Topic: Miners that refuse to include transactions are becoming a problem - page 3. (Read 16972 times)

Okay, you are entitled to your own opinion, even though reality doesn't care what my or your opinion may be.

This is wrong from the start.

Haplo I demand you back up your LIES with a cite.

Please provide a specific cite where I stated:
a) fees should always remain the same regardless of the market value of BTC
b) miners "deserve" 50 BTC per block in fees
c) how much fees are necessary to regain 50 BTC per block.

You will learn reputation is earned here and with only 49 posts you are starting out as a blatant liar.  I expect that when you can't find evidence to back up your lies you will correct your blatantly false statements about me.  We can disagree on the facts but I won't stand by when some fraking noob start lying to my face.    Regardless of your actions the quote will remain so others can see how little your word is worth.


Quoted for prosperity how little value Haplo's word has:

The only purpose of miners in BTC is to include and record TX.

1: Any miner not including tx for technical reasons is a communist parasite collecting a welfare check and contributing nothing. These people should be excluded 100%. One day there will be 1000tx+ per block, and miners trying to take shortcuts to just win their block subsidy will fuck the network if they dump 1000tx per block on everyone else. This is especially true once people are actually paying a fee to get their tx into a block faster, since they will still end up paying a fee but will get nothing in return.

2: The blockchain is a contentious shared resource, the only purpose of which is to record and verify tx. Any miner who excludes all tx for fee reasons should be considered a monopolist and be excluded.

3: Because miners are a communist entitlement-sucking union, they should be forced to meet a quota in order to regulate how much control they have over prices. If miners are forced to include 10% of tx, then getting into that 10% spot means bidding on the price of those slots. This doesn't stop miners from choosing the 10% they decide to include, or from excluding everyone who doesn't pay their price assuming enough people are willing to do so.

Case in point: D&T ran out as fast as possible and commissioned fee software, calculating how much he has to charge people in order to regain 50BTC per block after the subsidy is cut down, as though miners "deserve" to be paid 50BTC per block in fees, regardless of the market value of 1BTC, regardless of the value of the service they provide, and regardless of what users are willing to pay (100% exclusion for those paying less than his fee). Because the blockchain is a time-limited, shared resource, if all miners adopt similar policies, either because of what they think they're entitled to or because they want to save bandwidth so they can find a block a few seconds before the next pool, then the whole blockchain might soon be empty of tx.

The same type of problem occurred in the whaling industry, and because individual whalers had no personal incentive to limit their hunting, collectively whalers nearly hunted all species to extinction. By the same token, a subsidy with no regulation is likely to lead to all miners destroying the currency, because it is in their stupid, short term self-interest to cut corners and extort high fees regardless of whether or not BTC collapses as a currency.

Meanwhile ... some pools send an LP with no txn to speed up the fact that they need to send an LP to every miner in the pool as quickly as possible when a block is found.

e.g. it's not hard to find a quick DeepBit block with no txns.

So yeah your not gonna get too far with trying to enforce rules like this silly one.

The "absence of coercion", see that is what I really doubt can or will be possible anytime soon or maybe anytime at all, which makes its true implementation a theoretical exercise for me. You don't really know much about my worldview, but you are right, I don't buy into that "absence of coercion" scenario. That is an ideal.

This says all that I need to know about your worldview.  The free market isn't theoretical at all, it's the default emergent order in the absence of coercion.  And capitalism isn't remotely the same thing.

The "free market" really is about freedom. The market may indicate a price where supply meets demand, but each individual is free to trade at whatever price they like.

People often trade away from the "market price". In my city, there is a product which can be bought at the market stall on Saturdays for £1, but which costs £1.99 at the department store and £2.99 at a specialty store. The free market doesn't "dictate" that everyone sells for the "market price".

I think you are right, except that it isn't contrary to the spirit of freedom. It costs what it costs to use bitcoin, very low now, likely more later. But anyone can get access by paying only the legitimate costs, no licenses, permissions etc.

In addition to BitPalish companies any WOT of any kind that springs up will be able to save fees too. Be it as small as my family or a large Bitcoin OTC style thing or Hawala.

One thing that might change over these tx fee discussions is the network topology:

One large network has pros and cons, and suppose the fee on the 'large'  global BTC network is comparably large, and the tx processing becoming slow (as mining rewards will continue to be much more profitable than transactions for a while). A large global enterprise with a sufficient number of customers (like Facebook, Google, Amazon) decided to use BTC at some point, they could make transactions between users fast and without fees. Only going outside of their network to the open P2P BTC network would require high fees to get processed in time. That's comparable to international wire transfers vs. PayPal's internal payments. I see no reason why this shouldn't happen with bitcoin. I see the ad for BitPal: "Sign up to be a member of our xyz BTC community, that has thousands of members and transfer bitcoins quickly without fee". While the global BTC backbone mines primarily for coins and only concerns itself with large, expensive transactions. It really feels so contrary to the open concept of bitcoin, but I see that as a possible option at some point.

