Strictly speaking, it is not the hash, but just one of many possible hashes. It's always possible to come up with a hashing function to make a specific trivial password look complex from the standpoint of Shannon -- and, in the absence of information about what that hashing function actually was, there's a good argument for saying that it is complex. After all, the suggested string is also a hash of the word 'easy', and it is a hash of the word 'trivial', and it is a hash of the word 'oops'. However, if I don't tell you what the hash function actually is, it is unlikely that you would actually discover it.

The problem comes not from choosing a word like 'password' to run through a hashing function; it comes from choosing a well-known function with which to do it.

In principle, I suppose someone could translate all the common cracking dictionaries using all the common hashing functions in an attempt to provide a tool that could tell you not to use a word like 'password' run through one of those common hashing functions. But given the one-way nature of hashing functions, I suspect the exercise wouldn't tell you anything you didn't already know: if you're dropping a dictionary word into a hashing function and using the output, you already know what you have done, and a coming up with a tool to confirm that for you seems fairly pointless.

Sure -- the source code is on github, linked from the top of the page I mentioned.


The zxcvbn tool shows 185 bits of entropy and a crack time of centuries. It's very easy to type these in yourself and see. 

As several folks have alluded to already, the relevant aspect of the system's security (i.e., excluding any other potential problems) comes down to the properties of the passphrase relative to the capabilities of available cracking tools.

Unfortunately, our intuition is not always a good guide about the level of entropy in a given string, nor does it necessarily help much when trying to factor in the risk from dictionary attacks. If you'd like a quantitative evaluation of entropy for a given string, together with an approximation of crack time and the relevance of particular dictionaries, I'd encourage you to have a peek at zxcvbn.

Note that while this does offer a quantitative look, as is so often the case when Shannon-style entropy is involved, it is not by any means the only way of looking at the problem. See the original article (zxcvbn: realistic password strength estimation) for a comparison with a handful of other guessers of password strength.

Exchange Traded Product is just a broader, less specific term that can encompass Exchange Traded Funds, Exchange Traded Notes, etc. Some people have also started using ETC (rather than commodity ETF), for Exchange Traded Commodity.

Regardless of the final initial in the acronym, though, they all have that same central feature of wrapping up something (or some things) into a form that can be as easily traded as a stock. In some cases, as with GLD, SPY, QQQ, etc., options also become available on the securities.

I hear you!

Having said that, though, a decent derivatives market would be a godsend, and approval for the Winkelvoss trust could be the first step in making that happen.


Who knows how the characteristics of the market might change, though, were it to be opened up to more participants -- and especially if derivatives were to become available on the ETP? (For investors hoping for a dampening down of some of the crazy exchange rate swings -- or at least looking to protect themselves from those swings -- derivatives are the first port of call. Of course we have ICBIT, but it barely qualifies.)

Many thanks for your welcome and encouragement -- I appreciate it!

And yep, that's my real identity. (In case you're wondering why I'm using my real name, I wrote up some separate thoughts on the distinction between anonymity and privacy here: In the Bitcoin Economy, Anonymity and Privacy are Not the Same Thing. I'm big on privacy, but I don't necessarily think that anonymity is the best way to achieve it.)

I'm not sure that the total volume of Bitcoins in the trust would matter as much as the total volume being traded via the trust -- especially once the influence of derivatives on the trust are added in. I.e., if that's where people are active in bouncing the value of BTC/USD around, it doesn't matter so much how many actual underlying BTC are really there, just how many trades are occurring. Other things being equal, a single Bitcoin regularly traded one million times per day would have much more influence on determining the exchange rate than one thousand Bitcoins each regularly traded ten times per day.

BTC doesn't need to be USD 2.0 if the volume of investment and speculation behind the trust and eventually options on the trust outweighs the volume on existing exchanges. In fact, it probably doesn't even need to outweigh it, since as I mention in the article, the amount of gold currently held in ETP form is only around 6.5% of global demand, and yet the ETPs and options on those ETPs exert a very strong influence on the price of gold itself. Whether it's the tail wagging the dog, or whether the tail becomes the dog, it's hard for me to see how an exchange like Mt. Gox could continue to 'lead' when so much investment interest and speculative interest may be brought to bear via the trust.

I don't know whether you've had a peek at the original article, but I also suggest there that if the ETP were to be approved, it would also provide a strong nudge in the direction of creating a 'real' exchange -- one which acts as counterparty to all trades, rather than merely lining up individual buyers and sellers. If that should come to pass, the days of trading network style exchanges may be numbered.

