Goooo AnCap!
If Somalia were AnCap, you'd have a point. But it's not, now is it? Away from the cities (where state influence was the worst, and where violence is now the worst), Somalia is primarily Anarcho-communist. Oops! You just proved my point. Check out this research: http://www.peterleeson.com/Better_Off_Stateless.pdfLOL! I think you should write your asshat libertarian friend with the news that Somalia is AnCom. His paper will need to be rewritten with fresh fabricated data. |
|
|
|
(this all assumes that the guy who said this problem wasnt self adjusting was correct i did get two different contradicting responses)
I believe that when jgarzik said: It is economically self-adjusting.
He was saying the same thing that I was. "Economically" self adjusting in that there would eventually be an equilibrium between the fees that people would be willing to pay and the number of transactions in a block. Listen, this is not rocket science. Sure, there could be a positive equilibrium fee. But the equilibrium is chosen based on throwing darts and it is quite likely to produce a shit turd equilibrium. Worse yet, because of the rigidity involved in making any change to the rules, we could end up mired in shit turds for the indefinite future. The moniker Buttcoin could prove quite apt. So if there was no limit and the blocks were no longer rewarding miners, the market should arrive at a natural equilibrium that is a tranaction fee which is just enough to incentivize miners to include your transaction in the block. this same minimum transaction fee should prevent the government or large corporations from using micro transactions to bloat the blockchain. It seems like satoshi should have coded maximum block size to double at the same time as block reward halving. No, if there was no limit, the market would arrive at a txn fee of approximately zero (just enough to verify the ECSDA signature and transmit the block to peers). (To see this, set the limit to an arbitrarily large number in the previous example. The txn fee will be negative, but that is not feasible so it will stop decreasing when it hits 0.) This could actually manages to do worse then the current plan and much worse than monopoly. The outcome assures that the blockchain open for rape by any common vandal. At least the other plan has a chance of succeeding due to dumb luck. Surprisingly, this is actually another one of the developers' plans. Support the blockchain using 'donations'. LOL! You can start by donating to miners today! I'm sure this will top everyone's list of charitable causes. The only realistic way to get an acceptable solution to a problem like this is through voting. Probably coin-owners should suggest a fee. And the fee proposed by the median coin should be selected. They could also vote on a block size limit. As far as economics is concerned the two are equivalent. In the blockchain, you could probably instrument this by including votes in each txn and weighting them by output. Probably a vote to move the current fee up by 0.1% or down by 0.1% is sufficient. i.e. this is just one byte of extra info per txn. You could even encode it in a pre-existing piece of information, like include 5 satoshis in the fee to vote up, 6 satoshis in the fee to vote down, and anything else to abstain. Algorithmic solutions are hopelessly complicated and error prone. I'm an economist and do this kind of stuff for a living. I wouldn't trust an economist to design an algorithm to do this and not fuck up; let alone a computer scientist. The only reasonable thing an algorithm could target without a reasonable chance of seriously fucking up is the monopoly solution (total fee revenue maximization). I would say this is a second best option to actual voting. Finally, recall that proof-of-stake solves this problem entirely without meaningful fees. There would still be a fee to vote on, but you could vote for an extremely small fee just sufficient to deter spam. |
|
|
|
It's fundamentalism that's moronic, independently of the flavor  Amen! |
|
|
|
(this all assumes that the guy who said this problem wasnt self adjusting was correct i did get two different contradicting responses)
I believe that when jgarzik said: It is economically self-adjusting.
