Outside of the billions incident a while back, I don't think there has ever been a reorg involving more than a few blocks. And we sorta think that no one has the ability to actually redirect a transaction right now, or even will have the ability in the near future.
This is true only on the "public" block chain. I'm inclined to believe Tom Williams' story (of Mybitcoin's fame) that his machine had seen some sort of prepared, non-public block chain due to an attack. Having seen the inside of the software and having heard of the testing methodologies that it undergoes I' thinking that the further failures related to mishandled chain reorganizations are inevitable. Or, if you want to make a new thread on the subject, I can try to fill in the details for you.
Personally: I'm disinterested. But for the wider community some sort of anti-propaganda would be an asset. Something along the lines "Conscientous accounting auditor's guide to the bitcoin internals." I think the saddest part is that you (presonally) are one of the very few bitcoin-positive people who is actually willing to admit problems and discuss the deeper philosophical issues related to its implemenatation. I have this unfortunate experience that the full 100% of the bitcoin-positive people I've got personally acuqainted with were either fraudsers, gross-incompetents, fly-by-night operators or some combination of that. |
|
|
|
This is not a well-researched statement. There is no reduction in security of ECDSA by using the same key on multiple signatures. If there was, then no one would use ECDSA, as most other applications for it cannot accommodate new keys for every exchange.
[...]
It is an insult to cryptographers/mathematicians to, in any way, compare WEP to elliptic curve cryptography.
I fully agree with you that ECDSA is mathematically sound. And comparing it to WEP was an insult. But I will disagree with you from the standpoint of implementation engineering. In my career I was involved in several fracas where a mathematically sound idea got corrupted by the cargo-cult style of its implementation in software or hardware. Side-channels are hard to detect, and the way the current Satoshi bitcoin client development is progressing, I will probably be willing to bet a small sum on an interesting crypto-snafu that's going to happen in some of its branches. The above isn't a mathematical theorem, it is my hunch based on past experience with implementations of patented cryptographic methods. I have signed at least two NDAs related to the above, as of now I don't remember if they had already expired. |
|
|
|
So are you storing money as a float or just displaying it rounded to nearest cent?
There's nothing wrong with using floats to represent money, so long as the floats used are decimal. Twenty first century brought up the resurgence of decimal floating point. Nowadays pretty much everyone supports them. Check out the following page: http://speleotrove.com/decimal/Decimal floating point good! Binary floating point bad! I think Bitcoinica is implemented mostly in Ruby, so the following will be relevant: http://ruby-decimal.rubyforge.org/ |
|
|
|
very comprehensive explanation (it talks about shorting stock, but trading on margin is the same for stocks, forex, and whatever btc is considered).
Actually the Investopedia isn't the appropriate explanation. They deal with a regulated (or even over-regulated) market. What happened here is a so called "shakedown in a bucket shop". Wikipedia has a good starting point: http://en.wikipedia.org/wiki/Bucket_shop_(stock_market) But to really understand the economics of the operation of a bucket shop like Bitcoinica, you'll have to visit a physical library and look for some dusty paper books. Especially in the USA, where SEC was really successfull at shutting this type of scams. |
|
|
|
I believe that whether or not you use new addresses for any particular transaction should be a personal choice and you shouldn't try to encourage or discourage others from doing either. Some of us do not care at all about the pseudo-anonymity of Bitcoin. Some of us do.
This is a weak and not well thought out statement. Reuse of addresses could make it easier to attack the elliptic curve crytography that underpins the security of particular bitcoins. So the choice isn't just privacy vs. no privacy. The additional influential factor is: Do we believe that the reuse of the points on the elliptic curve weakens the resistance against the possible cryptological attack on private keys? I mean at one time WEP was considered "Wired-Equivalent Privacy" wiith no hint of sarcasm. Are you willing to make the same statements about ECC? |
|
|
|
What exactly is it that you imagine Gavin to be dancing around? Even better, how did you make it two months on this forum without learning about this?
I imagine Gavin dancing around GAAP (Generally Accepted Accounting Principles). Actually, I'll ask you the opposite. Do you see any easy-to-use software advertised here on this forum that is upfront about: 1) bitcoins could disappear from your wallet 2) transactions will get their timestamps changed in some obscure ways I'm not talking "upfront" as "discoverable here on the forum after two months of reading". I'm talking "upfront" as popping up a message when that happens or having any other mechanism to alert the user about those occurences. |
|
|
|
You should really stop saying "chargeback". It isn't even slightly similar to what actually exists in the bitcoin system.
Tell this to your accountant. How about "wholesale transaction reversal and freshly mined coin destruction"? Does this sound better to your ear? It certainly is more true. Chargeback is the proper term of art when writing about accounting in English. Here's the Gavin Anderson dancing around the issue in the section of this forum related to the alternative currencies. https://bitcointalk.org/index.php?topic=43465.msg520923#msg520923Since that section is the "wildest of the Wild West" it may have been missed by many of the readers of this forum. |
|
|
|
No one needed to audit the code to learn this, it has been very well known since day 1. It isn't anything like a chargeback though, since an arbitrary spender can't just make the decision to reverse one of their transactions on a whim.