The site which the screenshot came from is not representative of the UK market as a whole. The prices you see there are only a very small number of individual sell offers. In at least some cases, the folks behind the offers on that particular site became involved with Bitcoin only relatively recently and thus had significantly higher entry points than the current 'real' exchange rate. The available offer prices were much closer to spot when BTC was at 75-80 pounds.

UK buyers looking for far better rates can bypass individual person-to-person sell offers entirely and use speedybitcoin.co.uk or one of the other services which charge a small flat percentage over Mt Gox.

In an article called The Economics of Investing in Bitcoin Mining, I suggested that many Bitcoin enthusiasts are deeply attracted by the idea of mining their own, and that there is certainly value in that, but that at the same time, there is significant opportunity cost associated with investing in mining rather than investing in Bitcoins themselves.

As in the nineteenth century, some of the forty-niners traveling to California soon after the discovery of gold in 1848 became rich, especially the early ones, but many others wound up going home again with little or nothing to show for it.

Last week, I put together some thoughts on the potential impact of the Winkelvoss Bitcoin Trust on current exchanges such as Mt. Gox. At first glance, it might seem like exchanges would be dancing in the streets -- after all, shouldn't more potential demand for Bitcoins mean more potential business for exchanges? On the contrary, I'd suggest that the trust, if approved, would 1) wind up becoming the principal price discovery mechanism for Bitcoin, supplanting existing exchanges, and 2) drain much of the existing speculative and investment volume away from the exchanges.

In my view, this is partially due to the larger potential volume for the trust, which would remove many of the barriers that currently keep the ordinary person on the street from participating in the Bitcoin economy, and partially due to the sorry state of existing exchanges, especially their universal failure to grab the counterpary risk baton and run with it. (Existing exchanges just provide networks of buyers and sellers, as opposed to acting as counterparty for each trade.)

The full article is here:

Winklevoss Bitcoin Trust May Become THE Price Discovery Mechanism for Bitcoin

Comments, criticisms, corrections welcome...

That's an interesting question -- no idea what the answer might be now, but you can get a first glimpse from an historical perspective by comparing the graph of total network hashing rate with the graph of USD vs. BTC:

http://bitcoin.sipa.be
http://bitcoincharts.com/charts/mtgoxUSD

The latter has a checkbox for "Custom Time", which lets you set whatever interval you'd like.

When BTC took its sudden dip in mid-2011, for example, there was a less sudden drop in total network power which began soon after. No doubt there's more to the decision, but the exchange rate is bound to be a big part of it.

Without jumping into the rest of the to-ing and fro-ing, just a point of interest: this option would be severely undervalued relative both to pricing based on historical volatility and to other options currently up for grabs on BTCT. Assuming for the sake of argument a dividend yield of 30% and 5% cost of capital, this comes out to around 80% implied volatility. By contrast, most of the options available on BTCT run into triple digits -- 250%, 300% or more. In other words, if you were to persuade someone to sell you these options, you would, theoretically, be getting a great bargain.

To be fair, nearly all options pricing I've noticed on the exchanges has bordered on ludicrous. Maybe a realistic level would be somewhere between the bargain you've suggested and the nosebleed levels seen on the exchanges. 

Just to clarify a little, this is incorrect. While the original ratio will be 0.2 BTC per share, that will decay over time, and the trust will spend a portion of those Bitcoins. There is nothing sinister or odd about this -- it is, on the contrary, common and entirely expected by investors.

As I mentioned earlier in this thread, the relevant passage is on pp. 40-41 of the filing:

https://bitcointalk.org/index.php?topic=248013.msg2636138#msg2636138

Specifically:

"The Trust will transfer Bitcoins to the Sponsor Custody Account in payment of the Sponsor’s Fee and transfer Bitcoins to the Trust Expense Account and sell Bitcoins to raise the funds needed for the payment of all Trust expenses not assumed by the Sponsor... As a result of the recurring transfers of Bitcoins to pay the Sponsor’s Fee and the Trust expenses not assumed by the Sponsor, the net asset value of the Trust (“NAV”) and, correspondingly, the fractional number of Bitcoins represented by each Share, will decrease over the life of the Trust."

Again, there is nothing in any way surprising about this.

Unfortunately, nearly all the options available are priced with implied volatility through the roof -- three figures the last time I checked, and far above historical volatility...

One guaranteed method for eliminating click fraud entirely, a method which I used on my own advertising network (since closed), is not to charge advertisers by the click or by the impression at all, but rather to allow them to bid for placement only. Whatever they bid translates into a probability that their ad will be shown for any given page request on any given site in the network. Therefore, nobody cares whether a script kiddie in Timbuktu (no offense, denizens of Timbuktu!) wants to sit and click and click and click -- or re-load and re-load and re-load -- because nobody is paying either for impressions or for clicks.

