Topic: Miners that refuse to include transactions are becoming a problem - page 4. (Read 16972 times)

I don't see micropayments being a problem. Even if the average tx-fee paid rises, there will likely always be free transaction processing, even if it's just the company accepting the micropayments that mines those blocks. It will just take longer on average for such transactions to be processed and receive their first confirmation. So unless you're wanting that $0.05 you sent to your friend to be available for a resend in 30 minutes, micropayments should be handled just as well later on as they are now.

Got it.

I really like the suggestion on how to handle micropayments. What i like most about it is it provides miners a means of competition and distinction, and it also gets them into more business than just mining, which, long term, is probably  thankless, low margin work.

Because the network has no idea how much the tx is for.  It only knows the inputs and outputs which are greater than the amount being "spent".


It could but miners have no effective way to determine what that % is. 

I lay out some reasons why % based fee rules are not viable.
https://bitcointalksearch.org/topic/m.819583


There are alternative methods of handling micro payments.  Long, only read if interested in alternative payment scenarios. One would be bulk contracts.  Say WSJ wants to to charge 0.01 BTC per page.  WSJ could pregenerate a list of payment addresses to provide users.  It would provide a copy of the public addresses to a mining pool.  Since micropayments are unlikely to be a source of fraud and likely need to be 0-confirm anyways WSJ can wait for longer than normal confirmations.

The pool could create a contract with WSJ for a flat fee or a nominal fee per tx. So on a per tx basis maybe the pool has a fee requirement of 0.5 cents (USD nominal) however by signing a bulk contract with prepayment the WSJ effective cost is (1/10th of cent per tx).  WSJ pays $500 and pre-purchases 500K low priority tx.  The pool fills blocks as normal w/ "priority tx" and then fills up the blocks w/ these lower priority tx as space is available.  Sometimes WSJ gets 1 block confirms but often it is 4 blocks, 8 blocks, 20 blocks.

The analogy would be low cost bulk mailing from postal service vs Fedex overnight.  junk mail doesn't work at Fedex prices it doesn't mean everything needs to be one size fits all. 

Now hypothetically if enough micro tx demand existed a broker will acts as middle man.  It would contract w/ multiple pools for their "junk mail rates".  The broker would sell the services to multiple micro tx receivers.  The brokers cut would be based on the difference between what they contract mining pools for and what they sell that service to service providers which use micro txs.

There are two issues at hand, which are separate:

1) what TX fees should be
2) how to deal with mines not fully participating in the network

A flat fee of any significance is incompatible with micropayments. 1bitcent seems a very steep fee for something like a NYT or WSJ article I wouldn't pay more than 1bitcent to read

I've read the discussion about this from when satoshi was still around - I'm aware he didn't think Bitcoin to be well suited for micropayments yet, but I don't think that setting a flat fee is necessarily the right thing to do. People want to pay other people micropayments, and we should let 'em if we can. I've actually wanted to send such payments myself, and didn't because of the fee.

I'm not sure I understand why the fee is always discussed to be flat or per KB.

Maybe the client could recommend a Percentage based fee + 1 satoshi per output?

- Personally I like recommendations. Having the option to send for free for now seems important. It's not affecting the mining bottom line to allow free TX,  why not have a distinguishing factor in the market to use bitcoin? Some time in the future it may be determined that free TXs do not have enough of a chance due to blocks being full of paid TXs. It is only at this time that we MUST change the fee structure (on the network side)

I'm not an expert on #2, but it didn't seem to concern satoshi, and it doesn't look like anyone has raised serious technical concerns in the near to mid-term with MM not relaying TXs. It's not a serious security issue at this time. Technomage suggested taking time to consider the long term approach. This is correct. Making changes in haste will only create new problems down the road - problems which may not be so simple to correct then. If this system is to last 20 or more years, we can't think about this just in terms of what miners and users want today, but what will be compatible with what they'll want 20yrs from now.

Paypal costs $0.30 plus 2.9%. 

Current fees are roughly 0.001131751 BTC per tx. 
Even if tx fees on avg were 0.01 that would be an 883% increase but still a tiny fraction of Paypal.

However compared to Paypal (using current BTC:USD exchange rate)
$1 USD transferred  Paypal fee: 0.07 BTC (7x higher)
$10 USD transferred Paypal fee: 0.13 BTC (14x higher)
$100 USD transferred Paypal fee: 0.70 BTC (70x higher)
$1,000 USD transferred Paypal: 6.50 BTC (650x higher)

MtGox for example never includes tx fees.  One of the most sucessful bitcoin ventures and it provides nothing in compensation to the network.  The Bitcoin network will eventually need to be run on fees.  Nobody is saying massive fees overnight or that even 1 bitcent is going to make up the majority of block rewards but it starts the ball rolling.

