Wow, thanks for doing this.  Looks much better than our original (just using Google Docs to create the PDF).

We are frequently asked for details about Bitcoins and how they work.  So we have prepared this "white paper" that goes into a bit more depth than the typical introductory material available online.  While not designed to be a technical explanation, we hope that this will be of some use for people trying to decide if they should get more involved in the Bitcoin economy, focusing on the benefits and risk of using Bitcoins.

A Bitcoin Primer (PDF)

Here's an outline of the sections (it's about 8 pages long).


What is Bitcoin?
How does Bitcoin work?
What are the benefits of Bitcoin?
  Financial Self-Determinism and Control
  Irrevocable Transactions
  No Need for Middlemen
  Low Cost Transactions
  A World-wide System
  An Inflation Hedge for Long-term Savings
What are the Inherent Risks of Bitcoins?
  Irrevocable Transactions
  Underlying Value and Volatility in Prices
  Anti-Inflationary
  Computational Attack
  Regulatory Uncertainty
  Risk of Loss
  Is Bitcoin “The One”?
Applications Well-suited to Bitcoin
References and Links


Corrections or suggestions for improving this document as a general resource would be appreciated.

If you were to run mining as a business, you almost certainly would not have much in profits to report.

You should be able to deduct the cost of electricity, and depreciation of your hardware against your income.  I think most will find they are paying as much in electricity costs, as the value of the Bitcoin that they are receiving.

If you hold on to your Bitcoin, you should be able to claim all future gains in their value as a (hopefully, long-term) capital gain - which is taxed in the US at a much lower rate (15% max) than ordinary income (35% max).

I would propose that you could destroy some Bitcoin in exchange for coins in the new currency.  It easy to create a blackhole address such that you can be certain that no one will have the private key.

I was *close* to being fooled (bringing up the web site on a mobile phone, where my eyes have a hard time distinguishing the small text in the address bar).

I've not seen any way to resolve the problems with my original proposal; it seems too easy to use a large amount of cash to reach back arbitrarily far in the block chain to re-write history (and hence, invalidate transactions where you spent money).

I also don't like the proposals where participants have to wage a continuous battle to reinforce a disputed block with additional escrow balances.

Unless someone can come up with a brilliant way to salvage this, it seems currently to be unworkable (to me).

But I've enjoyed the discussion and the help poking holes in my proposal.

I would argue that there is a potential for Bitcoin mining to consume very large amounts of resources, on the orders of millions of dollars worth of electricity and computer capacity every day.  It's closer to $20,000 per day now.  People are fine "losing" that much, as they get at least that amount in Bitcoin in return.  But wouldn't it be better if we could have a system that had all the safety and security and decentralized control of Bitcoin, but WITHOUT the extra overhead?  I would think that would be a *good thing*.

BTW - after thinking about it, I would agree with other users here that a "lottery" per se may not be essential - I think it could be replaced with pro-rata return on investment.

I haven't addressed the bootstrapping problem.  You could start with a Genesis block, and then play nice until enough other people joined in and created enough of the currency to support the block commitments in a fairly uniform distributed way.  You could also generate "New Coin" in exchange for destroying a Bitcoin (i.e., enforce a 1:1 exchange rate - at least going one direction).

All the proofs of signing are there in the block for everyone to see and independently verify (signatures just use private keys for any Coin Address) - just like Bitcoin shows the proven small hash value and nonce.

Energy *is* being wasted if you consider that the same amount of *work* can be done for a lower *cost*.

Here's a modified proposal that addresses "problem 1" - how to allocate reward for those contributing to securing the block chain.

In most ways, this protocol is identical with Bitcoin's, except that it relies on "Proof-of-Stake" as opposed to "Proof-of-Work" (thanks, cunicula for the terminology).  Note that the utility of mining, is to certify a block chain, so that the entire community can agree on a single version of the transaction history.  Bitcoin's protocol is designed to make it expensive for any single bad agent to out-compete the rest of the network, in order to re-write history and invalidate transactions.

About $200 worth of computation resources (electricity, capital costs) are (currently) collectively consumed in creating each block.  So, someone who wants to "re-write history", need spend more than that to replace a block with their own (which is why you should wait for many confirmations before accepting a large transaction as valid in the block chain - it costs about $1,200/hour to go back in time a re-write history).

Given the volume of transactions currently in the Bitcoin network, this is quite a high "tax" we are paying; and it will presumably only be getting worse if the value of Bitcoin rises.

So, why not use the currency itself be staked in order to ordain a block as valid?

• Someone creates a proposed "next block".
• Anyone who wants to, can sign that block, by committing some of their Coins to it.  For example, I'll stake 100 Coins over the next 144 blocks (24 hours) - for total of 14,400 "Coin-Blocks" of commitment.
• When the block has collected enough signatures (the sum of the Coin-blocks committed), the block is considered valid, and a new block can be started.
• The required block bounty can be adjusted just like the current "difficulty" is adjusted in the Bitcoin protocol.
• In return for taking some of your Coins out of circulation, you earn a proportional amount of the block payout (e.g., 50 Coins).
• Each client deducts from the balance of any address used to sign a block, until the expiration block of the commitment (thereby taking it temporarily out of circulation).
• Clients break ties in potential block chains, by accepting the one with the largest total amount of bitcoin-blocks of commitment.

Sorry thus is not a fully fleshed out idea.  Maybe "proposal" is the wrong term.  I am hoping to find others in moving this forward (or prove that is not feasible).