Well, in a way what you say is true. And yet nobody important from the bitcoin community was willing to sign their name under the sacramental formula: I certify under penalty of perjury that the foregoing is true and correct. Executed on ...
In fact at least Gavin Andresen closed his company instead of facing an audit. Simultaneusly, I have to commend him for not accepting patches that change or completely remove the chain reorganization functionality. There's lots of patched and otherwise badly designed code floating around that makes the assumption about nonexistence of chargebacks. This is going to be an interesting subject for litigation: 1) on one hand the chargebacks "should never happen", 2) on the other hand this is the major feature of the protocol as described in the whitepaper. There will be interesting legal debate wheter removing the chargeback functionality was: (a) negligent accounting, (b) fraudulent accounting, (c) conspiracy to commit accounting fraud, (d) something else. Interesting times ahead. |
|
|
|
2. It's not a full NTP impl, just stupid SNTP w/ RTT correction (usually +-1ms on a box with properly working ntp)(and i0s original code doesn't even do RTT correction...).
3. Got a cross-platform way to determine if the system we're running on has a running and working ntpd? Does windows even have real NTP support? iirc even W7 only has a SNTP daemon to periodically sync ala ntpdate-from-cronjob.
In Re:2. I believe you are at least order of magnitude optimistic with +/-1ms standard deviation. To be safe I would assume 100ms standatd deviation with just properly working ntpd (and not openntpd). So two orders of magnitude, if you take your time over the net. It is easy to get much better time over the air thanks to Uncle Sam's (or Uncle Bill Clinton's) GPS with no SA (Selective Availability) and cheap GPS mousey receivers. In Re:3. I normally do the equivalent of "ntpq -c readvar" and "ntpdc -c loopinfo" agains localhost. I check the status bits in ntpq's output and the offset in ntpdc's output. It has been serving me well over the years. $ ntpdc -c loopinfo
offset: -0.000442 s
frequency: -6.781 ppm
poll adjust: 30
watchdog timer: 28 s
$ ntpq -c readvar
status=06f4 leap_none, sync_ntp, 15 events, event_peer/strat_chg,
version="ntpd 4.1.1c-rc1@1.836 Thu Feb 13 12:17:19 EST 2003 (1)",
processor="i686", system="Linux2.4.20-6", leap=00, stratum=3,
precision=-15, rootdelay=14.035, rootdispersion=22.854, peer=43382,
refid=ntp.example.com
reftime=d217a3e7.a71fcd24 Sun, Sep 11 2011 14:17:27.652, poll=6,
clock=d217a40b.dd65b20b Sun, Sep 11 2011 14:18:03.864, state=4,
offset=-0.442, frequency=-6.781, jitter=0.756, stability=0.015
In Re: Windows. MSFT has a decent NTP implementation if nobody is tampering with it, expecially in case of domain membership. But there are so many programs that tamper with the timekeeping on Windows that it is almost impossible to trust it by default. Additionally, MSFT made some "embrace and extend" extension to NTP, they are documented on the site for European Windows developers. In Re: pool.ntp.org. Due to rampant abuse, the members of pool.ntp.org deploy various countermeasures. One very popular is to deliver time that is obviously off by months or years. You really have to be carefull with polling pool.ntp.org more often than the default 1024 seconds. |
|
|
|
I swear I have a way better one around here somewhere (runs NTP in seperate thread, takes round trip time into account, only queries ntp once per hour after initial sync, compares local, ntp and median of peer times and warns user if theres a large difference).
I don't understand one thing: why would one incorporate NTP into *coind? In any other software that I know NTP is handled by split ntpd-daemon & in-kernel-timeadj in a closed feedback loop. The net effect is that gettimeofday() (or an equivalent) returns very accurate time (fraction of millisecond when stratum > 0). Any other software can just use gettimeofday() and rely on it being very accurate and monotonic. What so special about *coind that it would require a built-in implementation of NTP? |
|
|
|
Thanks for the explanation. However,
1. Bitcoinica is not a bucket shop, it's a brokerage. All the hedging will take place if our clients place orders in a particular direction. The spread seems high now, but will be eventually lower.
2. We don't bet against our clients. We never want them to lose. There's no incentive to do that.
3. There are large scale transactions coming in Bitcoinica. First day we got 3700+ BTC, ranked number 3 among all trading platforms. Bitcoinica has more than 5% of total Bitcoin/USD trades. A bucket shop can't survive this amount of volume.
4. Because of stop orders and margin calls, we are partly responsible for the crash. Our orders are already influencing the market.
5. I'm not a Singapore citizen. I just live here. Thank you for teaching me the culture.
I have nothing to add, I just wanted to preserve the above post in a quote agains possible later editions. As the item (4) shows, we now have an official "Boy plunger of Singapore (but not a citizen of Singapore)", who had engineered one of the bigest daily percentage drops in the history of BTC vs. USD trading. |
|
|
|
There are no options, derivatives, or short selling for Bitcoin because there are no parties who can be trusted to pay up when they lose big, and no enforcement system to make them. Just getting the current parties to reliably pay up on ordinary transactions is hard enough.
