Sorry if this has been addressed before:

In Ethereum (or in Bitcoin with Turing-complete scripts), could this idea be used by a miner to delay competing miners, rather than sabotage the network with a DDOS-type attack?

Namely, the malicious miner creates a transaction with a script that takes a long time to complete, and a fee that forces it to be included in the next block; and computes the script's outcome before issuing it.  Would that give him an edge over other miners?   

But if he thought that bitcoin was worth 390$, and bid for 390$, why would he not also buy on the exchanges, after the bids closed, until the price rose to 390$?

Suppose that, as you suggest, he refrained from buying those 375$ coins at the exchanges because he wanted bitcoin to remain underpriced and thus discourage other bidders.  If that is what he thought, he should have bid at some price between 375$ and 390$, probably closer to the latter.  But them after the auction closed, why didn't he rush to buy those underpriced coins?  If he thought that buying 2000 BTC at 390$ was a good deal, why would he not buy 1000 more at 375$?

Perhaps he was not a rational player, or was lazy, or became more pessimistic betwen 12-04 and 12-05, or had some other reasons; but what about all the other whales, including those who did not enter the auction?  On 12-05, obviously none of them thought that BTC was worth more than 376$.  

So, Tim may have bid for more than 375$.  But it is also quite possible that all the bidders (including Tim) acted rationally, and all of them bid below 375$.  

In the US you have the right refuse to say anything that could be used as proof against you.  (That constitutional amendment was intended to discourage the police from using torture to extract confessions from suspects, since such confessions are often bogus, and the police may end up torturing innocent people.  However, thanks to the CIA and TV, torture is now again fashionable in the US, so maybe they will amend the amendment soon...)

On the other hand, the Japanese police must already have interrogated Mark and other MtGOX managers and staff, for the investigation that Kobayashi requested.  But the police cannot be transparent while it is investigating a crime.  The findings will be published only after the investigation is closed, IF criminal charges are filed.  Then the relevant evidence and depositions will be attached to the case documents.

˜Provide liquidity" is a fancy way of saying "sell a lot of coins whenever the market price tries to rise".

Fixed that.  AFAIK we still do not know how much he paid last time.

Claim: Tim does not think that bitcoin is worth more than 375$.  Proof: If he thought it was worth 390 $ (say) he would buy at the exchanges until the price went above 390$.  But the price is barely moving from 375$.  Therefore he is not buying coins at 376$ or more.  Therefore he does not think that bitcoin is worth more than 375$.

Now, if he does not think that it is worth buying even 1000 BTC at more than 375$, why would he try to bid for 2000 BTC at more than 375$?

Same for all other people out there who still have money to invest.

Said another way, the reason why the market price is not more than 375$ now is because everybody who has money to spare, including Tim, now thinks that bitcoin is worth less than 376$.  So, if no one with money wants to buy 1000 BTC from the exchanges for more than 375$, why would they want to bid for them at the auction for more than 375$ ?

Would that be OK with the bartender?

There is only a fancy sports car on that cover.  Come on.  The get-rich schemes that I see advertised late night on my TV typically show a successful investor on a luxury yacht, with three girls in bikinis on each side.  Bitcoin infrastructure must improve a lot before we can see significant adoption.

Ok, it is possible that he bid for several lots of 500 BTC or less, all at different prices, and won only the highest ones -- which happened to add to exactly 2000 BTC. Seems strange, and a lucky coincidence, but I agree that is possible.

I can't imagine why someone would want to pay above or near market for less than 1000 BTC.  So their bids were probably all below market.

I can't imagine why Tim Draper would enter a syndicate with a bid for 2000 BTC, since he could bid directly for one lot.  Unless it was just to show support for Mirror.

My heart just melted.

I did not intend to call you anything; apologies if it sounded that way.  However, I got this impression that you are not very familiar with how the judicial system works.

I am not a lawyer either, but at work I have had to follow legal proceedings in various occasions, served as juror a few times, and even sued someone once (a lawyer, in fact -- and, amazingly, I won).  Sure, most of those experiences were here in Brazil, where the laws and legal standards are different from those of Japan; but the general principles that underlie the legal system seem to be the same (not surprisingly, since Brazil and japan both copied a lot of their government structures from the US).  In particular, everywhere it is the case that one must retain a lawyer to interact with the courts.  A court of law is totally unlike a private company, and you cannot demand or argue with a judge (or his appointed trustee) like you would with the manager of a bitcoin exchange.