Currently: ~0% fees, 100% subsidy
Someday: 100% fees, 0% subsidy

Even IF average fees were increased 800% (<1 bitcent per tx on average) and tx volume doubles after the block reward declines to 25 BTC fees would only make up 4%.  Still I consider 4% more healthy than 0%.


One Bitcent tx fee would increase fees

Perhaps the mystery miner includes transactions which meets his fee requirements as well?  He's just set a high fee requirement.  Try a transaction with 50 BTC fee, and check if he includes it.

The point of the proposed change is to reward responsible miners who actually contribute to the network with more blocks.  Satoshi suggested a part of each block to be free, and miners who believe in Bitcoin as a currency include those free transactions.  If Bitcoin transactions get more expensive than PayPal transaction (which you actually suggest for small transactions later in this thread), then people will use PayPal instead, making Bitcoin less valuable for everyone.  It is not in the interest of most miners.

Agree.

This is the broken window fallacy, by the way, Atlas.

This is off topic, but I honestly don't think that the first block reward drop will change things that much. It will make tx fees slightly more relevant but still not very relevant. If you presume that Bitcoin usage is going to grow in the coming years, then Bitcoin price will most likely more than double as well. Also the smaller inflation will probably increase appreciation in Bitcoin's value. That will basically compensate rewards in a way that the total miner reward in fiat is likely to actually increase significantly per block, which will mean more mining investments, more competition and a more secure network in absolute terms (although weaker in relative terms).

In relative terms the network is bound to get much weaker than it is now but that is actually fine, at least for a long while. The security level in relative terms is overkill right now and in absolute terms it can only get better if Bitcoin continues to grow. Eventually we'll reach a point where tx fees have a significant role and then we have an interesting situation where we basically switch to a tx fee market system that works or if that fails, the only other solution to providing a safe enough network is a PoS/PoW hybrid.

Now back to the topic. I have to agree that it's possible that this problem could get less significant in the future and simply go away eventually but we can't know for sure. It could also get worse if more botnets start to apply similar methods. I think it's healthy to think about possible solutions if this gets worse. It's bound to get better eventually just like you guys are saying but the issue is how bad does it have to go in the short term to be unacceptable.

My problem with this hasn't been the practical issues of it, it doesn't increase average tx confirmation time by much and that angle won't change significantly even if the percentage of no-tx hash power increases. The problem is that the mystery miner entity is partially leeching on Bitcoin and I don't like it, and apparently there are a lot of people like me. But it could be that trying to apply a quick solution to this could be problematic in the long term. The best solution to this would be if we could somehow expose the botnet (if it indeed is a botnet) and stop it that way. That is very difficult as well though.

It's a simple fact that any extra incentive to mining including transactions will make mining more profitable to those who do, and will push difficulty up, directly reducing the power of ANY individual entity.

Simple logic really: more reward to mining including transactions => more proper mining, more difficulty, less baseline reward for everybody.

Whatever else is done, this step sounds trivial and it's also fair. It also reduces uncertainty about the reactions and instability that block reward halving might cause.

Not sure what your definition for "forseable future" is.  As long as tx fees are essentially 0 so no matter what the block reward is ~0/anything is still essentially 0%.

Even a modest rise in fees can provide meaningful change to that dynamic.  Hypothetical scenario.  Over next 12 months tx volume doubles to ~100 per block and block reward falls to 25 BTC.  If miners start enforcing fees and avg fee ends up being just 0.01 BTC per tx that is 1 BTC in fees and 25 BTC reward.  Fees are a small but hardly insignificant 4% of total revenue in just 12 months.

Hopefully on a longer time frame say 3 or 4 years tx volume will quintuple (or more).  At say 250 tx per block and 12.5 BTC subsidy.  An avg fee of just 0.01 BTC per tx results in fees making up 16% of total revenue.

If USD:BTC remains the same then mystery faces falling revenue and will need to adapt (or make less).  If USD:BTC prices then global hashing power will also rise as miners chase larger block rewards (including fees) and "mystery" makes up a smaller portion of the network.

Going forward bitcoin will need to be supported more and more by fees anyways.  Any "problem" myster presents is temporary at best.  Breaking changes to protocol to "fix" a temporary problem seem dubious.

I really don't understand how some of you simply ignore the fact that transaction fees are irrelevant in comparison to block reward, and will be for the forseeable future.  Even if they weren't, there are plenty of people out there with the incentive to disrupt the Bitcoin network by mining without including transactions, regardless of what the fees are.

Nothing you have suggested so far changes that.

No. The path forwards is to discover how this is being done, so that people whose computers are compromised can get rid of the malware.

Meanwhile, empty blocks are a minor irritation but do not threaten the foundations of Bitcoin.

