This is not possible in the standard client.  You need this patch: https://bitcointalk.org/index.php?topic=24784.0;all


Again, not in the standard client.  It just applies the algorithm to determine your fee, but it doesn't give any explanation of why.  It'd be nice if it could at least spit out a screen of the relevant variables to explain the fees.  I don't know if there's a patch for this.


It should be possible with the patch above.  You send your full consolidation transaction using only the coins you choose.  If it fails, you subtract the fee then try again.


I don't see any reference to from-address in the API help.  You can do a "sendfrom <fromaccount> ...", but note that accounts are not the same as addresses - accounts are just which internal ledger will be debited for the transaction, and it does not affect which address is used externally to actually send the coins.

I'm pretty sure all the RTFM in the world won't save you here.  I Rd a lot of FM (and the source) and concluded it was simply not possible in the 0.3.24 API.  0.4.0 will be the same unless they pull the coderrr patch.

Related.  Where's George bills get the URL written on them so that future holders will visit the site.  It's a small marking done on a relatively large scale, and their site gives you a pretty good sense of how long such lightly-modified bills stay in circulation and how often someone cares enough to actually pay attention to something written on their money.

This works great as long as you have someone with $40M who's willing to burn a good chunk of it solving this problem.  They will be operating at a loss since this goes against market forces.  I don't have that kind of change in my altruism budget.  Do you? 

The objection that this doesn't reward miners enough is unusual!  This scheme would take most of the value growth and give it to miners instead of currency holders/speculators.  On average it's much more profitable for miners than the current system...  But it would certainly have greater variance.

Encoin's another interesting take on it...  Sort of a different approach at the same idea, just with some different tradeoffs.  It touches a lot more things... I'm interested to see how it pans out.

No, I know my method isn't yet baked to perfection.. If nothing else it will require a lot of tuning to get a set of constants that result in a stable market.  I still think it's an interesting mechanism that altcoiners should consider.  Even if it's never adopted I hope it'll inspire some more innovative ways to stabilize price.

7200, give or take a bit.  The difficulty readjusts every two weeks or so to bring it back to 7200 whenever it starts going too fast or too slow.

Network speed and difficulty charts: http://bitcoin.sipa.be/

The CPU is to verify all the transactions as it downloads them.

Once you've caught up to the whole block chain, future starts will only need to download and verify new blocks.  If you have to set up many wallets, you can copy the block chain files (blkxxxx.dat and blkindex.dat) from one to the others so you don't have to repeat the process for each.

Also, you can download a copy of the block chain here: http://eu1.bitcoincharts.com/blockchain/ .  That's much faster, but your client won't verify it... Depends if you're paranoid.

What block number does it say on the bottom of the window?  It may still be processing the block chain history, which requires a lot of CPU.

It varies with the difficulty level.  Right now you'll get about 0.3 per day.

"killall -w bitcoin", then try running it again.

Miners will always earn transaction fees, so it will never hit zero.  We can also set a minimum reward to guarantee security during an uncontrollable crash.

In Bitcoin, the 50BTC per block only lasts another year or so; then your earnings are cut in half.  It keeps decreasing in the future.  Will Bitcoin still be secure once we're past the initial currency distribution phase?


That's included, but I didn't want to go into it here since it's a pretty complicated subject and better suited to the Alternate Cryptocurrencies section.  I'm trying to stay somewhat on the "stability" topic. 


You can't make major changes just because we don't like it - the fork wouldn't be followed.  This option only works if the system is clearly broken and a fix is widely agreed to be necessary - in which case it will be quickly adopted.  Again, have a look at Bitcoin for precedent: with the rise in price, 0.01 transaction fees were becoming a burden; Bitcoin 3.23 changed it to 0.0005.  Or look at the plan if SHA256 is broken: cut off block chain at whichever block the developers consider to be the last before the break, change to a new hashing algorithm, then resume the network with the new client.

That's just the nature of distributed networks.  If you can convince a majority of the network that your change is necessary, it's adopted.  If you can't, they go on without you. 