The weighted probability method also enables advertisers to bid specifically for placement on individual sites, should they wish to do so -- where their bid for that site is then pooled with a share of the network-wide bids to derive a probability for any given ad to appear on that specific site.

The only 'catch' with this method is that it does not permit real-time bidding: there needs to be some window of time during which probabilities are fixed, and that window of time needs to be reasonable for a human to manage. In my own case, I used a 1-month window, meaning that all commitments to advertise for a given month were due by a certain date, and those payments then generated a weighted probability for how often any given advertisement would appear in response to any given request to the ad server during that month. If you were catering to very hands-on advertisers, that window could probably be dropped to two weeks or even less.

Give me a shout via PM if you're interested in further details; depending on what exactly you'd like to put together, it's vaguely possible some of my old and creaky PHP code for implementing this could be of some use to you.

There's a few different ponderings all wrapped up in one. 

The shares of many non-US companies, including several H shares (companies incorporated in the mainland but trading on the Hang Seng), trade on US exchanges via ADRs -- American Depositary Receipts. It's a bit of a stretch to put it this way, but ADRs are a bit like passthrough shares traded on the likes of BTCT.co: in the case of an ADR, a US financial institution holds the 'real' shares overseas and issues certificates (the ADRs) that trade on a US exchange. The underlying shares themselves are called American Depositary Shares (ADSs). This is how it works for shares of a single company, such as CHL (China Mobile), for example. (A bank can also issue certificates in more than one country, in which case they're called Global Depositary Receipts, or GDRs.)

Although it's ADRs rather than ETFs you'd be after if you wanted to trade shares of a single Chinese company on a US exchange, that doesn't at all address the part of your question about being denominated in a decentralized currency. For the 400,000 shares in ASICMINER themselves to become available for trading on any of the established exchanges, they would need to be registered appropriately and denominated in the currency of that exchange. IMHO, that seems unlikely.

Should Friedcat ever want to do it, though, I wonder about a reverse takeover -- a shortcut to an exchange listing, whereby a private company buys control of a listed company, and the private company's shareholders then exchange their shares for shares in the public company. (The most famous example of a reverse takeover was Buffett's taking control of Berkshire Hathaway.) Given Chinese restrictions on foreign investment, I imagine this could be a hornet's nest of complication and a potential disaster for non-Chinese shareholders.

Just to follow up -- yes, I personally experienced this when experimenting with a move from Bitcoin-Qt to Multibit. If you dump the private keys only for the addresses displayed in the Bitcoin-Qt GUI list of receiving addresses and then import them into anything else (e.g., Multibit), you can watch the transaction list rebuild in the new client, and everything will look OK at first as those transactions reappear, yet the total balance will be wrong.

The reason?

The list of receiving addresses displayed in the main interface of Bitcoin-QT does not include all the extra addresses created behind the scenes for receiving change. Those addresses are not displayed directly in the main GUI. So, exporting the private keys only for the receiving addresses which you can see in the GUI does not give you a complete set. As a result, if you move that partial set to a new client, your total balance will be missing all the amounts you received back to those hidden change addresses.

Although this is just a basic feature of how Bitcoin transactions work -- total inputs and outputs need to balance out -- the result when trying to move between clients can be a little counter-intuitive if you're not expecting it and look only at the GUI list of receiving addresses. This also explains some of the horror stories about people spending a small amount from a paper wallet and then accidentally losing the rest of their balance: they didn't realise that spending a small amount from the paper wallet meant that wallet became worthless, because the remaining balance from their transaction went to a different change address. (These kinds of details also mean it's often more straightforward just to transfer an entire balance to yourself when moving between wallets, rather than trying to preserve all of the original wallet's transaction history, addresses, etc.)

I think it's interesting that within the confines of a single exchange (Bitfinex) -- where, notably, borrowed funds cannot be removed from the exchange -- the demand for dollars to buy Bitcoins on margin appears to be much higher than the demand for Bitcoins to short. The result is a perpetual asymmetry in prevailing interest rates, with USD interest rates extraordinarily high, BTC rates exceptionally low, and the ratio between the two growing or shrinking depending upon whether the USD/BTC exchange rate is rising or falling, respectively.

Funds of this type are designed to generate revenue via management fees. The sponsor's fees and trust expenses are covered by the Bitcoins held in the trust itself. Thus the gradually decreasing net asset value (understood in BTC per share) cited in the filing, pp. 40-41; the same gradual process of NAV erosion also occurs in other commodity ETFs such as precious metal funds, etc.