Confirmed, Huobi's daily volume for Sat May/17 was 9.3 kBTC, the lowest on record since Nov/03 (when the price was 210$).

Bitstamp's daily volume was 2.3 kBTC, only a bit over Mar/15's 2.2 kBTC (were they down on that date?).  Apart from that, it was the lowest volume since 1.9 kBTC on Aug/11, 2013 (when the price was 95$).
 

IIRC a Japanese bank closed MtGOX's account too, not clear why.  And an Australian bank closed the accounts of Australian bitcoin exchanges.

This old article (Nov/2013) may be relevant:
http://www.forbes.com/sites/kashmirhill/2013/11/15/bitcoin-companies-and-entrepreneurs-cant-get-bank-accounts/

Could be simply "why are we serving a customer whose business is to undercut our business."

Could be the popular perception that "bitcoin is the currency of illegal drugs, weapons, and pedophiles."

Could be that the BIT's legal status is murky.  What is their legal nature and strcuture? Does SecondMarket hold the bitcoins themselves, or are they a proxy for some other company?

I suspect that there are Chinese who open an account at Huobi just for the pleasure of moving orders up and down the book, while carefully avoiding being caught by an actual trade.

Huobi's volume for today (Sat May/17 UTC) is only 8.8 kBTC.

There are still 4 hours left before 24:00 UTC, which would be 08:00 Sunday morning in China, so perhaps it may still grow a bit.  In any case it will probably be the lowest this year, much less than the 18.0 kBTC of Mar/23 (a Sunday when the exchange was down for ~4 hours in mid-afternoon), and the 15.3 kBTC of Jan/31 (the Chinese New Year Day).

Previous comparable lows were 9.6 kBTC on Dec/03, 2013 (following the first crash after the all-time high; was the exchange down during that day?); and 8.7 kBTC on Nov/03, 2013 (the eve of the big rally, after the price had been mostly flat at 200$ for some ten days). 

Considering the latter, this low volume may be taken as a bullish sign. 

With a link to an interesting article:
http://www.ofnumbers.com/wp-content/uploads/2014/04/Learning-from-Bitcoins-past.pdf

I don't know what you mean by "regulate bitcoin" or "inside the bitcoin system", but fraud in bitcoin-only enterprises and investments is absurdly common, even of we ignore MtGOX and other exchange scams.  For example, Danny Brewster was able to pull the Neo & Bee scam because its shares were sold for bitcoins through a Panamanian company that was not registered as a stock exchange, not even in Panama.  He probably would not have been able to do a real IPO with ordinary money on a regulated exchange; his past "career" in the UK and the lack of a real business plan would have been exposed.

User @JorgeStolfi tried to do that for a while, but gave up some time ago, partly because he found that several large Chinese exchanges (perhaps 30% or more of the total) were not listed in any convenient source, not even in bitcoinwisdom.

A weird Korean (maybe Japanese?) movie I saw on cable:  A young doctor of traditional Chinese medicine lives holed up in his apartment caring for his beloved wife.  She is presently dead, but thanks to constant herbal baths and other treatments she has been in a stable condition for months, and the devoted husband is patiently waiting for her to recover.  (Which is only fair, since she did him the same favor some years before.) I don't know why I thought of that movie while looking at Huobi's charts.  

The numbers that I find googling are closer to 0.5 J/GH a couple of months ago.  Is 0.2 J/GH the current value?

IIRC Mark had registered the "bitcoin" trademark, not the "biitcoin.com" domain.  Is this right?

Thanks.  You mean that a 1 GH/s machine costs 2000$ and consumes 0.2 W? Not 0.2 kW = 200 W?

Wait, I am pointing to charts issued from the Holy Sanctum of the Bitcoin Church -- and *I* am the FUD machine?  

Is it?  What factor should one apply to those number to get the real cost?

But if I bought even one satoshi, I would not be able to brag about my trading performace any more.  

And I would be racking my brains now, trying to decide whether I should sell half of my bitcoin holdings and use the fiat to double the other half, or create my own private exchange so that I can trade with myself and not worry about the price anymore.  

Sorry to ask again, but it seems important:

could you please print "..." (three dots) above and below the order book summary (the rightmost table, above and below the current price) when there are blank lines in those positions?  