Based on my limited experience, I would say that, if you wish to have any influence in the liquidation, you must absolutely hire a lawyer that understands the Japanese bankruptcy laws and precedents, ask him what can and cannot be done, and let him interact wth the trustee.   If you don't have the means or patience to do that, you should just wait for the trustee to figure out a plan that the judge approves, file your claims according to his instructions, and accept whatever you get from him.

But the word "disappointed" in the URL is misspelled.  A minimally self-respecting troll would not make that mistake. That detail strongly implies that the article was real and was written by a typical WSJ reporter.

Some random statistics about the presumed "BitPay.com wallet" in this site:
http://www.walletexplorer.com/wallet/BitPay.com

From the first entry on 2011-07-02 to about 2014-12-02:

About 690'500 transfers into the addresses in this "wallet", adding to about +1'270'000 BTC

Largest single input into those addresses:  +9483.46000000 BTC on 2012-06-24 00:39:05
( transaction d4f7fbbf92f4a3014a230b2dc70b8058d02eb36ac06b4a0736d9d60eaa9e8781 )

There were

    47 inputs of at least 2000 BTC
  137 inputs of at least 1000 BTC
  316 inputs of at least  500 BTC
  578 inputs of at least  250 BTC

In 2012 there were two inputs of ~230 BTC from addresses in the "SilkRoadMarketplace" cluster.
Also one input of ~590 BTC from wallet "Btcst.com-pirateat40".

KnC miners made several large inputs between Nov/2013 and Feb/2014 (buying equipment for their centers?)

The payment for the 0.5 M$ home of BFL's Josh Zerlan, which was reported as being made through Bitpay
http://www.kmbc.com/news/bitcoins-used-to-buy-johnson-county-home/26078516
seems to be this 1060 BTC transaction on 2014-04-28 22:08:49 UTC :
http://www.walletexplorer.com/txid/470dd9660d62b60ce8451c22392ba119b1c0b97d0927d6d5ef3ee0c2da043585

There were ~633'000 inputs of 1.0 BTC or less; ~440'000 of them were in 2014.

It would be more informative to use dollar amounts instead of BTC amounts, since the price varied so much in that time span.  Coming soon.

However, as observed before, the BTC amounts of the largest inputs seem to be roughy constant over time. 


 

In principle, the syndicate is not trying to buy coins for Barry Silbert, SecondMarket, or the SMBIT fund.

The syndicate collects money from small bidders (starting at 25 k$ minimum bid) and joins them into the 1000 or 2000 BTC bids for the USMS auction.  If any of these merged bid makes it to the winners list, SecondMarket distributes the coins to the corresponding small bidders, and collects a fee for its service.  Otherwise they return the money to the small bidders, and shrug off their wasted work.

They had 104 small bidders which together wanted to buy 124'127 BTC.  Obviously when they assembled the bids for the USMS auction it would be pointless to bid for more than 50'000 BTC.  So they must have considered the highest among those small bids, until completing 50'000 BTC, and rejected the rest.

One cannot tell from those numbers whether any of the bids were above or below market.

OK, discussing what you would like to happen may be useful as a philosophical exercise, or if you intend to lobby Congress to change the laws at some time in the future.

However, that is pretty useless for the goal of recovering some of your loss at MtGOX.  Surely you agree that the judge cannot change the laws, nor ignore the precedents, right?

So, for example, if there is a law that says that any payout must be in yen, then the payout will be in yen, no matter how much the creditors dislike it. 

If the law says that the claims are to be computed as deposits minus withdrawals, that is what the judge will order, no matter what the creditors think of it.

If the trustee points out to the judge a case from 1957 that looks sufficiently similar to this one, in which the Mongolian Naval Treasury bonds that the company still had were auctioned rather than distributed as such, the judge will want to take the same decision his colleague took in 1957 -- even if the creditors find it absurd to compare bitcoins to those bonds.  It would take a good lawyer -- not a letter from the self-help group, not a blogpost by Antonopoulos, not a video Roger Ver, not an apparition of ghost of Satoshi in the courtroom -- to convince the judge that the 1957 precedent is not germane; preferably, by pointing out another case from 1962 that was decided the opposite way, and arguing -- withe the proper legal language and reasoning -- that MtGOX resembles said case more than the 1957 one.