In any case, the fourth "nuclear option" safety is only invoked if we fail at getting the design right.  The better solution is carefully tuning the algorithm so it can't be gamed in the first place.

Also: the dynamic inflation is only intended to get us through the initial growth phase.  If this works as intended, that value growth will be much more even than Bitcoin's price chart.  If it takes off as a currency there will be a very large number of coins available to miners for a long time; then the goal is to gradually weaken the stabilization as the market cap grows.  If it ever becomes so large that it becomes a reserve currency it will end up completely floating; the end game would either be a fixed money supply (all mining profits from fees) or a small, fixed rate of inflation.  (There are other options, but again it's well beyond the scope of this thread.)

Correct.  That's one reason that the inflation rate algorithm has to respond slowly and smoothly.  Gradual changes don't break the network - we've seen a 10x devaluation of Bitcoin over the last 4 months, greatly cutting into mining profits, but the difficulty is able to drop fast enough to compensate.



That's why the relaying nodes get a veto: they drop dishonest blocks.  Miners don't want their blocks to be orphaned, so they have incentive to mine blocks that get the best propagation possible.  In fact, getting price quotes from several sources and averaging them will lower their orphaned block rate by reducing the chance that a relay node using a different quote source will drop their block.

It's possible the relay network could be subverted by a large number of colluding relay nodes.  Thus there's a second check:  future miners refuse to include dishonest blocks in newly mined blocks.  If you want your generated coins to mature, your blocks have to be legitimate.

To prevent a 51% collusion attack, a third safety is possible: create a "decentral bank".  This would be a second mining system where people perform proofs-of-work solely to manage monetary policy, which means authenticating price quotes and the inflation rate.

The fourth safety is the strongest: anyone can verify that the price quotes in the block chain are correct, and the inflation rate responds very slowly and smoothly.   If it becomes apparent that the system is being gamed, we'll have time to figure out why and how before coins start being generated excessively. We then release a new version that adjusts whatever is necessary to close the loophole.  In the worst case, we revert to 50BTC per block.

The target block rate wouldn't change.  Only the number of coins generated per block would change.

Mining is a great technique for decentralized consensus.  My idea is to mine the exchange rate into the block chain.  Miners can use any source they want.  The rest of the network would evaluate the price quote.  A little tolerance would be allowed due to different data sources, but blocks with obviously wrong quotes would not be relayed, and the next miner would fork the chain from before the block with the lie.

I assume you mean devaluation.

Short answer: Same as Bitcoin: you don't.    You can't prevent devaluation of an unbacked currency if everyone decides to abandon it.

Longer answer: I would use a feedback mechanism to detect large devaluations and temporarily reset the target exchange rate to a lower level during the rebound.  This is to prevent losing control of the speculation-damping mechanism.  So, after a crash, it will gradually rise back to the target exchange rate, instead of having a sharp rebound.  (This is a bit oversimplified.  Instead of switching to a "recovery algorithm", it would just be implemented as a weighting factor in the regular inflation-rate algorithm.)

This would actually deal with the current situation better than Bitcoin: during rapid devaluation, it would greatly reduce the generated coins to much lower than 50BTC per block, thus reducing the devaluation rate.  It wouldn't prevent a blowoff as large as we've had the last few months, but the goal is to never have a bubble that big in the first place.

Overall, this would be a very heavily-smoothed, slow-responding mechanism.  This is not meant to provide short term stability or hold a specific exchange rate.  It is only mean to respond to long term trends, to prevent the get-rich-quick problems.

Not really...  Futures and options help protect you against price instability, but they don't actually improve (and can harm) the stability of the overall market.



+1.  I think this is a design flaw in Bitcoin's economics.

My proposed solution is to use dynamic inflation levels (by increasing coins generated in mining) to prevent excessive price growth against a currency basket until there's enough stored value to provide inherent stability.  This would eliminate: 1) wild speculation in the form of "you need to buy as many coins as you can NOW to own a slice of the world's future currency!"; 2) the "early adopters are unfairly benefiting from the pyramid" problem.

