions it enables ( decentralized notary, unrepudiable statements, etc. ) are creating value in the same way that making jewelry from gold creates value. And that's an interesting argument. Much more interesting than a bunch of people ranting incoherently about "fiat".

If I were in search of a dissertation topic as an economist/anthropologist; I would be looking at the current wave of cryptocurrencies as a fascinating natural experiment in the social construction of value.

If the gold price collapses, you can sell it to jewelry and electronics manufacturers at some (low) price point. Gold has a weakness as a store of value because this price point is a lot lower than its market price during good times - and this is why Krugman is also not in favor of it as a store of value.

If the price of the dollar collapses, you can sell it to the Fed (usually indirectly - the Fed sells debt and then sits on the proceeds). This floor is pretty solid, cf. Volcker's recession in the early 1980s.

If the price of Bitcoin collapses, you can't sell the electricity used to produce it onto the grid - that electricity is a sunk cost. So what provides a floor to the Bitcoin price if for whatever reason its value suddenly falls? Probably only its utility as a medium of exchange, which in an era of electronic exchanges lasting fractions of a second probably doesn't provide a huge demand for the stuff.

I think this is a dubious argument. It describes what happens most of the time, when the value of the dollar fluctuates. But what if the collapse of the dollar is accompanied by a downfall of the Fed? Hyperinflation has happened in many, many countries that used fiat currencies, and even the US has had high inflation:

http://en.wikipedia.org/wiki/Hyperinflation#Examples_of_hype...

Gold originally was the only metal colored like that (shiny) so it stood out, coupled with its scarcity it became a status symbol. It was the stuff of royality. It has some properties that make it useful in jewelry but pure gold is actually too soft for jewelry and is used mostly as alloys.

I am not saying gold (like other metals) doesn’t have its industrial applications, it does!

You can not convince me however that gold is inherently a better metal for jewelry than alternatives. It is only there to give jewelry that golden shine, which has been considered a status symbol. Now, if gold somehow became a "cheap" metal I don't think people would be keen on displays of their gold.

I don't see how it is not the case that mining Bitcoins is throwing away energy.

If someone had made a crypto currency based on, say, protein folding [1], I could get behind that.

We could just create a crypto currency where we just hand out 100 units to every person. That certainly wouldn't require ASIC farms and such.

[1] Yes, I realize that the whole 'hard to do, easy to verify' bit is not completely obvious in the case of protein folding or SETI signals.

Not without a trusted central party you couldn't. The work that goes into bitcoin mining is the 'distributed engine' that prevents double spending or forging of bitcoins. Bitcoin mining is not just a way of handing out new coins, its the very engine of the whole system. The only known way to break that engine is to control a bigger engine of the same spec (the >50% attack), and thats looking increasingly unachievable by anyone.

Bitcoin couldn't and wouldn't exist at all without that promise. So a simple system based on allocating 100 coins to everyone just wouldn't do at all.

The 'work' is actually the verification of transactions on the blockchain - the reward from mining is just an incentive to do this (mining & verification are tied up in the same process).

The energy expended in mining is necessary to validate transactions in the network.

You might be able to make it more energy efficient but you can't remove the energy expenditure.

Ideally, the BOINC processing would somehow be directly integrating into the mining itself. But again, I don't have any idea how that could actually be done. But it would be nice if that is possible.

Mining is not that valuable, there are other ways to do it that don't waste so much energy.

> It's not simply throwing energy away into solving useless math problems

Yes it is, since there are solutions like proof of stake that solve the same problem using vastly less energy.

The complaint Mr Krugman is registering is that Bitcoin's don't even have that. If my car is totaled, I can always sell it for $100 for scrap. There is no bit scrap dealer.

Bitmessage doesn't use the same blockchain as bitcoin. The work of bitcoin miners does nothing at all to help bitmessage.

Only inherited protocol design or reused code can be said to contribute, and that's not relevant to this discussion.

But it is also an arms race of energy spent. And the more valuable BTC becomes, the more energy a rational agent will use to mine it. Unlike mining gold, BTC mining does not have a fixed cost.