So it is not a problem to you then if people like me, who actually want Bitcoin to succeed, stop relaying blocks with no free transactions?  And when we become a majority, you are basically shut out of the network until you start behaving.  Interesting idea.  Because I don't want to support greedy miners who stop perfectly legitimate transactions just because they can run a denial of service attack against transaction processing and demand a ransom from users.  Initially I thought you opposed Revalin's idea, but I can see you actually support it wholeheartedly.  You just don't want to take part, but that's up to you.

Haha, now you take the value in USD into account.  Earlier your arguments have been about percentages of block reward.  Now you are basically admitting intellectual bankruptcy.  Well done.
If you want to adjust it for the price of Bitcoin, be prepared to increase it every week.  Transaction processing does not cost you a single cent.  You will notice the downward spiral in both price and the number of transactions as people stop trusting Bitcoin, because Bitcoin transactions are slow or expensive.  Users dependence on miners seems to be clear to you, but you don't grasp the fact that miners are dependant on users as well.  Short term profit sounds good to you, but you don't understand how bad your idea of unpredictable service is in the long term.
If you think this is a problem, please find a better algorithm.  Adjust it dynamically according to total network hashrate perhaps?

By artificially decreasing service to everyone, making transactions unpredictable and Bitcoin less useful and less valuable.  It's just a malicious activity, and I will support every means to stop it.  Transaction processing is essentially free, and the price should reflect that.  If block size becomes a bottleneck, or when the block reward approaches 0, fees may be necessary.  In a hundred years or so.

The TL;DR of it is: No changes to protocol necessary thanks. (Too many updates as it is - "patches to my money" makes me nervous!)


Too much money is being spent on mining compared to the transaction volume I think.

I get it increases security, but surely a "few supercomputers" are enough capacity if not too much for now.

When the reward is lowered alot of these rigs will shut off.

Even botnets have costs:
1. Risk of being arrested.
2. Programming/hacking time.
3. Updating after anti-virus updates (having to do major virus software updates regularly isn't exactly free)

Fees will rise to some set level; because the cost of processing each transaction is always similar.

If it approaches the actual hardware cost then it will be quite low since the BTC network could theoretically run on a few racks and some well connected nodes.

Lets say its 0.01 btc or in that area, you can still do nano-payments:
1. Transfer 1-5$ to your used site.
2. Use this to pay smaller fees they throw at you.
3. BTC network cost is only and needs only to be paid once.


So there is a small chance the network will trend to become very small and insecure though widely used.
I can only guess, but there are at least a few facts that may prevent this from happening:
 
1. Hardware sitting unused and depreciating in value can earn back some of the loss by mining for transactions.
A few idle corporate super computers doing this and the network is pretty safe.

2. BTC banks may ramp up idle hashing power immediately after big important transactions, then throttle down once its done. This would secure the network and their transactions against double spends.

If they detect an attack later all the banks and others will online their power to defend their interests (a double spend attack messes with every single transaction even if only one party is getting directly screwed).

This would be as simple as not working on the longest chain, but the longest chain YOUR transaction is included in.

The 51% attack would only look successful for a few hours.

3. BTC businesses might run a few nodes at a "loss" to further their user base.

4. Humans are not machines; its possible that people will NOT trend towards paying the minimum hardware cost, but will regard 0.01BTC (at current value) as arbitrary and just pay it.
Doing so you get an oversized and secure network like the one today.

Hrm, it occurs to me that any miner charging a lower fee would be able to collect higher fees in addition to lower fees, so between two miners with equal hashing power, the one with the lower fees will make more money.

The main problem arises directly because of the subsidy. If a merchant is doing a significant transaction and wishes to ensure a fast verification, then the customer must pay for it. When a miner gets a block reward, both the customer and the merchant are paying for it.

1: Because of the subsidy, a miner with significant hashing power could charge impossible fees and still get paid virtually 100% the same amount. The advantage of doing this is limited, but not zero, and generally reduces the use-value of BTC as a whole. Mystery has already demonstrated a good reason for doing this.

2: If the miner in question was a 51% attacker, they could legitimately charge impossible fees on purpose in order to DoS the network. If someone had the resources and the intention, perhaps a government, then virtually nothing could be done about it.

3: If 50%+ of a miner's income is coming from tx fees, then only the situation of the 51% attacker would really be relevant, since miners that did include txs could easily out-earn and outhash a miner smaller than that. If, on the other hand, the exchange value of BTC increases at a rate faster than the subsidy decreases, then charging more than a token fee is stealing. Unless, of course, the costs of running a mining rig increase faster than the value of the subsidy, which seems fairly unlikely since the cost per unit of power or storage decreases exponentially.