The future of bitcoin is totally irrelevant for the liquidation.  As in any process, the judge is bound to think only about the matter of the case, which is to fairly distribute the present value of the assets to the creditors.  I have pointed out before some practical reasons (apart from any explicit laws that may exist) that may convince the judge to order the auction of the coins, rather than their distribution as bitcoins.

Madoff's personal property, that was seized as part of the liquidation, included art pieces and luxury cars.  The courts in that case couldn't care less about the value that those items could have umpteen years in the future. It auctioned them, implicitly trusting that the auction bidders would consider such speculations when submitting their bids.

In the case of MtGOX's bitcoins, the question of future value is even less relevant -- because bitcoins are not unique works of art, but fungible commodities with a large market.  220'000 BTC is a lot bitcoins for one person to own or buy, but is still less than 2% of all bitcoins issued so far, and less than the volume traded in the exchanges in one day.   If those coins are auctioned for 37'000 ¥ apiece, the judge will not understand why giving 37'000 ¥ to a former client is less fair than giving him 1 BTC.  And, frankly, neither do I.


I understand that you are questioning whether the company really owed >800'000 BTC, and/or whether it had really lost any BTC that it owed to the clients.  Is that so?

The decision to liquidate had to be taken according to the only data that the court had at the time, namely their books -- that showed the impossibility of MtGOX honoring its obligations to its clients.  The numbers that Mark submitted to the court are a matter of public record.

If the judge had considered the possibility that those numbers were false, that the insolvency was a big lie, and that the actual assets might in fact be sufficient to meet the actual obligations, that would be even more reason to liquidate the company.  How could the judge let the company keep working, if the company's insolvency was faked?  He would be approving the theft of real bitcoins from the real clients.

The suspicion of embezzlement and doctored books is in fact automatic in bankruptcy proceedings; that is why the court appoints a trustee with no connection to the company and no vested interest in it, and orders the company to turn all assets and internal records to him.  It is part of the trustee's job to check whether the books are consistent (and we already know that, in this case, they aren't, although the extent of the discrepancies is not known yet) and request criminal investigations by the police if it sees evidence of fraud (which Kobayashi apparently did)

That implied suspicion is also why the clients will have to explicitly submit their claims to the trustee, and he will have to validate them, before any payoff.  He cannot trust that the obligations in the company's books are legitimate.

The Madoff ponzi was much bigger than MtGOX: 64 billion US$. WorldCom, Enron, Lehmann Bros, Washington Mutual, were many times bigger than MtGOX, and much more complicated, with dozens of subsidiaries, branches, etc.. Even the Brazilian oil company OGX and the Brazilian MLM scam TelexFree were much bigger.

In Japan there was, for example, memory manufacturer Elpida (5.5 billion $) and Japan Airlines (at least 10 billion $).  (JAL was one case when the bankruptcy did not end in liquidation: the government bailed it out in part, and put pressure on creditors to suspend their demands and give it a chance, so it was back in normal operation after 2 years.)


You cannot consider only the pluses and ignore the minuses.  When it closed, MtGOX did not have 220'000 BTC, it had minus 660'000 BTC.

Sunlot wanted to do just that: convince the Japanese court to stop the liquidation and sell them all the MtGOX assets -- including the 220'000 BTC -- for 1 BTC, but without the 660'000 BTC debt to their clients.  That would have been very sweet idned -- for Sunlot, not at all for the clients.

Those 220'000 coins indeed are a relatively uncomplicated part of the assets (although they may have to be auctioned; if not, returning them as BTC may be a big legal and accounting headache).  The messy parts include the loans of 13 million dollars to mark and his other companies; the money in various bank accounts; the suspended Coinlab lawsuit; the assets of the US subsidiary (and others perhaps); and understanding what happened to the 660'000 missing coins.


The decision to liquidate the company was totally inevitable given that its obligations were  ~500 M$ more than its assets.  Far from being "fraudulent", the decision to liquidate MtGOX may be the only part of this saga that was totally transparent and undeniably correct under the law...

The purpose of the bankruptcy law is precisely to suspend any creditors' lawsuits already underway, and block any new ones.  It will not prevent more specific lawsuits in the future, civil or criminal, against the people involved -- e.g. for fraudulent management (if Kobayashi does not start some himself).  But the civil lawsuits against the company MtGOX will be aborted, and there will be no more point in starting new ones  -- because even the corpse of MtGOX will have ceased to exist.

