There's also Primecoin, which proves the concept that mining energy can be put to a useful goal (for some value of "useful").

https://en.wikipedia.org/wiki/Primecoin

Wait... *what* uninformed developer(s) predicted that? The previous record (six months ago) was set by Titan, at 17.59 Petaflop/s. So to pass the exaflop barrier this time around would require over a fifty-fold improvement -- something never before seen in the history of the Top500 list. Did someone *really* make this prediction, or is author Kevin Fogarty just making shit up?

<quote>The mapping from your pseudonyms to your "real" identity can be done as soon as you trade your bitcoins for anything not bitcoin.</quote>

That's a good point, assuming that government regulations will require exchanges to record traders' identities (which seems likely). On the other hand, one could use the money to buy something with the Bitcoin and remain anonymous.

Tangentially: I used to think that Bitcoin couldn't be anonymous because one could build an identifying profile based on a series of purchases or exchanges. Then I learned that quite a few holders of Bitcoin use a *different address for each transaction*. That really does muddy the waters!

I think you're underestimating the coordination and effort required in that "administration and such". That's the central authority.

Decentralized authority works for some things -- the Bitcoin system demonstrates that. But some (such as judging insurance payouts) require the human touch.

We disagree on what "identity" means. A cryptographic token that stands for "Tom Geller" isn't the same as the meatspace Tom Geller. Heck, even the name "Tom Geller" isn't the same -- it's just a token as well.

The meatspace Tom Geller can be arrested and held. He has to personally show up at the bank to verify his identity when he opens an account. He has a picture on his driver's license that more or less matches his face.

<quote>bitcoins are not anonymous.</quote>

That's true in a technical sense. But practically speaking, they can be mostly so.

<quote>No. Pseudonyms with private keys can achieve the same thing.</quote>

We disagree. I think you're underestimating the reasonable requirements of governance.

<quote>This is done in bitcoin today without any centralized authority.</quote>

No, it's not. The spender is never identified in a real-world way.

Oh, duh -- I guess your units are from Butterfly Labs. (Sorry, was reading comments in non-threaded mode and got confused.)

Thanks for the details! Always welcome.

What makes are the units? This sounds quite out of line with what others have reported.

(Consider that Blockchain.info reports a net *loss* for mining -- see https://blockchain.info/stats . Maybe that includes only transaction fees or something?)

Good lord, why are the two of you spending so much time and effort on this? Go play.

Hi, AC -- thanks for your comment. I understand why you think there's a contradiction, but I don't believe there is.

I think that future Coin will be based on essentially the same specs, procedures, and algorithms Satoshi described -- in that sense, it's decentralized. However, I think there'll be a layer above that which *isn't* decentralized. And there will be a tie between the two layers.

Here's an example. Let's say that Bitcoin continues to be the leading cryptocurrency -- anonymous and decentralized. However, if you want to insure deposits in it, you have to submit to some kind of identity proof through a central authority. People who need that service will submit, thereby taking part in a system with elements of both decentralization (the Coin) and centralization (the services supporting it).

It could also be that the centralization has deeper roots, making it essentially impossible to work in Coin without taking part in the layer of centralization. For example, what if merchants could only accept coin if it was authenticated to its identified spender? Then everybody would use the "top" layer, or be unable to spend their Coin.

Thanks for all the details! Great summary.

So how do your numbers jibe with BFL's homepage claims of 5GH/s?

I doubt this. Could you explain more? What (currently available) computer could do this? What part would the botnet play? How would it be coordinated?

We are of similar minds. :)

I'd moderate this up if I could. :)

Your points are why I think a government-like body will get involved -- and I do believe that there will be an increase is costs when that happens.

Will those costs be greater than for current systems? I don't know. I'm guessing they won't, if only because of Coin's inherent self-defense mechanisms (crypto). On the other hand, most attacks are against "wetware" -- i.e., tricking the people who hold value. Those attacks will need just as much protection -- at just as great a cost -- as for any other medium of value.

Anyway. I don't see much difference between Coin and existing electronic transactions in the situations you describe.

<blockquote>Oh, and its easy to lose your bitcoins. Gee, just like "paper" money.</blockquote>

It's a matter of degree. I think it's an order of magnitude (i.e., 10x) easier to lose Coin than paper currency.