In addition to the above: some potential players actually audited the source code for the Satoshi bitcoin client and know that it contains the code to reverse transactions (chain reorganization) and change transaction timestamps in direct violation of the GAAP (or their equivalent throughout the civilized world). And yet the bitcoin promoters continue to advertise that there's no chargebacks in bitcoin. |
|
|
|
Zhoutong... I'm very impressed with what you've created so quickly. People are understandably skeptical, but I'm sure you'll be proving yourself over time. Bitcoin is much stronger because there are ambitious people like you building new systems. Keep it up! If my Dwolla clears into MtGox today I might try out some of this margin stuff on your site  The "Boy plunger of Netherlands" endorses the "Boy plunger of Singapore"! I always disliked Something Awfull forum posters, but they do get at least one thing right: bitcoin community continously delivers excellent entertainment in terms of human folly, abuse, fraud and plain stupidity. |
|
|
|
What's so special about the country I'm living in? And what? "financial fraud"?
While I don't really understand what you mean, I have to say that the Bitcoin market is open for all, and new ideas are coming out every day. I urge you to look at the product itself instead of judging by other factors.
I'm not interested in unethically making other Bitcoin fans' money in this united community. What we want is to enlarge and enrich the Bitcoin market and Bitcoinica does the job of bringing in Forex traders.
Zhoutong, I truely think that you are a genius. It so just happens that you seem to have an innate ability to come up with evil schemes. I truely believe you when you are saying that you don't know what I'm talking about. Wikipedia has a good introduction to bucket shop concept and why it is considered a fraud in most of the civilized world. http://en.wikipedia.org/wiki/Bucket_shop_(stock_market) Jesse Livermore was the most famous bucket shop promoter, although he never was a bucket shop operator. In fact the actual bucket shop operators went to the extreme to avoid being associated with him. http://en.wikipedia.org/wiki/Jesse_Lauriston_LivermoreYour country is significant in my mind, because the South East Asia was/is a heaven for bucket shops. In not sure about Singapore, but Hong Kong and Macau have a long and continuous history of fly-by-night hole-in-the-wall financial "exchanges". You have extended the long tradition with your offer of accepting single USD accounts to allow wagering on bitcoin. |
|
|
|
|
This is awesome. A 17-year old in Singapore invents and implements a bitcoin bucket shop. I'm truly impressed. Imagine the financial fraud he'll invent and implement when he has at least 10 or 20 years of experience.
Ladies and Gentlemen!
We now have a new Jesse Lauriston Livermore and he lives in Singapore.
|
|
|
|
|
Yeah, you've missed chain reorganization / rollback / chargeback. Pretty much like anybody else who tries to modify the Satoshi client without understanding of the Satoshi whitepaper.
Loading a new wallet requires a rescan of the block-chain.
Theoretically, wallet.dat could be modified to store the hash coordinate of the tip of the block-chain when it gets "unloaded". Then upon "loading" instead of the full rescan the code could do a partial rescan "up the chain" (or "down and then up the chain" in case of the reorganization).
The amount of coding to do it properly is significantly nontrivial. About the same order of magnitude as developing a properly architectured split "bitcoind" & "bitcoinui".
|
|
|
|
OK, thanks, I understand now. |
|
|
|
|
OK, I'll bite again. Please don't take my reply as a personal insult.
I think you are nuts. "You" meaning "you plural", i.e. anyone who shares this idea.
"Satoshi bitcoin client" development speed had reached the stall speed. It is about to go tailspin anytime soon. Gavin is making heroic efforts to reject the patches/pulls that subtly change the behavior of the Bitcoin protocol, especially with respect to the chain reorganization.
What needs to be done is the following:
1) bitcoin needs to be refactored into two separate modules:
2) bitcoind which is a wallet-less daemon that handles the P2P network presence and caches the blockchain using some sort of general purpose database query language
3) bitcoinui which is block-chain-less wallet management utility that uses any one from the existing variety of security technologies. Some versions of it could handle people wallets in a really secure way (TPM, Sim-card, CONOPS, whatever else). Other versions wouldn't handle normal wallets but would be just interfaces to the existing mining pools.
4) the communication between the bitcoind & bitcoinui will require a known, well tested two-way communication protocol, e.g. FIX or QuickFIX FAST or anything else that gets some support in the existing securities industry and existing exchanges.
Trying to plow forward with JSON-RPC (or any other one-way protocol based on HTTP) or trying to reinvent and re-implement any of the well known in the securities industry methods of AAA (Authentication,Authorization&Accounting) is a form of slow-motion project strangulation.
Thank you for your (plural "your") attention.
|
|
|
|
|
MagicalTux had posted that they no longer use dark pools on Mt.Gox. For the very big trades they use immediate-or-cancel and iceberg orders.
|
|
|
|
|
Show Posts