Well, not "free" - but I'm looking for a system without the inefficiency of the Proof of Work system of Bitcoin.

Imagine something very similar to Bitcoin, except, instead of "mining", you risk some amount of your currency in order to receive some in return when you help secure the block chain.  But you risk losing your coins as well, so you won't have people gaming the system expecting to receive something for nothing.  There is no software yet, as this is just the beginnings of an idea - with many parts not solved yet.

I'll look for your posts.  I'd love to help build something along this line ... but there are still quite a few issues to iron out to convince ourselves that participants couldn't cheat (or it would cost them more to cheat than to play nice).

Worked for me.  I just made 0.1 BTC for posting this tweet.

Sweet!

Isn't the the subsidy is a useful incentive for managing/securing the block chain?

And yes, there's a big hole in the scheme until 2 is solved.  There has to be an unbiased way to decide between competing block chains (perhaps by preferring chains that have the most ticket value spent in them).  But we need the equivalent of the secure timestamping service that Bitcoin includes.

Though I love the Bitcoin protocol, and it's elegant solutions to double-spending and cheating, it still bothers me that it takes a prodigious amount of energy to mine for new currency (something like 25 Megawatts are being consumed by miners right now).  So I've been trying to think of ways to substitute a less costly process for the current hashing problems required by BitCoin.

If you think of mining as a form of lottery, each computation of a nonce hash is like buying one "ticket" - the more tickets you buy, the higher the probability of your winning the 50 BTC+ prize.  Why don't we replace mining with a more DIRECT lottery?

Every 10 minutes, say, each person that wants to participate buys how every many tickets they want (using the same currency), and then the winner is chosen randomly such that your odds of winning are proportional to your ticket purchases.  The winner not only receives the 50 coin bounty in the block, but also all the tickets purchased in the block.

This has the same incentives and rewards as Bitcoin, but reduces the net cost of "mining" to near zero (all the "costs" of mining are returned to the winner of the block).  It's "fair" since your chance of winning is proportional to the amount of coin you risk in each auction.  A simplification is to treat the amount of "tip" included in each transaction as the "ticket purchase" amount (you can enter a NULL transaction with a tip in any block when you want to mine, but not sending a real transaction).

The remaining problems to solve are:

• How to fairly decide the winner of the lottery (without relying on trusted 3rd party).
• How to decide that an accepted block is "canonical".

The first problem can be solved by hashing all the user addresses of the ticket purchases in the block, and using that as a seed to a cryptographically secure random number generator.

The second problem feels non-trivial to me and still a source of possible cheating.  Any ideas?

Has this all been discussed before???

Someone who releases a Lean Startup style Minimum Viable Product (MVP), may be indistinguishable from a scammer.  There is a risk that BFL may not deliver products as promised.  The difference between them and a scammer is that we would hope they offer refunds instead of disappearing with customer funds.  They have said as much here ... I hope we can trust them ...

I'm not seeing a problem logging in...

I'm liking the new Bitcoinica UI.  But, as someone who is not very experienced with Forex trading concepts, it seems to me the interface is unnecessarily complex.  Let me propose an alternative, simpler model; I'd love to hear from those more experienced in trading interfaces why I'm wrong.

Suppose Bitcoinica displayed only:

• Your $ account value.
• Your BTC account value.

Every time I "trade", one account loses some value, while the other gains.  Bitcoinica could lose the concept of an "open position".  A margin call could come into play if my total account value ($ + BTC) becomes less that a fixed percentage (say, 20%) of my LONG account balance.

An example:

• Suppose BTCUSD = $2.50
• I send 100 BTC to Bitcoinica ($ = 0, BTC = 100, Net Value = $250)
• I buy 100 BTC at $2.50 ($ = -$250, BTC = 200, Net Value = $500 - $250 = $250 => 50% of LONG)
• BTCUSD decreases to $2 ($ = -$250, BTC = 200, Net Value = $400 - $250 = $150 => 37% of LONG)
• BTCUSD decreases to $1.50 ($ = -$250, BTC = 200, Net Value = $300 - $250 = $50 => 16% of LONG)
• Bitcoica force liquidates 166.67 BTC @ $1.50 ($ = 0, BTC = 33.33, Net Value = $50)

I can grok what is happening much easier with this system.  There is no need for Open Positions separate from my Currency Accounts, and it's much easier to see if I am truly LONG or SHORT which currency (and by how much).

Thoughts?

Here's the problem with penny auctions, from a rational perspective.

All your past bids are sunk costs.  You can rationally increase your bid (as small increment) if you think you have a probability of winning the item.

If you have two "rational players" like this, then will continue bidding to infinity.  At each step, they realize that all their past bids are "lost", and then are willing to risk the small increment in order to win the (high value) item.

It's like a game of chicken.  The only "correct" strategy is never to start playing at all.

One wrinkle that some penny auction sites add is that you don't actually "lose" your bids, but rather you can apply them to buy other products.  The terms are probably not that great, but at least you can spend your sunk costs on something.

So, if you want to buy one of the consolation products, and don't mind sinking up to that amount in the off chance you can win a random lottery, that sounds fine.

I've not looked at your site to determine if that's what you're doing, or are just a "traditional" penny auction site.

I disagree.  I think it relies on many users not understanding what they are getting in to.  Since it's a winner-take-all model, and you can risk an arbitrary amount to win, most bidders are going to lose all their money.

I would never place a single bid on a penny auction site, nor would I advise anyone to do so (just like I don't buy lottery tickets - it's a losing proposition).