Every couple of weeks there are people asking on this forum, "why are there so few bids at Huobi" etc.  People do not know that the order books are truncated by the exchange's server, so they think that the summary shown comprises the entire book.

In some cases the server may return the whole order book, without truncation; so that the summary shown on the chart spans the complete order book.  Even in that case, printing "..." after a complete book summary is not as misleading as showing a truncated book with no indication that it is truncated.  

Yeah.  Somce clients create zillions of tiny book entries, with small BTC amounts and at prices that are often 1-2 cents of yuan apart. Thus when the books are truncated after N entries they are still very close to the spread.

Credit cards make huge profits from their fees, so the cost is much less than what they charge.

That's a good question indeed.  I am  looking at these plots:
https://blockchain.info/charts/n-transactions?showDataPoints=false&timespan=&show_header=true&daysAverageString=7&scale=0
https://blockchain.info/charts/cost-per-transaction?showDataPoints=false&timespan=&show_header=true&daysAverageString=7&scale=0
Note that there was a marked increase in the number of transactions around 2013/Dec/14, and another smaller hump around 2014/Mar/20.  At those times, the cost per transaction actually increased.   Thus it would seem that the cost is more than proportional to the number of transactions, i.e. the marginal cost is greater than the average cost.

Perhaps this effect is due to some marked change in the character of the traffic at those times, such as a decrease in the number of transactions per block.  (I understand that the expected cost of processing one block is largely independent of how many transactions it contains; is this correct?  If so, the cost per block should be a "cleaner" parameter than cost per transaction, right?)

Only if you ignore the fee.  My bot paid generous amounts in fees over the years, to put it mildly.
[/quote]
Since MtGOX was not doing what clients thought it was doing, the fee

I am pretty sure that the court will consider a creditor to be a person (or company), not an account.  Creditors will have to identify themselves and state their claims.

If you thought MtGOX was going bankrupt of course you would not have put your money and coins there in the first place.

People with multiple accounts may try to be smart and claim only accounts where they mostly deposited, omitting those from which they mostly withdrew.  That would be an attempt to defraud the court (like a contractor claiming a debt that was actually paid in cash), which means jail if caught.  For money withdrawals, the risk is too big.  For bitcoin withdrawals, the risk may be smaller but an audit with expert assistance may identify the owner of the omitted account.

The accounting was bogus in the sense that the coins that clients thought they were buyng and selling and/or the money they thought that Mark was keeping for them  did not actually exist.

It is risky to state publicly suppositions about crimes, but in restrospect there is only one plausible explanation for the withdrawals delays that started many months before the total block, including why the "malleability bug" was not understood and fixed promptly. 

There is another website devoted to the analysis of the database (the one with many colorful plots), whose link I cannot find now.  Some of those plots are hard to understand, they seem to show lots of trades at prices well removed from the then-current market prices. 

One of the main worries about that database is whether accounts of some "friends" were inflated in order to snatch a large portion of the remaining coins.   Also there is the possibility of insider trading, fudging with "dead" accounts, mislabeling of managers' accounts as accounts of friends, etc..  Basically, since the previous management was caught lying (about MtGOX's solvency), and their role in the disappearance of hundreds of millions of dollars has still to be explained, one cannot trust any information that comes from them.

Yes, pyramid schemes usually have that effect. Those who invested will only make money if more people invest (and hold) so that the price goes up so that they can sell with profit.

You don't see people who bought Apple or Google stock pestering everyone in sight to buy too. Real investors know that other real investors do not pay attention to hype, they look for fundamentals.

As for "increasing adoption", do not see it in the blockchain:

https://blockchain.info/charts/estimated-transaction-volume
https://blockchain.info/charts/n-transactions

Also this chart seems to say that the bitcoin system currently costs 30 USD per transaction:
https://blockchain.info/charts/cost-per-transaction
The cost of running the system is currently hidden in the inflation caused by mining, but in the long run will have to be paid by users as transaction fees.  If the numbers above are correct, bitcoin will be inviable for payments below 1000 USD.

Huobi is weird, sometimes 10-15 minutes without reporting a trade and then suddenly retroatively reports some large trades and order book changed completely then freezes again.

You could use 4 weeks (= 28 days) instead of 1 month. Close enough to a month, but strictly regular.  Ditto 52 weeks (= 364 days) instead of 1 year.