Oh boy, you still have some things you better not see...

Does anyone expect an intelligent move from the anti-dudes? They'll either convert none of the funds or way too much, possibly burning a lot in a stupid move. I think generally being against monetary systems is related to being unable to handle money, and not understanding what is going on in general, so don't expect these guys to be of much help to us. They just cause a ruckus and damages in their wild flailing and then hopefully dissipate.

What I think?

Uhm... I like jumpiness. Makes things look more animated.

Also, I don't know how to justify it, but it feels like liquidity has improved. And I don't understand why people care so much about what other people do with their orders. It's a few thousand coins jumping, wow, world's not gonna collapse.

How about renaming the thread to "Properties I can add to a stable crypto-currency to make its price diverge"?

Seriously. The point of speculation is that it corrects price. Adding an arbitrary feedback that goes against the correction -- or magnifies it, for that matter -- is a pretty absurd idea.

If the goal is to emulate a central bank, altering supply to achieve some murkily defined goal, I have a hint: don't. The last thing we need is more strange dynamics nobody understands that blow up the currency at random. I would probably have used a Bitcoin+demurrage system to keep the amount of circulating coins constant at all times, but Bitcoin is good enough. Run it for a few years, and speculators will make it increasingly stable. Voilà, there is something usable, without the threat of some failure in system dynamics suddenly causing a hyperinflation or whatever.

I agree fully that C++ is not a suitable language to maintain the main BTC client in. Yes, I was doing C++ programming in the past, and I see there is some kind of "coolness" in being the "advanced" hacker who juggles in the subtleties of template programming or the insanity of pointer magic, implicit type casts and the likes.

But this is totally out of place for Bitcoin. Aspect-oriented programming is the way to go here. We need contracts, and anything else that improves security and makes outcomes crystal-clear. This is not some toy project where one can just fix a bug after it became apparent, and the "security" offered in C++ is beyond outdated. Most people have probably never seen a competition on harmless-looking backdoor programming. It's just marvelous what possibilities C alone offers in terms of "oops, this pointer is going somewhere else now". Attacks on Bitcoin will be a whole different level than what anyone has faced in the past.

That said, it is a very good thing to have multiple clients. If one gets wiped out, then maybe Bitcoin survives if enough others remain (or were offline at the point of the attack).

I'm still watching this, but still not satisfied. I've just been shown a phone on which the wallet shows inconsistent data: the transactions do not sum to the balance. What the heck? In addition, the balance is wrong, it does not agree with blockexplorer and shows coins that supposedly appeared out of nowhere.

It is usually not possible to buxfix quality into a program, and if this is done in a cryptographic context, the result is probably not secure. How does that programmer's saying go... "Do it. Do it right. Do it right now." I really like the UI, and the back-end (BitcoinJ, I assume?) does not seem to eat coins all too much, but whatever does the processing in between really needs an overhaul.

A lot of the output I saw was more than one error: the error itself, and the fact that the program doesn't output something like "Sanity check failed. Send a report to the developer, backup the last valid wallet data?"

That's the true Bitcoiners, estimating a possible >130% rise and calling it a "minimal upside".

I lol'd.

Wow, this is quite the bear party. A drop below 1 would be about 65% down again, that's quite heavy even for Bitcoin.

I like how the market is more liquid now, with short-selling and all that going on. And how there are many bull-bear hybrids now, who don't only argue one way. I'm not selling any now. It's less likely the trend stops right away, but if it does, there's quite a possible upside.

I don't really know whether I want it to go up or down right away. It's an immediate loss if it goes down, but then again, I do have some more money I'd spend on cheap coins.

The ability to transfer funds worldwide without major timing, fee or regulation issues is a killer feature. Combined with the ability to store value, Bitcoin is incredibly awesome for pretty much... everyone.

Everybody who sells software online, virtually the entire financial sector, and websites with donations, hell most of the internet longs for something like Bitcoin. Silk Road is nothing more than a little media hype that becomes meaningless if Bitcoin ever gets serious.

Just wanted to inform: 20 hours later, my withdrawal was broadcast. All the coins are visible again.