By the way, when people say "file for bankruptcy", that is a short for "ask the courts for protection from creditors' lawsuits as specified in the bankruptcy law."

My understanding of what happened in the US courts is limited.  I think (not sure) that several things happened, in some order:

* MtGOX USA filed for bankruptcy protection in the US too, after MtGOX KK did so in Japan.

* An US judge aborted that MtGOX USA bankruptcy process because he concluded that it was best for the Japanese court to handle it too.

* Greene and Lack sued MtGOX in the US.  They claimed to have evidence of wrongdoings by MtGOX managers.

* The court requested Mark to show up to testify in the Greene & Lack case.  Mark was very relctant to do so.

* The Sunlot guys convinced Greene & Lack to withdraw their lawsuit and support Sunlot's plan to stop the liquidation and take over MtGOX.

It must be noted that Sunlot had already promised immunity to Gay-Bouchery and McCaleb, even before starting any investigation.  And they never replied when asked about whether they would extend that coutesy to Mark too.  if we are talkng of cover-ups and evading justice, then...

Also Sunlot never replied when asked whether GB and McC and Mark would get their share of the 220'000 together with (or before) the other clients.  (IIRC, Japanese liquidation law, like the US one, says that management is always last in the queue of creditors, so they should get nothing in this case.)

The site http://www.walletexplorer.com/ tries to analyze the blockchain by clustering addresses into "wallets", where a "wallet" for them is a set of addresses that appear to belong to the same owner.

I could not find an explanation of their algorithm, but I would guess that two addresses are assumed to have the same owner if both appear as inputs to the same transaction -- since both private keys are needed in order to sign such transaction.  Thus, if even if a bitcoin-accepting business uses a separate address for each customer, once some coins in those addresses are dumped at the same time into a common bucket, this algorithm will identify them as belonging to the same owner.

The person running that site has identified several famous "wallets", such as the receiving addresses of BitPay, Coinbase,Bitstamp, etc.. Thus, for example, the following link shows all the traffic into or out of all the addresses that they associate to BitPay, merged together (instead of separated by address): http://www.walletexplorer.com/wallet/BitPay.com

With some kludgy hacking, I have collected the data of that "wallet" and plotted the number of transfers into or out of them, per day, and the BTC volume in those operations:





[ Click on the images for the full-size versions. ]

Beware that there may be addresses that belong to BitPay but were missed by the algorithm, so those plots may undercount the actual traffic.  on the other hand, some famous transactions (like the 0.5 M$ house purchase by Josh Zerlan of BFL, or the alleged downpayment of 1 M $ by hashTrade to BFL) can indeed be found in that "wallet".

There are a number of puzzling features in those plots, such as the nearly flat number of deposits per day over the past year (apart from weekly variation).  Basically, those addresses have been receiving 1000--1500 deposits per day, amounting to 500-1000 BTC per day on the average.

There is a Black Friday spike in the number of input transfers (3190 on Nov/28, about 2x the usual number) but not in the total amount of BTC transferred; which presumably means that the ~1500 additional Black Friday purchases were mostly small.  

The data file used in that plot is here.
The coumns are the date and 3 groups of 4 numbers: outputs, shuffles (coins moving between addresses in the same wallet, with zero change in its BTC value) and inputs.  In each group, the four numbers are op count and total BTC moved in that day, then the same accumulated since the start of the file.

Hope it helps.  I hope to post more details later tonight.

Thanks for the clarification! 

However, is it necessary to buy the same dollar amount every time?  Is that strategy optimal in some sense, for that model?

If the increments in log(price(t)) have a normal distribution, the expected value of price(t) as t goes to infinity will slowly increase; won't it?  If so, that strategy works in theory because the assumed model is optimistic.   

I find it strange that they used a generator too, but perhaps they had no quick way to upgrade their wiring or feed?

On the other hand, would it be OK to run a diesel generator 24/7 in that area of town?  Is it an industrial area?

Since they did not pay for the machines, they could pay more for the elecrticity and still make a profit.

Phase II: Online Auction Period
Date: Thursday, December 4, 2014, from 8:00am EST to 2:00pm EST
http://www.usmarshals.gov/assets/2014/dpr-bitcoins/?m=1