I would assume that your average merchant, from time to time, might want a transaction confidence rate of at least 75%, and 80-90% or higher would be ideal for sensitive, high value transactions. If a large miner is charging more than 100 times the amount than for a 50% confidence tx, then 100% confidence, or only 2:1 the value, is costing 100 times as much, or more. The cost to value for using bitcoin as a currency then becomes something like 1/50 as much or worse in that case. In the competitive case, this wouldn't matter much, but in the current case there really is no competition since any reasonable tx fee will be an insignificant source of income to any miner for the foreseeable future. A miner who includes tx doesn't, and can't, make enough money to employ more hashing power than one who charges enough that no tx are included, so a lower cost service has no advantages at all.

The only advantage of lower fees at the moment is that, if all miners charge a 1 bitcent tx fee, bitcoin dies tomorrow and all their earnings from mining become worthless. However, for someone who only trades their BTC for fiat, this risk is almost totally externalized and can't be corrected by the market.

I think the disadvantage of floating price control is far outweighed by the advantages of keeping high-confidence verification costs at least half-reasonable, and by the resistance to a 51% who tried to DoS by excluding every tx by using an unpayable fee. Also, if you publish the fee along with the block, then it becomes very easy to verify that the expected tx for that fee have actually been included, so working the system a la mystery becomes virtually impossible. If the network becomes flooded with idiots like D&T who demand both hefty taxes AND a welfare check, then the network will fail even if their claim is "legit", so the disadvantage is really moot.

Hmm.
A) Isn't this nearly the same (if the broker is powerful enough) as actually having 51%+ of the network? What if the broker made a deal to get miners to not process other brokers' TXs?

B) even if the above is no, how can the broker guarantee to the customer 51%+? He could be unaware that his competitor has seen his brokered control over the network, and contracted to increase the total network size to remove the possibility of the first broker executing any 51% attacks. I don't see how this can be offered.

I do like the general idea of the broker though.

As long as they are including valid transactions, who cares if they have the entire block chain, or by what method they are selecting transactions to include in the block?

Any invalid transaction inclusion will result in their hard work being wasted, as that block would be disregarded by all.  Miners only get paid if the blocks are relayed, accepted, built upon.  Picking random transactions seen since last block, without any validity checking, is a valid strategy, if inefficient and potentially costly.

I can't believe how freedom scares so much people. Miners are free to choose what transactions are going to add to the mined block the same way we, the users, are free to select a fee according to our needs.

Aren't you agree? No problem. Mount a mining rig and start mining allowing free transactions.

The same is doing D&T. He's paying for a patch wich allows him to select transactions to add in his blocks. He's not forcing you to follow him. But if a lot of miners agree with him.... well... that's the power of democracy and you must accept it or start your own mining rig.

In the future I want a free mining market where miners can pick the tx with the fees they want AND, very important for fighting MtGoxtatorship, some big companies offering laundering tainted coins included in fake high transaction fees .

Except then you've killed de-centralization, and added more costs between customers and merchants, and as has been mentioned, between customers and their own private addresses. Also, if the network as a whole doesn't guarantee against double spends, then the security isn't high enough for a merchant to be confident using it in the first place.

The miners don't deserve to be throwing that much weight around, as it just loads costs onto the coin without providing any value that isn't available by cheaper and more reliable means.

I think per-byte or per-kb fees are reasonable, but there has to be some way for the market to regulate fees, or else a single large miner charging 100 times as much as everyone else could cause tx to become so unreliable that the currency is choked out of existence. For example, if "mystery" included txs, but charged 1BTC to do it, just getting over 85% confidence rate that your tx will commit would cost 1BTC.

Certainly 100% commit should be more expensive than 50%, but the blockchain is a tragedy of the commons type problem for one, and for two you don't need tx fees to operate, and you probably won't need them until 2025 or later, if BTC even survives that long.

I would say, every block should have the fee rate of the producing miner published with it, and if the fee is > 2 orders of magnitude greater than 50% confidence, or if the miner doesn't include the tx expected for the fee they charge, then the block should be excluded. The only problem with doing it that way is that, since currently most tx are free (which I think was a bad decision on satoshi's part), there's no comparable fee at the 50% confidence mark. Using max(MinFeeCharged, 50% confidence) instead could probably get around that. With that system, if 50% confidence or MinFeeCharged is 0.0001 BTC, then you could charge up to 0.0999 BTC and still get published, with no penalty for charging less or nothing.


You get paid a subsidy, idiot. You aren't subject to the laws of supply and demand, any more than mystery is dependent on including tx for his income. Also, there is little/no room for competition of service in this market; there's only one quality of service and the only variation is how much time it takes to verify a tx. The only thing a greedy miner can accomplish by charging an excessive fee is to slow everyone's tx down so that BTC is less competitive as a currency.