He was saying the same thing that I was. "Economically" self adjusting in that there would eventually be an equilibrium between the fees that people would be willing to pay and the number of transactions in a block. Listen, this is not rocket science. Sure, there could be a positive equilibrium fee. But the equilibrium is chosen based on throwing darts and it is quite likely to produce a shit turd equilibrium. Worse yet, because of the rigidity involved in making any change to the rules, we could end up mired in shit turds for the indefinite future. The moniker Buttcoin could prove quite apt. Q_D(p)=a-bp Txn demanded per block (Q_D) is a function of p, the price per txn. I'm assuming it is linear here just for simplicity. Q_S=min{a,4000)} 4000 is just the number of txns that can fit with the current block size limit. If a<4000, then the fee is approximately 0. This is the current equilibrium with 0 or minimal fees. If a>4000, then there will be equilibrium fees. You can just set Q_D=Q_S and solve the equation to get equilibrium fees. 4000=a-bp p*=(a-4000)/b That is the fee with a block size limit. It goes up as demand increases (a). It will go down if substitutes for bitcoin are developed (this would increase b). Consider the monopoly situation, where the monopoly picks block size. In this case, the monopoly maximizes total feel revenue (ignoring costs because those are hard to predict right now). That is p*q, or ap-bp^2. The solution is just given by a=2bp, or p=a/2b If a/2b<(a-4000)/b, then the monopoly (like Paypal), leads to lower fees than the current plan. Rearranging. This is a>8000 Essentially this says that if people demand more than 2 MB blocks at a txn fee of 0, then monopoly leads to lower fees than a block size cap of 1 MB. The whole example is ad hoc because I assumed a linear demand curve. Nevertheless, the example illustrates what utter and complete garbage a block size cap is. Usually we try to do better than monopoly. Here it seems quite plausible, even likely, that we will do worse on all counts. There are many other solutions which make more sense. Many of these alternatives are also bad and arbitrary. However, the solution at hand is really fucking extraordinarily bad and really fucking extraordinarily arbitrary. It can simultaneously fail on both security grounds and txns cost grounds. If the cap results in a functional outcome, you can thank dumb luck because I'm sure as hell that no thought process went into this. In short, the current plan is utter and complete fail. Anyone who says otherwise is completely full of shit and cannot be competent to make this type of decision (i.e. if they were competent, they never would have suggested anything remotely as stupid as this.). The developers know this is utter and complete shit. They suggest otherwise when speaking in public to avoid damage to confidence. Of course, if they wanted to solve the problem, they would say: "Hey we have a problem, does anyone have any ideas?" |
|
|
|
yea i already figured that one out myself from what i understand about economics but this is not a very good solution once it moves beyond cleaning up the dust problem.
You are correct. This is a ridiculously bad solution. It could lead to ass-raping fees that dwarf paypal. However, that's the plan. Proof of stake solves the problem completely, but that's off the table for BitPal. |
|
|
|
Okay now we're messed.
Would a more logical answer be that if major crisis like that happens (transactions not included in blocks without large exorbitant fees) there would be developed a new digital currency based off original Bitcoin but with this problem inherently solved already?
That could be the case if everyone using Bitcoin could not cooperate in changing the code at once...oh boy
EDIT:
or maybe just as easily that software could be built ON TOP of bitcoin...
I have a design for a proof-of-stake algorithm that generates consensus without proof-of-work and without time stamping. I need to find developers who are interested in implementing such a project. I don't think you can build anything on top of bitcoin. |
|
|
|
Also, to cunicula. I have been running some simulations of t-test results on data from various random distributions and it looks like ones similar to the pareto distribution are most likely to cause false positives and exaggerated estimates of effect size. Have you ever seen any econ literature saying this? I searched a bit but got a bunch of noise.
If the random distribution generating individual observations does not have a mean and/or variance (true for some pareto distributions), then the central limit theorem doesn't hold and you are in trouble. Similarly, if the variance exists but is just really large, then it will take a huge sample size to get an approximately normally distributed test statistic. I suggest you look at Hayashi's textbook: Econometrics. |
|
|
|
if you don't want the government to do research and put stuff in the public domain, then you have to support private research through strong IPR. There are other ways. Most of known history has proceeded without either, progress has only slowed as government has become more involved. Of course researchers are dealing with ever more complicated problems as well. Right, we can return to the 16th century before the government became involved in IPR protection. Britain was really idiotic to reform IPR in 1622. That whole industrial revolution thing really set us back. The US was really really stupid to improve upon Britain's IPR protections. It just made the mess even worse. Is this some kind of joke? Your head is already comfortably lodged in your ass, there is no need to try to stretch it further. You asserted something without providing reasoning or evidence as well.