I guess it just needs a little patience. I still wonder what causes this though.

Over 12 hours now, no broadcast. Hundreds of coins have currently disappeared for me, this is not something I like to see.

Has anyone had a successful withdrawal lately? What is happening at Gox this time? With a hack once and burned coins another time, silence on the issue is unnerving.

*Sighs*

If what you say is true, the correct tactic is to delay the rally and secretly buy as much as possible at low prices.

Why does everybody believe some idiotic buy-high-sell-higher rampage manipulator would be so very effective? You always try to buy low, a rally *while* you are buying is bad, it should happen only afterwards. Otherwise, how does "the manipulator" (I'm so tired of that nonsense) get his volume through?



I want to commend your tactic. It is one you can't regret, but can make money with. If more people would think that way, Bitcoin would gain support without high risk for anyone! Your rough estimates are correct, you're certainly on the good side of Kelly's Criterion, which means not betting too much of your money, and on average, that tactic gives good returns. Though waiting for 1MUSD per coin is probably overdoing it.

As things are, I recommend that people who don't feel to do serious speculating still buy 20 BTC or so. It's not something to cry about if they're lost, the innovative project is supported, and it gives them good profit if Bitcoin happens to survive.

I'm more of a physicist, but I did think about a trust-based distributed payment system before Bitcoin came up. I stopped doing that, since I think it's better to unite on one crypto-currency than to fragment the market right now.


I don't really want to avoid round numbers, I only do it because it helps statistically when everybody else prefers them for no reason.

I believe there's a fundamental error in how your proposal treats likely prices. Just because the price is more likely to go to a certain position, it is not necessarily the best idea to bet this way. To give an extreme example, let's assume the price will drop by 0.10 USD with 80% probability, and rise by 0.90 USD with 20% probability. Both reach the same "magic value" behind the comma, whichever it is. But the expectation value of profit on holding a Bitcoin would be +0.10 USD! The correct decision would be to buy, which pushes away from the magic number.

Now, if you approach the magic number very closely, such a scenario becomes more and more likely. "Real price", as in the natural one, is decreasingly likely to be within a small corridor produced by people trying to enforce a certain number.

In short, just because something is likely or usually easy to maintain, it is not already a good speculation method. Speculators are there to determine the real price, not try to enforce their own.

I ordered a withdrawal multiple blocks ago that wasn't broadcast yet. Very unusual, why delay this? The only reason I can think of is manual security at mass withdrawals. But it's day in Japan, so what should take time there?

I always wanted to see some of the major players out themselves. Pleased to read from you, old_engineer.

Your trading technique is almost disturbingly similar to mine. I'm not that big, <1% of volume the last 30 days. But I could quote you on so many things. Not using a bot, mostly placing small orders to be executed on price swings, heck, "friction on price" are words I use exactly!

The only thing we disagree on is the idea of setting the price to a round number. I think this is a systematic mistake, so when I have a hunch which way the price *should* be, I prefer pushing away from round numbers, even if it appears risky on the short term. It's nonsensical to expect real BTC price to just happen to be 3.00. Within 30 cent of it, well, could be, but not exactly the magic value. PatrickHarnett's proposal of setting it to a value nobody knows won't help against the fundamental problem: we don't know the real value, and it might change with time.

Whoa, that's not what I meant. First of all, I compare nothing of NASDAQ after 2003, only the last big rally and the time where the market is not in "bubble destruction mode". Also, the comparison needs multiple corrections. I wouldn't expect Bitcoin to be as close as the NASDAQ and DJIA bubble were between one another, simply because the timings in the digital age are heavily distorted, Bitcoin is much smaller, and even with that said, everything happened at blazing speed in the first half of this year.

