Optimal fee strategies for miners

[jonny1000]

Peter

Yes, I agree that in the case of zero marginal costs and perfect competition that fees fall to zero.

Good to clear this up.

You seem to think that (total orphaning risk) ~= (total fees), whereas I only agree that (marginal orphaning risk) ~= (marginal fees). I believe total orphaning risk can be significantly less than total fees and I think I've convincingly demonstrated this in sections 3 - 5 of my subchains paper.

I agree with you here. I agree total orphaning risk can be significantly less than total fees, I just think there are some potential circumstances in which it won't be.

You seem to worry that the marginal cost of block space in a free market might be too low (leading to higher costs to operator a node), whereas I worry that it might be too high (leading to fewer people using Bitcoin)

Like Gavin, I optimistically think improvements in technology are likely to mean these marginal costs are eventually insignificant.

I disagree that we will be operating in the top-left sector of the table. I suspect that we are presently operating--and will continue to operator--in the mid-right sector

Do you agree that competition and low orphan risk are positives, therefore the closer we are to the top left the better? I am sure you agree now that better technology and lower orphan risks are positives. Are you also not in favour of competition and a highly distributed mining industry?

You "suspect that we are presently operating--and will continue to operator--in the mid-right sector ", I agree with you. However, do you think its a possibility the industry could change and we could move to the top left for some periods of time? This may be the difference between us.

Please recognise that moving to the top left is possible, and therefore please work with me to find a dynamic market driven blocksize limit that works in this scenario, as well as the others. I think BIP100 meets these requirements.

Note:
In addition to perfect competition, I think there is another scenario of mining incentive time horizons shifting to the short term, which could have a similar impact on this fee analysis.

Many thanks

 

[Erdogan]

I think the cost of potential orphaning is irrelevant to the function of the market. It just means that we are closing in on the limit to the blocksize. The market will itself focus on minimizing it. Now, the focus is on making the best chips, when that competition levels out, other aspects will be the focus of competition. Orphaning risk is one, political risk (concentrating mining within a certain regime) another, power source stability, operating stability, risk of exploding administration cost (right-sizing) are others.

 

[Joseph Woolfridge]

I still think a tiered service as OP described isn't possible naturally. But I might be wrong. If i'm wrong then it already exists to a discernible/acceptable degree. My reasoning is that wallets should be smart and not guess significantly higher for priority than for the minimum fee. Maybe it can happen naturally through luck when there are peaks and troughs of transaction volume throughout the day.

I think Matt Corallo's relay network is an example of a cartel-like structure in action that is benign.

I have only seen a few proposals provided for the fee "death spiral" absent marginal costs:

1. Assume some kind of gentleman's agreement for a minimum fee
2. Use the block size to create an artificial scarcity, dynamic or not
3. Penalize the miner the larger his block becomes (was it Friedenbach?)
4. Create artificial orphaning risk (I believe bitcoin unlimited does)
5. Haven't seen this, but it is the opposite of #3. Instead of penalizing a miner for a larger block you can reward a miner for "standing his ground" and not including a fee that is too low. Not sure it is possible.

 

[albin]

@Joseph Woolfridge

I would argue that tiered service is inevitable, because it arises naturally out of optimizing against large variations in demand preferences (this is the basis for all price discrimination phenomena), but it's likely to be a long time before it's a foregone conclusion, because it's going to take some more halvings before any miners actually go out of business for failure to optimize to collect maximum fees.

 

[painlord2k]

Suppose there are three different groups of people with different fee/confirmation time preferences:
"A" : willing to pay up to $1 for transactions confirm in the next block with high probability
"B" : willing to pay up to $0.10 for transactions that confirm in the next 6 blocks
"C" : willing to pay up to $0.01 for transactions that confirm "sometime today"
​

I think Gavin ask the wrong question because his example assume people in Group A is not people in Group B and Group C. But usually this is not true.

Individuals in Group A could be willing to pay $1 if the transaction is confirmed in the next block, $0.10 if it is confirmed in the next 6 blocks AND 1 hour and not confirmed at all if it is not included before an hour or 6 blocks. The groups don't are mutually exclusive and partially overlap.

Every miners should act in response of what he actually know and he should know the costs and risks to add 512 bytes or 1 KB data to the block he is actually mining.

With tx.locktime and locktimeverify users of evolved wallet should be able to push out transactions with different fees minable in different time frames.

If current block is 400,000 group A user could push out many transactions:
1) pay 1 $ fee if included before block 400,002 or tx is invalid
2) pay 0.10 $ fee if included before block 400,006 or tx is invalid
3) pay 0.01 $ fee if included before block 400.100 or tx is invalid.

If (1) is confirmed the others are automatically invalidated and dropped by the mempool and so on.
I want to remind to all (myself included) often people could prefer (or need) the opposite strategy (increase the fee paid to get the tx included ASAP)

This strategy have a problem: the transaction fee depend only on the time frame irrespective of the fact the actual blocksize is 100KB, 1MB, 10MB. But could make sense for the users to tie the fee to the size of the block including the transaction. Bigger the block, bigger the costs to include an additional transaction, higher the fee I'm willing to pay to be included or, maybe, the reverse.

Problem is, can a transaction be conditionally valid depending on the size (or version) of the block it is included in?

The decision on the fee structure should be left to the miners (what to accept) and the users (what to pay). Competition between users and between miners should yield the best results in this way.