Topic: Cryptografic currencys in future (Read 2051 times)

Verifying inputs, ECDSA singatures, transferring data, storing blocks, etc.  are technological problems. If computing power grows quickly enough, these problems are solved.
If computing power does not grow fast enough, more efficient uses of hardware will be needed and I'm pretty confident that they will be found. Using hardware more efficiently would make everyone involved better off. If you find ways of using hardware more efficiently, it will not be hard to convince people to adopt these problems. In sum, the problems either solves themselves, or people become highly motivated to solve them, and everyone else becomes highly motivated to adopt any solution they find.

Sorry if you feel like I am minimizing your contributions here. The above is not my area of expertise. I'm sure it is important stuff. I just feel like the developers have a handle on the above problems. I have confidence in this area.

Securing the blockchain is a political problem. Growth in computing power doesn't matter in any significant way for blockchain security. The game is determined by allocation of voting power and rewards for voting. Altering voting rules and rewards creates winners and losers, and this creates conflict. Even if a superb solution is found, it will be very difficult to coordinate adoption of that solution. Identifying a good solution is not a rewarding endeavor. For the most part, it creates conflict and pisses people off. Adoption of a bad solution early on will have a persistent negative effect on long-run outcomes (hysteresis). Solving a problem like this from the beginning is very important. Adoption of a bad solution, like that in bitcoin, is damning. I think that makes this a pressing issue.

I don't think bitcoin developers have a handle on this issue at all. Slowly removing block rewards and waiting to see whether the blockchain survives is not a viable plan. 'wait and see' will eventually blow up in their face. For the same reason, I am against the gradual introduction of proof-of-stake in PPC. The proof is in the pudding. If you plan to secure the blockchain long-term using mechanism A, then put mechanism A in place now and show that it works. Otherwise, why should I trust it. Particularly when game theory suggests that it is not likely to work at all.

This is why I'm talking about divide-and-conquer optimizations...


I mean that if there are costs associated with mining then somebody needs to pay for them.

So it is about reducing costs. Which is not only proof-of-work, but all of the other stuff.

It is absolutely different thing in terms of security. Both for those small guys and for double-spends. (If you'll have large amounts of stake concentrated in hands of ewallet operators double-spending will be trivial.)

You know, I'd rather put my money into USD bank than into PPCoin ewallet. At least USD bank is regulated  and can pay some real interest. (Here where I live banks pay about 8% per year for deposits in USD...)


Oh, have you analyzed how it works? I did.

Consider a simplified scenarios where all stakes are of same size. Mining is essentially a lottery where chances to win are inversely proportional to number of stakes which are actively mined.

If number of actively mined stakes is large then chances to win in one round (after one block) are small.

E.g. suppose there are 52560 active stakes -- one for each block within one year time frame -- then it takes something like a year to win a stake if you check your stake each block.

If you do not check each block you have much lower chances. To the point where you'll die before you win in lottery. And at large time scales compounding starts to matter too.

And don't forget that to mine you need to download and verify blockchain... So you don't win that much from mining only from time to time.

You guys are arguing the same thing.

The question is whether miners' security services could be provided without accompanying expenditures on equipment and electricity. If so, miners could be paid much less without any reduction in the quality of service received by their customers. If miners are paid less, inflation and/or txn fees can be reduced. This would make the currency more useful for everyone.

I recognize that this is not an issue most people think about. The fact that everyday users ignore future fee hikes comes from a subtlety of blockchain design.

tax on spends = (txn fees + block reward per day) / txn volume per day = (22   + 7200) / 177,079 = 4.1% [I think this is very much an underestimate, but cannot come up with better numbers.]

Users are effectively taxed 4% on each txn. But users ignore this because fees account for only 0.3% of the total tax burden.

Later txn fees replace inflation. Bitcion's competitiveness is largely based on its low fees. 4.1% isn't especially low, however. For bitcoin to succeed this number has to go way down.
I think 0.1% is a reasonable and achievable target.

That can only happen if proof-of-work is replaced with a more efficient security system. One which does not rely on continuous input of external resources.


Blah blah blah... prove that it is "fundamentally unpossible"

1) You are ignoring financial intermediation. Smallholders who spend coins often can't mine effectively, but this doesn't mean they can't earn interest too.

Smallholders who want liquid accounts will put their money in online wallets. Wallet services will mine and pay out interest to depositers. The interest will likely be only slightly less than one would get from mining oneself. You are right that there will be a tax on spending, but it's going to be negligible.

2) Mining PPCoin doesn't work like you say it does.

You are paid 1% * coin-years. You can mine once a day, once a year, or once a decade and you still get the same interest rate. (There is an issue with compounding here, but it is negligible.)

Actually it's more subtle that this for ppcoin. Let's assume that cost of mining doesn't depend on size of a stake.

Then whether it makes sense to mine depends on size of a stake! E.g. if person has X coins and cost of mining per year is C, then it depends on whether (0.01 * X - C) is positive.

So bigger miners get bigger nominal net income out of mining, while small miners will be losing money on mining.

Also note that it is nominal income, we haven't taken inflation into account. With inflation real net value of mining might be negative!

So we get this:

• people with small quantities of money are taxed
• people with bigger quantities of money might make nominal profit, but still lose value
• people with large quantities of money actually make profit on mining

When did it become a bad thing to pay for services rendered?  That is what the tx fee is, I don't see that as a tax, which is an arbitrary confiscation of value for likely unrelated purposes to the commerce taking place.  The tx fee is not arbitrary and directly paying for the services of miners... with that said the tx fee in cryptocurrencies that have no upper cap on block rewards is not needed economically, sadly they are needed for security and health of the block chain itself, so devising a plan to overcome spam/block chain bloat in an uncapped currency while eliminating tx fees would make for a better system all around.  The next evolution would be a cryptocurrency that is unbounded while not having any effective inflation or deflation, this eliminates the effective "tax" on both creditors and debtors, as inflation is a "tax" on creditors (and savers usually) and deflation is a "tax" on debtors.