Based on current interest rates and the usual amounts of currency that are exchanged, yes .01 BTC is exorbitant, and would kill the market if everyone used it. The fees you're suggesting are in fact higher than much of the business that is done, and are multiplicative since any business transaction will generally involve several re-organizations of coin by each party. That's exactly the sort of thing that needs to be prevented.

Compare your fee of US$0.05 with the usual tx fees charged by a stock exchange router: US$0.0005 or less, which is even less than the current spam fee and they don't even get subsidies.

sturle I don't think you get the idea of a free market.  

You think 0.01 is too high..  Fine don't pay it.  some miners set fee requirement at 0, some at a single satoshi, some at 0.001 BTC.  Some at 0.01.  The market determines the price.  The intersection of supply and demand.

You make it sound like if I (me little ole D&T) set min fee of 0.01 for my 0.15% of network suddently no tx which a fee less than 0.01 will ever be processed.  Take all your arguments and replace 0.01 with 0.001 instead.  Problem solved.


Extrapolating the cost when BTC is worth $30 or $100, or $20,000 is just intellectually bankrupt.  Obviously no miner's fees would be fixed in stone for the life of the GPU.  I think a tx should be worth a couple cents.  If BTC price doubles the tx fee can be cut in half and miner generates the same.  As BTC rises competitive pressures will ensure prices are held in check.  That is also a part of a dynamic fee market. Even the spam fee eventually will need to be adjusted.  OH NOES BTC has a 0.005 hardcoded fee on tx smaller/younger than 1 BTC/1day.  When BTC is $100,000 each that means all tx will cost $1,000.  It will cost $1,000 to send $1.00.  OH NOES OH NOES SELL YOUR GPUS NOW BTC IS HORRIBLY HORRIBLY FLAWED.

Ironically you just pointed out something the total cost to users (in terms of fees + inflation) is falling.  Even with a 0.01 fee on every tx without exception (as if I could control 100% of all hashing power on the planet) the implicit cost is lower today than even a year ago and will be significantly lower after the subsidy drop.


Of course not.  Why would someone pay more fees than necessary?  That is the point.  Charging a higher fee gives those who pay more a higher level of service.  

Cross border personal transfers cost 1%, which will make Bitcoin more expensive for any amount < current price at a fee of 0.01 BTC per transaction.  Double that for the transfer to your spending wallet (free in PayPal).  A quick check of my main spending wallet shows that 88 of 320 transactions are 2 BTC or less.  With your fee policy, PayPal would be cheaper for more than 1/4 of the transactions from my personal spending wallet!  Take away one of Bitcoin's most important advantages, and see what happens to the value of your mined coins.  Please try this experiment on one of the alternative chains first, and see how successful it is.
Currently all normal transactions (with the exception of spam and fast laundering) are free.  For transactions resembling spam there is a fee of 0.0005 BTC.  In very rare cases, like when a block is mined by a malicious miner, a free transaction have to wait a block.
This is an increase of 0.01 BTC for almost all normal transactions.  Currently 0.05 USD.  Less than a year ago it would be 0.31 USD per transaction, which is higher than even the normal payment PayPal fees.  And it comes in addition to the collective fee implied by inflation.
If MtGox had to pay fees, guess who would pay the fees.  Their customers.  This goes for all commerce.  Nothing comes for free.  If you want to pay 5 BTC for some service, you will have to buy 5.01 BTC.  Or 5.02 if you pass through your own wallet.  Only 10 US cents by the current value, but you know as well as everyone else that the current value is quite low compared to what it was.
Substitute "subsidy" with "inflation", and you have something there.  A tax on everyone who own bitcoins.  And "someday" is not in our lifetime.
Yes, and 25 BTC is still a very high block reward.  4% fees in addition to that is irrelevant.  Free transactions is a major advantage, and it will solve itself if it becomes a problem because free transactions will no longer run smoothly if transaction volume increase enough.  If you want those 25 BTC to be worth something, you need to mine a fair share of free transactions.  And follow the standard rules to make fee polices predictable for users.  Most of us don't do any adjustments to the fees.

Out of curiosity: Do you pay 0.01 BTC fee for your transactions?

Similar system would provide non-51% double spend protection.  Broker gets contracts for 51%+ of hashing power.  Merchant sends tx to broker who sends it to contracted pools.  Pools guarantee they will include tx in the next block and will not replace it will a double spend under any conditions.  A merchant (like say Walmart) would have a high level of confidence they couldn't be double spent economically.  Walmart would process tx instantly using the guarantee provided by a broker.  Likely that 0-confirm guarantee would be worth a lot more than say normal tx processing so it would provide an additional revenue stream for pools/miners.