Maybe I'm not getting your proposed algorithm. Letting Ti = time between block i-1 and block i, the current algorithm is (disregarding the off-by-one bug):
What I think you're suggesting:
Is that right? If so, can you explain what c is, and how does this work? If not, what is it?

This absolutely cannot work. The harmonic expectation of the exponential distribution is 0, so there is no way to retarget using it.

Even if you use a geometric mean or something else that does converge, it will have very high variance causing retargets to be all over the place. Not only that, but an attacker could release blocks with reported timestamp a second after the previous ones, causing a massive drop in the geometric mean and a spike in the difficulty.

Also, messing with the timestamp of the last block may have some short-term effect, but not so much on the long-term because of the rules for validity of timestamps.

This isn't to say I think the retargeting algorithm is perfect. It is basically a P controller, where PID may have been better.

You do this using vouchers. You withdraw USD as a redeemable code, and send the code to the recipient.

They used to have an option to transfer to a member by his email address, but it was problematic.

After not receiving a response from HowToVanish, I contacted RunToGold. They gave me a direct link to the download and said they corrected the mistake.

There are no doubt many things I should have done differently. As I said this is still a proof of concept, the only thing I am committing to come hell or high water is to abide by the contract, anything else is prone to my newbie mistakes. I will appreciate if you bear with me and let me learn from my mistakes - and compensate for any losses caused by them - instead of writing me or the asset off.

So let's look at some of my recent decisions.

1. Offering at an initial price of 0.5 BTC.
This is arguably too high. As I said this seemed to me a fair price considering the alternatives, but I may have been mistaken in my evaluation.

2. Lowering to 0.4 BTC.
This seems like a reasonable thing to do given that there was little demand at the higher price. Please tell me if you think this is wrong.

3. Not offering a compensation before lowering the price.
No excuses, I just didn't think of it at the time.

4. Offering compensation after the fact.
I didn't think of the possibility that early adopters would have sold, and it still makes little sense to me. If you bought it at 0.5 BTC, it means that you value a guarantee to be paid the equivalent of 1 MH/s as more than 0.5 BTC. Why would you want to shortly after sell it at 0.4 BTC?
Under this assumption, compensation after the fact would have been more or less equivalent to compensation in advance. And as I said, I think some form of compensation is fair because early adopters shouldn't be penalized for my overpricing.

5. Not trying to find early adopters and paying them directly.
That would have been cumbersome and there's no way for them to prove they were indeed early adopters.

6. Not announcing the compensation in advance.
That would have just invited freeriders who would then buy the shares and receive compensation, for an effective price of 0.3 BTC.


I would like to clarify that I am not committing to anything but what is stated in the contract, and potential buyers should base their decision first and foremost on the contract. The future traded price at GLBSE, which could be affected by my decisions, should be secondary. But if anyone else feels they have been wronged by my decisions, please let me know and I will try to set things straight.

Also be aware that it is entirely possible that the issuing price will decrease further in the future, and that I will handle it better that time (again, I'm not guaranteeing that there will be compensation to current bondholders).

To avoid any confusion, I do not currently have a policy of making a post about every coupon paid, only about any news and such. Scheduled payment history is available at http://bitcoinpuremining.com/ ; blocks marked as paid are indeed paid, otherwise they are listed as "Pending" (in theory a technical error could cause wrong stats to be displayed, but I make an effort to maintain accuracy). So far coupons have been paid for blocks 168335, 168671 and 169007.

I've paid a bonus 0.1 BTC per bond. Because of the recent price drop from 0.5 BTC to 0.4 BTC, I don't want early adopters to be penalized for their support. This guarantees that people who bought at 0.5 BTC are not worse off than people who buy now.

Note that I reserve the right to change the price at which bonds are offered without any compensation given to existing bondholders.

A similarly easy, and much more effective solution is to limit withdrawal rate per person. For example you could say withdrawals are limited to 25 BTC per day. So someone who wants to deposit 500 BTC will be accepted with open arms, but will do so with the understanding that it may take up to 20 days to withdraw everything.

Also a good idea. But its main advantage, the liquidity provided by GLBSE, might only be realized when GLBSE gains in popularity (which might be brought about by GLBSE 2.0), and with relatively large lenders. Otherwise, I can see how someone wanting to exit the investment by selling on GLBSE would suffer significant slippage. (Bidding on one's own bonds could mitigate this, but is not easier than having reserves from which to withdraw CDs.)

Not a fan. It's much easier to trust one's honesty in honoring the terms of a bond, than his competence in running a successful lending business.

In particular:
1. "Bitcoin" - capitalized and singular - is the name of a protocol, the concept of using that protocol for a decentralized digital currency, a project to bring about this currency, a software to handle this currency, and an open source project to develop this software.
2. A "bitcoin" is a denominational unit of currency in Bitcoin. So you can have one bitcoin or 3 bitcoins. When used at the start of a sentence, and only then, it is capitalized.
3. There is no such thing as BitCoin, Bit Coin, Bit Coins, bit coin or bit coins.

Did you try deleting everything except wallet.dat in the Bitcoin Application Data directory?

If you need to verify that payment was sent but don't need to immediately use the funds, you can use blockexplorer.com or blockchain.info.

You should consider contacting Amir Taaki (aka genjix) who lives in London and, among other things, is a core Bitcoin developer and operates an exchange; and Vladimir Marchenko, who lives in Sheffield and last I heard, owned a large contract-mining farm.

There's a lot to be desired in these two areas but I think the current options are good enough, I don't think that's what is holding Bitcoin back.