Pretty rich statement coming from Mr. Asshat himself. Mr. Asshat speaks of alternatives. It is not clear what he means, but let's be charitable and assume he has a concrete idea. One libertarian alternative is to have a coalition of potential inventors bribe the initial patent holder to put his IPR in a pool that is commonly held by the coalition. Research suggests that this is more harmful for innovation than the initial monopoly patent right. It creates a persistent cartel which is more difficult for outsiders to break than the original monopoly. http://www.immagic.com/eLibrary/ARCHIVES/GENERAL/NBER_US/N090616L.pdfWithout reading the whole thing I skipped directly to the data and wondered why they fit a "fourth-order polynomial", why is 1905 or whatever it is less than 1900 in figure 5. Why is number of patents not normalized to total patents for that year. The fact these figures exist makes me immediately suspect crap paper with crap reviewers. Perhaps it is explained within, if so where? You need to read the entire paper and weigh all of the evidence presented. |
|
|
|
so if government were to spend $1 Billion to fuck the bitcoin system they could get the best miners and hardware that would only solve blocks if they included a transaction fee of a specific special number that they give and design people that they wanted to use the bitcoin system
Sure. But... 1) There are plenty of cheaper, easier attacks that would disrupt the bitcoin system anyway. 2) You could just as easily spend $100 million to destroy the market value of bitcoins, by manipulating the free market. 3) The community would want to continue, and so, everyone would agree on a new algorithm, rendering all that expensive hardware useless. There are plenty of reasons why mining attacks cost far more than other, more realistic attacks. $1 million is a more realistic estimate of gov't costs than $1 billion. As usual, jgarzik claims that disruptive attacks which cause the attacker to lose money are more realistic than regulatory attacks that cause the attacker to gain money. Of course, this makes no sense, but apparently the dev community wants to reinforce this misunderstanding. |
|
|
|
if you don't want the government to do research and put stuff in the public domain, then you have to support private research through strong IPR. There are other ways. Most of known history has proceeded without either, progress has only slowed as government has become more involved. Of course researchers are dealing with ever more complicated problems as well. Right, we can return to the 16th century before the government became involved in IPR protection. Britain was really idiotic to reform IPR in 1622. That whole industrial revolution thing really set us back. The US was really really stupid to improve upon Britain's IPR protections. It just made the mess even worse. Is this some kind of joke? Your head is already comfortably lodged in your ass, there is no need to try to stretch it further. You asserted something without providing reasoning or evidence as well.
Pretty rich statement coming from Mr. Asshat himself. Mr. Asshat speaks of alternatives. It is not clear what he means, but let's be charitable and assume he has a concrete idea. One libertarian alternative is to have a coalition of potential inventors bribe the initial patent holder to put his IPR in a pool that is commonly held by the coalition. Research suggests that this is more harmful for innovation than the initial monopoly patent right. It creates a persistent cartel which is more difficult for outsiders to break than the original monopoly. http://www.immagic.com/eLibrary/ARCHIVES/GENERAL/NBER_US/N090616L.pdf |
|
|
|
while the state evaluates social benefit vs. research cost.
Lol. As usual, I am overpowered by the careful reasoning supporting your libertarian convictions. |
|
|
|
If you think funding research to put it in the public domain make sense, you are absolutely welcome to do it. If you're spending your own money, it's much more likely you'll get the cost/benefit analysis right. Perhaps you are right, perhaps I am getting it wrong. The beauty of a free market is that everyone gets to do their own such analysis.
Sure, but the private problem is much, much simpler. The entrepreneur evaluates private benefit vs. research cost, while the state evaluates social benefit vs. research cost. The fact that private calculations are more accurate isn't much help. The entrepreneurs are trying to solve the wrong problem. Unless the entrepreneur is megacorp, a la Singapore Inc., then the entrepreneur will fail to undertake socially beneficial projects. If the alternative is mega-corp, I will take a US/European/Japanese democratic state any day of the week. By the way, in developing countries, megacorp is often the way to go. The state's suck so much that citizens are often better off under megacorp's monopoly. |
|
|
|
There are only a couple smart ways of managing BFL.