Speculators don't like it because my altcoin is only mildly deflationary during the growth phase (as opposed to Bitcoin's highly deflationary nature).  However, I think it will allow a much more stable value store - by undermining speculative bubbles, you don't have to worry about buying in at the top and getting screwed like the guys who bought BTC at $30.

I agree with all of this except the word "speculation".  We need market makers, not speculators.  Market makers and arbitrageurs create market depth and liquidity, but neither perform speculation, and they make money whether the market goes up or down.

Indeed, speculation does not create market depth - speculators buy up (or sell) inventory and then hold that position until they capitalize or capitulate.  They do not put it back on the market as they are speculating that it will be more valuable in the future.  They perform an important aspect of price discovery (pricing in the future), but right now it's dominated by people who don't seem to be well informed.


There are plenty of coins in circulation to handle all current business at $1/BTC.  With only 2% of coins on the market (and matching fiat) you'll have over USD$1M of depth.  That's enough to absorb a few days worth of all personal transactions that are occurring without moving the price.  It's not enough to buy a house or do commercial business, but we're not there yet.

The current depth is 0.05%.  That is directly from excess speculation.

Raising the price does not create more fiat depth, and reduces BTC depth.  Right now there are only USD$100k of bids on MtGox.  If you raise the price to $50, there will still only be $100k of bids... Just at a higher price, and for fewer coins.

At some point a higher price is justified to allow more depth and value store, but for current use, we're still order of magnitude past where it needs to be.


It's a lot easier to regulate centralized markets like commodities exchanges.  Bitcoin's decentralized and pseudonymous nature makes regulation much less effective.  Just like the internet, they can regulate all they want, and it will react in similar ways: moving prohibited actions to another jurisdiction; increased use of anonymity; widespread disregard for the law.

Take a look at copyrights.  There is widespread cross-jurisdictional agreement on the terms of what's covered.  There are fairly strong laws to enforce it, and occasionally they bust a few people.  So what do the pirates do?  Build decentralized networks and move all their coordination to more friendly jurisdictions.

That's not to say the regulation will have no effect.  Without regulation copyrights wouldn't exist at all, and content sharing would be much more widespread.  It will just be much harder to do than the SEC's usual fare.

I don't know a site that gives you a chart of exactly that, but:

http://bitcoin.sipa.be/ - (hashrate/difficulty) is proportional to block rate.
http://blockchain.info/ - you can go back however many blocks you want and compute the rate yourself.

Yes, it'll have an effect on the price.  An extra 2400 BTC supply dumped per day will move the market.  The effect should be small but measurable.

Let me qualify my argument a bit:

A low price doesn't provide stability, and indeed reduces it - I agree with you there.  However, a low price indicates that the destabilizing jackoffs have left the market.  For that, I believe we will therefore see stability correlated with a lower price.

Of course, even greater stability results from a high price driven by commerce.



Absolutely.  This is preventing me from doing my business in BTC today.  I'll try to switch to BTC as soon as the market depth supports it, but we're a long way from there.



Artificially inflating the price to $30 makes it a joke.  I don't want to stay at $2 forever.  I just want it to grow with real commerce instead of crazy speculation.

With a million dollars I can manipulate the price up 5000% today.  I wouldn't be able to hold it there very long since I'd be bleeding money fast, but I'd probably get a couple hours before someone with a large enough wallet came along to sell through me.

Price manipulation potential is not a simple ratio of your cash on hand to the market cap.

That's correct - a low market cap can be cheaply manipulated.  However, the high market cap simply means that we have a very large hoarde of speculative investors who're pushing the price around without regard to fundamentals.  A low market cap means that the hoarde has left, and the price is beginning to be controlled by commerce again.  Pushing the price around by large percentages against a heavier fundamental anchor just increases your losses in doing so - I'll remind you that cornering a market is a fool's errand.  That's where the stability comes from.

Real stability will come from increased commerce which will result in a high enough market cap to make the price harder to push - stability from both sides.