The biggest problem is AML (anti money laundering) laws which attempt to put a handle on activities that could be an opening for money laundering. At best, the licensing and reporting required to comply with such laws has costs involved and hassles customers. I don't know what specific problems Tradehill have been having.

Odd, blockchain.info doesn't work for me with Firefox 3.6.27, but it does work on Chrome. So you can also verify with http://blockchain.info/address/1KpoMx7ZcR45qMo1Uf3iQ8fsurhgtnQfcK or http://blockchain.info/block-index/183665/00000000000009ab6e1c90b41cae85772383f7b4a05076c5e8195ff190406d35 if your browser supports it. (EDIT: Apparently everything works on Firefox too, except for the homepage).

fuktbyth, can you check how many blocks your client has downloaded (hover over the icon in the bottom right) and how many connections you have (the signal icon next to it)? If you have less than 168451 blocks and no connections, can you restart the OS?

When you do get your bitcoins back, if you're looking for a way to exchange them for USD which will work with such a small amount, you live in the US and can accept Amazon payments, I can recommend http://coincard.ndrix.com/.

You were unlucky enough to start using Tradehill when they were forced to shut down due to circumstances beyond their control. In the "non-Bitcoin world" businesses acting in good faith close down all the time, and often customers suffer from it, in amounts much more than $10. I understand that losing everything that you had to do with Bitcoin must hurt, but that is a statement about you, not about Tradehill or Bitcoin.

I don't believe Fussy Prick is lying about having sent. I know it looks bad, but I examined it some more and noticed that blockexplorer.com only reports blocks up to 168424, while my own client reports knowing 168543 blocks. blockchain.info reported something about re-importing the block chain and are now down completely. I think this might be some sort of forking or incompatibility caused by architecture changes scheduled for recently. There ought to be posts on the forum about similar phenomena which are worth looking for, but I suggest everyone should update their client to the most recent version and then just wait.

You need to distinguish score (which in some other places I call "units") from reward. Score is s=1/D and is the thing that should total X. (from the OP, "a score of 1/D... until you reach a share which brings the total score of the shares counted above X".) Reward is what you pay for each included share, and is s*B/X.

I hope there are some pieces to the computation you're not showing here, because some things are missing with respect to the accurate variant. In particular, you need D and B to be the values at the time the share was submitted rather than at the time payments are handed out (the OP clarifies this for D, but was written before I realized this is also true for B), and for absolute accuracy you need to proportionally discount the reward for the earliest share (the one that brings the total score to X).

WTF. The Tradehill people are outstanding members of the Bitcoin community who have for 8 months provided a valuable service, fighting various hardships and losing tons of money. Even now they are staying in the game and might one day resume operations of the exchange.

There's no sense being worked up about some loose change that is worth less than what it would cost to transfer it. I have some $7 left there too and I don't at all mind it going into the development of bitcoin.com.

I completely sympathize with wanting to start small with Bitcoin and being upset when things seem not to work out. I too had an unpleasant experience when I started out (didn't have any funds vanish, but couldn't withdraw USD as easily as I'd hoped). In the great scheme of things these really are drops of bad in a bucket of good, you only need to realize that Bitcoin has a long, bumpy ride ahead of it.

Anyway, if you've already downloaded the blockchain you should see the transfer immediately, maybe you're not connected to the network.

Edit: Maybe the sender is disconnected from the network, I don't see a transaction to 1KpoMx7ZcR45qMo1Uf3iQ8fsurhgtnQfcK in either a block or the list of floating transactions.

A simplified model gives a result of 5 hours, but finding a more accurate estimate requires some work.

Yes, higher hashrate means the attacker can close the gap faster. With 51% or 50.001% he is guaranteed to outpace the network from any disadvantage (say 6 blocks), eventually. The guarantee doesn't tell you how long it will take, and in fact with 50.001% it would take on average about 6 years (it's random so it could be much less). With 51% it's 2 days. (These estimates are with a very simplified model, actual numbers could be different).

Yes, he has a chance of success but no guarantee.

First coupon, for block 168335, has been paid. Should have been 0.0015883564 BTC per bond, except I misunderstood how payment works in GLBSE, so bondholders can enjoy the larger amount of 0.01443959 BTC per bond.

In other news, bonds are now offered at 0.4 BTC per MH/s.

It's worth being careful with your logic when any mistake can drastically alter the conclusions you make.

It's not a secret that the LW community leans towards the asperger and nerd clusters, but it is markedly much more grounded than you'd expect from the stereotype, not confusing the everyday reality with highly simplified idealized models of it.

This is an excellent example of how you've completely missed the point of LessWrong. Eliezer isn't criticizing the art - art exists to entertain, and if it does that that's fine. He's criticizing the thought processes that would lead us to think this is a plausible scenario - thought processes which are harmless when applied to SF, but which could be very dangerous when applied to some real situations.

No, in fact common LessWrong opinion objects to the notion that reality exists in the mind. It is its perception that exists in the mind.

What you call pedantry, I call being careful with ideas, where mistakes can aggregate and make you reach very wrong conclusions.

That is an important observation in many LW analyses, but is a corollary, not a core principle, of rationality.

So it's better to say human systems are magical and mysterious and nobody could ever understand them?

Apologies for practicing the dark arts of necromancy, but this had to be said.

Let's just put it this way. A person who reads LessWrong with some regularity is not the same as a person who doesn't. There really isn't any other site like it.

Anecdotally, LessWrong is how I learned about Bitcoin, and Wei Dai, whose b-money proposal was an inspiration to Bitcoin, is a top contributor.