Here are Satoshi's thoughts on the matter from the early days of Bitcoin:

Even though benjamindees seems to be on a warpath against the miners, he has some good points. This issue, even though it's not critical, is one more thing that makes Bitcoin suspicious overall. I'm one of the biggest Bitcoin-enthusiasts out there and even I don't like this at all. I can't explain something away to other people if I'm bothered by it as well. In addition I'd like to point out some important aspects to this.

First of all, some of you claim that it's just a matter of finding a tx fee high enough to incentivize this mystery miner to start adding transactions. I think this is entirely false. From what some people have speculated, the benefits he could potentially gain from not adding any transactions, or not verifying transactions at all, can be sort of huge for him. There is a possibility that he simply doesn't care what the tx fees are in the transactions because he doesn't even look at them.

This means that we could have a tx fee of gazillion and a gazillion people paying those tx fees, but he still wouldn't care, because it would possibly lead to many of his miners to be discovered and thus his whole mining operation would be threatened. It could certainly be that Bitcoin needs to become the most expensive money transfer method in the world for him to care, and then it wouldn't even matter anymore. Who would care about Bitcoin if it was that expensive?

So not doing anything is basically ignoring the issue and hoping it goes away or doesn't become too bad. My other point is somewhat related to this, I'm starting to feel that there is an ideological conflict going on right here. We have people who want to desperately hold on to a pure market approach even when we have a clear exploit that allows miners to leech on the network. Then we have people who think that some restrictions are appropriate to stop the leeching.

Then there is this notion that the fixed block rewards are for creating blocks and tx fees are for adding transactions, this has to be the most ridiculous idea I've seen in this thread so far. Who came up with this? I'd like to see Satoshi saying that, I have a hard time imagining it though. Bitcoin is a cryptocurrency and for a currency being able to transact is critical. Ín the case of Bitcoin it's critical to be able to transfer money cheaply, that is one of Bitcoin's major benefits.

Again, I'm not saying that transactions should be free nor that tx fees should be dictated in a socialistic way, but being able to intentionally not work with transactions at all and still get a 50 BTC reward is one hell of an exploit that is likely not going away with the block reward drop.

So we will at least have to put up with this for 5 years, possibly much longer. This is because if Bitcoin continues to grow, bitcoins will certainly be more valuable in the future. That will make sure that nothing really changes as far as mining is concerned, our network just gets a bit less secure relative to the market cap. From the perspective of our mystery miner, nothing changes.

Finally, I think that this issue is not big enough to put a solution in place that either heavily restricts a future tx fee market or has other major side-effects. For example I feel that jgarzik's proposal was too drastic and now it seems, at least according to theymos, that basically all of the recent proposals are problematic. I don't support doing anything hasty about this issue so let's just continue to think it through.

My point still holds. Alt-coins being attacked by legitimate bitcoin pools that take it upon themselves to police the world of cryptocurrencies has a chilling effect on dissatisfied bitcoin users who might otherwise have tried to create a non-pump-and-dump alt-coin.

I personally don't need to be saved from the scammy alt-coins. I'd rather be saved from all those trying to change bitcoin into their vision of an alt-coin.

An alt-coin which uses a different PoW (or PoS) would be immune to that particular attack.

So would any alt-coin that wasn't a dump and pump piece of junk.  You know one that actually submitted a proposal, took in feedback, had a launch date announced well in advance, had a working testnet, got support of multiple pools and miners to have sufficient strength on day one.

Of course no alt-coin has shown any innovation, any progress.  They are wothless pump and dump scams. 

I've found myself asking this more than I would have expected lately.

Although to be fair, when you have an entire mining pool willing to use it's resources to attack any new alt-coin that springs up (in similar fashion to this, I note,) then it's not as if people are very free to just create their own alternatives and go run with them... they're pretty much stuck with Bitcoin if they like the idea of a cryptocurrency but want changes to it.

The law of unintended consequences rears its ugly head again.

Spell out the contradiction, friend.

That'd make things a little more difficult for botnet operators, but it can be bypassed. The botnet software could listen for blocks and transactions on the Bitcoin network and include them without checking them or storing them. The resulting blocks will usually be valid, since legitimate nodes don't relay bad transactions. The botnet could even "check" each transaction by seeing if its other peers either already have the new transaction or will accept it.

I just realized that this attack also applies to other proposals requiring some amount of transactions from the memory pool to be in the next block.

You seriously think, at current prices, that a miner charging a 0.01btc per-tx fee is exorbitant? And in the same context as comparing Bitcoin to Western Union, no less?

While I can see how many would think that's a bit much (as I do,) I hardly find that to be anything close to "lunacy." In fact, such a strong reaction to the idea has me wondering who exactly it is that's desperately grasping for every last BTC they can get at.