Timing issues I expect:

• Bitcoin is illiquid. Times in which the price appears very stuck, especially until a real bust started in August, should not be considered much in comparison with the much more liquid NASDAQ.
• Similarly, the MtGox hack is something one cannot just throw in there. It probably delayed the process, just cutting it in as natural is most likely not a valid approach.
• These saw-tooth like down-up spikes are distorted because of Bitcoin's extreme timing. What is important is that a similar pattern repeats a few times, and includes a characteristic drop, turnaround, and vain rally multiple times and on similar time-scales.
• I personally would not match the "bursting" phase of initial sell-offs with the same scale factor as that of the following "bust" phase, where price is repeatedly lowered in spiky price curves. They have different psychological components that might have changed separately: during the "bursting", the inflow of people and their reaction time to a changed scenario is important, which is much faster in an internet-driven community. But in the bust phase, panic thresholds and the time people take to change their minds is important, which has not changed as much. Thus, in Bitcoin, the bust phase appears longer compared to the initial bursting.

The point of this wasn't to become an exact price prophet, but rather to make qualitative forecasts on price. After the superexponential breakout, prices were very likely to drop, and after that had happened, the following rallies ended in that typical bust-pattern, each attempt to return driven down by the great bears until all denial was destroyed.

It's cheating to match the graphs in a non-trivial way in retrospect, so I won't. Also, it's quite extreme to intend predicting an exact price curve. This analogy was good enough to make predictions on expectation value, that is, produce money on average. In my case, it saved me from massive losses, and gained me a little in day-trading. I know it working once doesn't prove a thing, but once you "get" what the analogy is about, it is quite striking IMO. I wanted to test it again on silver -- I was quite frustrated my order on put options wasn't executed, just before the 40-30 drop.

Anyways, I'll shut up on this now, as it's a thing of the past now. Bitcoin "bubble mode" might have ended already. At least I can't use the analogy to predict when, or if, the trend stops now. All I want to say is: I think this was more than a coincidental similarity.

I never understood why deflation is supposed to be so horrible. We have massive other problems with money, what's the big deal in some initial liquidity troubles?

Initial distribution is largely irrelevant; someone hoarding from the start can sell his coins only once. So it will happen a few times, lol volatility. Life goes on. Speculators will have to take this into account and do some calculating, then the problem will fade as someone figures it out. A bit of the money will go to these speculators. Oh well.

What's the big mythical problem here? The problems with international payment are huge, it's nowhere near the scale of some money supply trouble.

What is more, it's all out in the open. Once people know how the market behaves, rules won't suddenly change as they did in the Euro in the last years. This is very, very, extremely important! I can live with quite a few issues if I get clear and stable rules in return. The deflation issue just isn't that important.

Oops, misread, "or maybe yesterday" made me click yes before reading all options.

To escape the dilemma, I bought a few more. Now I'm certain my answer is correct.

Seriously, it's all a matter of infrastructure.

Ever heard of The Cayman Islands, Jersey, Gibraltar, Liechtenstein?

The amount of money waiting for a system like Bitcoin to work is probably of a magnitude around 10^10 ~ 10^11 USD. A trade balance of 10 GUSD (low estimate) would amount to an average 476 USD traded per Bitcoin, which does not take future expectations, hoarding, lost coins, inflation and unfinished minting into account.

In a full-success scenario, 5k USD/BTC is possible and not unlikely. Why do you guys think I shrug when people keep flaming me with how I'm burning money if this fails? If I was a real investor, and not just a newcomer wannabe with play money, I'd be busy buying now.

If a certainly surviving Bitcoin is worth 300 USD, which is conservative, the market currently claims that Bitcoin will fail with over 99% probability! This is absurd!

I like the idea, but it's not fun to watch because it doesn't really move much. At the very least, the second hand should move once a block.

Personally, I would add something even faster than a block to make movement smooth. For example the integrated probability that a block would have come since the last one, indicating whether it's coming late. Doesn't have to be on the clock disc, just add a bar or anything neat.

That would be the kind of thing to stare at when waiting for a transaction to confirm. There are lots of other cool things one could add, like color-filling the clock's background with the amount of BTC minted over the total amount ever.

I'm still feeling quite good, thanks for asking. Yay Bitcoin!

If you want to see me in a less positive mood, wait for another >50% down. But unless Bitcoin really looks in danger of dying, I'm not so much shaken. Bubbles don't have soft landings, doesn't everyone know that? Relax 'n watch the touchdown.

Picking a better topic text would improve the thread btw.