The "tax" on debtors is an easier fix by devising a negative interest rate system, thus you can have mild deflation and no "tax" burden on debtors.  Inflation or "tax" on creditors/savers is the harder one to solve because a positive interest rate system simply increases the burden of inflation and thus the effective "tax" on creditors/savers.

If it's cheaper to mine tax can be reduced.

Think about it: tax = costs of mining.

Not really. It taxes both spenders and non-mining savers.

And even if I was wrong, that doesn't make you right.

The tax we have been talking about is txn fees per block (tax on spenders) + block reward (tax on savers)

Schemes which maintain security while simultaneously lowering this sum enhance security.

It is all about security and has no direct relationship with scalability.

Hmm...

Let's just look at what you wrote that I was responding too.
And then at how you backed up your argument...


Embarrassed yet?

I think discussion about reducing "tax" is useless unless you know some concrete scheme to do that. It is equivalent to working on scalability of Bitcoin.

Actually I've made some research on this matter. But it doesn't mean I can't write on other topics, right?


I think it would make sense if fees will be proportional to costs. E.g. for spending it is costs of sending data over network and doing lookup and ECDSA verification on all nodes. For hoarding it is cost of storage.

Even if storage was free, hoarders would still benefit from network security, so it makes sense to tax them. Although it's hard to quantify how much. Perhaps we can find some equilibrium here.

I think it would make sense if fee structure was transparent and fair, rather than arbitrary.

I didn't know you are so math-impaired. This is wrong on many levels...

1. It takes time to mine a share, likely a very significant time, like years. But prices will be adjusted continuously. So suppose in a year prices are 1% up, but you haven't mined your share yet. It is equivalent to losing 1%. So effectively it is equivalent to tax on spenders.

2. Furthermore, coin-age is reset after sending coins. So money which is in active circulation does not earn interest. So, again, it is a tax on spenders.

3. To mine you need your wallet to be attached to an active mining node. Which means that people who are using thin/mobile wallets simply won't be able to mine. Moreover, people who care about security of their coins will keep then in an offline wallet. So clearly people who can't/don't want to mine pay to people who are mining.

4. Mining has its costs. With ppcoin you'll need to expend resources all the time just to keep your money intact. So how is it different from taxation? You're just paying taxes with your hardware resources. And, no, you don't secure your money yourself since you'll be securing other people's transactions, and also resources which need to be spend is not proportional to money you own.

5. Mining has its costs. It is fundamentally unpossible to pay for it without paying for it. Thus if you want to reduce costs of spending/saving you have to invent more efficient transaction processing. For example, based on divide and conquer. Otherwise it is just about shifting costs around.

No. This is clearly wrong.

PPC coin "taxes" miners by printing coins. The burden of taxation is proportional to the amount of money saved. This would be like "robbing savers," but ...
PPC rewards miners by printing coins. The rewards are offered proportionally to the amount of money saved.

There is no net taxation at all.

I don't think discussion about who to expropriate is useful. It is a complicated issue and rational arguments can be advanced for almost any position (see below).

Money is supposed to have two properties.

1) Cheap to use for exchange
2) Cheap to use as a store of value

If we have large txn fees, we are interfering with use (1). If we increase the money supply over time, then we are interfering with use (2).
If we have a good which satisfies properties (1) and (2) better than bitcoin, then this good should dominate bitcoin as a form of money.

So those are the reasons why we should avoid taxes. However, because of externalities, we can also make arguments in favor of taxes. Bitcoin has maintenance costs and these are related to user behavior.

Txns generate externalities. Txn has to be stored and transmitted by everyone else forever. This is costly. If people just sit on their coins, no data is generated. Therefore, txn taxes seem sensible in the bitcoin system.

Hoarding of coins generates an externality. All else equal, hoarding increases bitcoin market capitalization and therefore increases the economic motivation to destroy the bitcoin.
Accordingly, hoarding increases the need for security measures protecting bitcoin. Hoarders should pay the security costs that they impose on other bitcoin users. Therefore, taxes on hoarding are sensible in the bitcoin system.

Seriously? People who do not lose their savings are compensated for what sloppy users do?

This is anti-insurance.


I don't think it's possible to take away something one never had.

You can as well call food prices "a tax on hunger" or something.


This issues are orthogonal. We can argue about whose money to expropriate while trying to reduce the need for taxation.


Doesn't ppcoin incentivize mining with interest rate? Then it falls under "robbing savers" category.

This is exactly what ppcoin project is trying to achieve, to demonstrate that work is not required to secure block chain, thus further reducing the cost burden on users of future cryptocurrencies.

J. Becker et al in a lengthy and possibly prejudiced paper 'Can We Afford Integrity by Proof-of-Work?' argued that Bitcoin is not affordable compared to existing infrastructure. I don't necessarily agree with that as Bitcoin may be affordable enough but high energy consumption is one of the very few legitimate criticisms on Bitcoin in my opinion.

once asic there is ltc for the gpu farms.

Appreciation compensates savers for the possibility of loss. You would like to take this compensation away and distribute it to someone else. That is a tax.

To reiterate, the focus should be on reducing the need for taxation, not arguing about whose money to expropriate.