Scam A
1) Sell to consumers.
2) After 1 is finished sell more to yourself.
Scam B
1) Sell to consumers.
2) Lower prices, sell to more consumers.
3) Lower prices, sell to more consumers.
...
n) After (1)-(n-1) are finished, sell more to yourself.
It can only end one way. Scam B is just a longer, long con.
That's why competition is so important. It forces BFL to show their hand eventually. Competition is not good, here. For the current front runner, it increases the attractiveness of scam A relative to scam B. |
|
|
|
Sure, Opportunity Cost 101. But not everything is measurable in money. What about sentimental things like health, education, or family? How much money is your health worth if you lose it? $1000? What if you can afford a million dollars - does that make your health worth more? That's a decision people have to make ahead of time. We are failing to make that decision and it's one of things destroying our health care system. If you want hundreds of thousands of dollars spent to extend your life a month or two at the end, that's your choice. But I don't see why my health insurance should be unaffordable because of it. Our health and education systems are going down the tubes precisely because we don't conduct cost/benefit analyses and thus we don't focus our costs where they get us the greatest benefits and have to overpay for the benefits we really want. It doesn't make any sense to apply cold-blooded market efficiency to certain things. Quite the reverse, it doesn't make sense not to. Otherwise one ends up going down a slippery slope towards analysing the financial pros-and-cons of things like euthanasia and eugenics. It's even worse when you don't do that. You wind up driving costs through the roof and people can't afford the services they really do want. The protection against euthanasia and eugenics is individual rights. I agree that in many cases government research may be a waste of time. If people are allocated a budget, they better use it or they'll lose it!  However, my point gains the most significance in cases where there are clear moral values at stake, yet its worth is not measurable in money. E.g.: cancer cures, search-and-rescue/disaster relief tools, plentiful sources of potable water.... Even loss-making space exploration has value. I agree. But if you factor in the probability that the research will not pan out, I think these all flunk any sensible cost/benefit analysis. I don't think you can make a coherent case against a cost benefit analysis. Joel, if you don't want the government to do research and put stuff in the public domain, then you have to support private research through strong IPR. However, IPR generates a monopoly rent that has a negative knock-on effect on all subsequent innovation. The negative knock-on effects work as follows, if I invent B and you have to license A to invent B, then A is going to get a bunch of profit from my invention. If I invent C, and I have to license B to do this (and thus A as well), then A and B are both going to profit from me. Continue down the chain and there is no point in bothering to do D,E,F,G,..., the pie has to be divided among so many rent-seekers that there is nothing left for me to gain. [Consider for example the case where SHA-256 and ECSDA belong to Microsoft. Does bitcoin happen?] There is plenty of evidence that putting research in the public domain stimulates subsequent invention. (i.e. that IPR reduces subsequent innovative output) e.g: http://economics.mit.edu/files/6803http://www.sfbtr15.de/uploads/media/Moser.pdfThus, even if the government is many times more inefficient than private research labs, it can still be better for society for the government to fund research. This is particularly true for basic research that has wide applicability (i.e. which can produce a long chain of dependent innovations). Alternatives: No IPR and no state-funded research. For things that are very easy to copy, such as life-saving medicines, this leads to almost no research being done. Massive corporation that owns IPR to all science. I guess that is the libertarian solution to this problem. To me it sounds like a movie in the sci-fi/horror genre. Your talk of cost-benefit analyses is naive. Credible cost-benefit analyses are extremely difficult to do in this area. The problem is dynamic and large-scale. This makes it very complex. |
|
|
|
|
There are only a couple smart ways of managing BFL.
Scam A
1) Sell to consumers.
2) After 1 is finished sell more to yourself.
Scam B
1) Sell to consumers.
2) Lower prices, sell to more consumers.
3) Lower prices, sell to more consumers.
...
n) After (1)-(n-1) are finished, sell more to yourself.
It can only end one way. Scam B is just a longer, long con.
|
|
|
|
Stratum/Electrum
Only need to solve the web-based wallet part.
