No I am not. I am talking about miner revenue and costs. If orphan risk cost is 50% of miner revenue, then "things break down" yes, this does not really imply much about the orphan rate itself.
If orphan risk cost is 50% of miner revenue, then half the 'solved' blocks are being orphaned, no? Or do you have some other definition of 'orphan risk cost'?
Sorry I do not understand your point here. It is pretty clear two miners in the same location may have lower propagation times between them than one on the other side of the world. It is pretty logical, isn't it?
Yes, but a single 'miner' is often geographically distributed itself - not in the same location. Further, you ignore my latter assertion that, should it take considerable time for miner A's block to propagate to the majority of the network, miner B's block, perhaps actually mined later in time, if propagated quicker to the majority of the network, may orphan miner A's block.
If retarded propagation was in any way a net benefit, miners would merely insert gratuitous latency before announcing their blocks. Whether maxblocksize were tiny or huge.
I do not, you seem not to understand my point. The logical thing to do is keep adding transactions up to the point where the fee on the most marginal transaction is equal to the marginal orphan risk cost. At the same time it is logical for wallets to only pay a fee which is a small premium to the expected marginal orphan risk cost. This dangerous dynamic ensures orphan risk costs are high relative to revenue.
You are correct - I no not understand your point. Why do you think that the point where marginal orphaning risk is equal to marginal additional fee has anything to do with the point where absolute orphan risk cost is equal to absolute income? These two curve crossings are not the same.
This is ALL I am saying. We should not have the fee market driven by orphan risk. Whenever I make this point it is constantly misrepresented as if I am saying something else.
I still think you are saying something else. I am not _intentionally_ misrepresenting you. Care to explain further?
I think wasted work is the most understood one.
In what way is 'wasted work' any sort of a threat to the system?
I do not think the orphan risk cost would be high. I even put RELATIVE TO MINING REVENUE in big letters, in a desperate attempt to avoid people misrepresenting me, it did not work.
Ohhkaaay.... Why do you think that orphan risk cost would be high RELATIVE TO MINING REVENUE?
The limit as in the major point of competition in the industry, driving decision making. If you do not think it would be, answer me this question:
Why would wallets pay a fee any higher than a small premium over the expected marginal orphan risk?
Because miners will have no incentive unless there is profit. I don't know where mining margins will come to equilibrium, but many old-school industries shoot for ~7% net. Accordingly, if we ever equilibriate, I assume that wallets will trend towards paying about a single digit percent more than that point where the fee would equal the expected marginal orphan risk. So I am with you so far, I believe. The additional step that I don't get is why you think that this implies this is the point where
total orphan risk cost is equal to
total mining revenue.
Oh - if that is your definition of 'pushed right to the limit', then yes. I fully expect that the system will be driven by each miner making a series of decisions that result in him 'pushing right up to the limit', as defined in that manner. Why do you think this is a problem?
[doublepost=1469270212][/doublepost]
No I do not. I want Bitcoin to be a system where a small minority of financial elites or any other kind of minority can prevent a hardfork. This is not control, it cannot impose changes but only prevent them. The reason we need this is to differentiate ourselves from the existing fiat system where elites can impose changes.
Stop pretending that because I want a minority to prevent a HF that I want a minority in control.
I realize that this is for cypherdoc, but....
But any change is ok as long as it is a soft fork? No, I'm not following the logic behind your desire.
Now you think changing the limit to 2MB would be keeping things the same don't you?
I recognize that bumping maxblocksize above the point that it affects most generated blocks is a technical change. However, it would _maintaining_ the _economic_ dynamics. And in my mind (despite being a technologist) when it comes to the criteria that should drive the design of _money_, the economic should take precedence over the technical.
And I realize this is not a point you are making in this particular post, but The SegWit Omnibus Changeset is far and away a greater technical change to Bitcoin that a simple bump to maxblocksize. On a fundamental level. While still imbued with its own quirky economic changes.
You think keeping 1MB is a change? If you think that then that is the disagreement, stop misrepresenting the other side.
That is merely one element of the disagreement that I seem to have with the Core adherents.