Quote from: exstasie on March 30, 2016, 11:15:58 PM

Quote from: adamstgBit on March 30, 2016, 10:53:21 PM

Quote from: exstasie on March 30, 2016, 10:39:28 PM

Quote from: adamstgBit on March 30, 2016, 10:21:54 PM

Quote from: exstasie on March 30, 2016, 09:48:56 PM

Quote from: adamstgBit on March 30, 2016, 04:26:36 PM

i believe there are significant incentives for miner not to create blocks that are TOO big

The relay limitations of all miners are not equal. Orphan risk is unequally distributed based on these limitations, which threatens the viability of groups of miners controlling a minority of hash power = miner centralization risk. See Pieter Wuille's simulation: https://www.reddit.com/r/bitcoin_devlist/comments/3bsvm9/mining_centralization_pressure_from_nonuniform/

What defines "too big?" Without some hard limit that defines this, there is nothing to prevent miners from forking networks.

if they have a shitty internet connection whats stopping them from creating a smaller block so they can propagate that block faster?

if they have shitty internet it only incentives them further to create smaller blocks.

you've unintentionally drove my point home?

I have not. At all. Not even sure what you're referring to with that.

"A shitty internet connection" is not the issue. Rather inter-regional internet infrastructure is, which is out of the control of miners.

You're suggesting that the lack of any block size limit would result in a single cohesive ledger (Bitcoin) vs multiple ledgers. I assert that without a hard limit, miners would disagree on the block size that they are willing to relay. The only possible result of that is a chain fork because of incompatible rules.

a miner with poor inter-regional internet infrastructure need simply mine a block in which contains less data, as to achieve a higher proportion rate.

better?

That doesn't address the assertion that larger blocks contribute to miner centralization. Mining viability over time (due to subsidy halvings) depends on fees becoming a greater share of block rewards. You're then putting some miners between a rock and a hard place: 1) increase orphan rate or 2) decrease block reward. Either way, their viability is threatened, and this disproportionately affects smaller miners/pools, thus contributing to miner centralization.

See the link I posted: The group of smaller miners loses a % of relative income. If they publish larger blocks, their loss percentage goes up slightly further.

Perhaps more important: If a group of miners is willing to relay and build on a 1GB block, it doesn't follow that the rest of the network is willing to. Are we supposed to trust miners never to disagree? Sorry, but no thanks. I prefer that node software makes disagreement (hard forking) impossible.

you've convinced me. in order to make life fair we need rules which level the playing field.

decentralization is the key to victory, if a there are no miners in africa bitcoin will fail!