I think a great way to encourage people running nodes is by making it easy to connect a feature rich and network light mobile wallet to you own personal electrum-type server. That way your phone, tablet, or frequently disconnected because you drag it everywhere laptop can have an instantly available wallet without trusting third party servers. The electrum server and stratum are freely available, but it would be nice if some of the bigger bitcoin news and information sites put together a guide for the mobile bitcoin user doing business in person. This remote computing seems like a good idea. Mobile Phone stores private keys and connects only to home computer. Home computer handles blockchain, node, etc., but not wallet. |
|
|
|
Rergarding the other comment. They will already have 0.9x in the bank as USD. I am suggesting that they risk another 0.1x to get revenue of 0.5x. If they get caught, worst case they lose 0.1x. If they do npt get caught they gain 0.4x. Is the probability of detecting them higher than 80%? If not, the wise business decision is to go for it. (note this is based on accepting your claim that value drops to 0 if they are detected. I don't believe that, but it doesn't really matter.)
I didn't claim it would drop to 0, but I think it would get pretty close. Even if they only maintained 40% or 50%, it would still be a massive red flag to anyone not only that one entity has such a large percentage of the network, but that they're also the (most likely) far and away largest provider of mining equipment. Where would the massive red flag come from? How do you know they aren't mining right now? Regardless, your assertion that it is worth gambling 0.1x to gain 0.4x is very shortsighted, as it ignores future earnings. BFL has an apparently not insignificant investment in infrastructure and training as well as brand recognition. They've spoken at length about the desire to make this a long term business and bring out new generations of hardware. They would essentially be throwing that away since eventually such a scheme would come out.
If they are assured a permanent monopoly on hardware then I agree that we will be okay. If not, then building up a huge stockpile of private equipment makes perfect sense. Such a stockpile drives down the returns to mining and discourages entry from competitors. |
|
|
|
just to be clear, will i be able to buy a kind of "currency" using your system? One thing you can do is accept an offer to exchange any kind of currency somebody is willing to sell for anything you already have. You can also place a standing offer that others can accept. Is there any early adopter advantage? Since there's no mining, there isn't the early adopter mining advantage that Bitcoin had. Early adopters, however, will have the first crack at any opportunities the system does provide while later adopters will have missed at least some of them. This is, unfortunately, a dangerous area for me to comment on. (It's much the same reason Bitcoin folks can't say things like "Buy bitcoins as an investment, they're almost sure to go up".) One possible early adopter advantage is that the system may have less liquidity in the beginning. Lower liquidity could mean larger spreads. For example, if nobody is offering to exchange a particular currency pair for a decent rate, you may find somebody who needs to exchange that currency pair to make a payment might accept an offer that's really not very good. You can easily place a number of "not very good" offers and if anyone gets desperate and liquidity is poor, that's a profit opportunity. However, there's no guarantee liquidity will be or stay poor. And you have to remember to cancel your offers if the exchange rates change. So this is purely speculative. Could you use some sort of pyramidal marketing scheme to create early adopter advantage or are you trying to avoid that entirely? Early adopter advantage is very good for getting something off the ground. It provides the reason to initially adopt a network technology and thus partially solves the chicken and egg problem. Most economic models of network technologies suggest doing something like this to bootstrap a network. My view is that bitcoin could not have gotten off the ground without early adopter speculation. |
|
|
|
No - by spending the dust, the dust makes it into blocks. Any transaction that's very recent, dust or otherwise, will make it into the main transaction set just for being recent.
My concerns are: 1) there may be incentives for miners to ignore dust. [not necessarily a bad thing, but we should consider carefully whether that would be the case.] 2) I'm not sure how useful it will be able to verify 98% of txns in a block if I can't verify the other 2%. I will have to operate under the assumption that someone else is verifying these for me. However, if I'm doing that, why store the blockchain at all? After all, if the block gets rejected, then the other 98% of txns get rejected with it. Note: I'm not sure I'm right here. I just want to raise the issues. Still wondering about this Say we mop up all the new dust created after each block arrival (introducing new stuff will turn some old stuff into dust). Is there a quick algorithm for doing this? Or would it require a resource intensive search and thus maintenance of a dust index?
|
|
|
|
|
Show Posts
