Quote from: jonny1000 on October 08, 2014, 08:31:25 PM

Quote from: Gavin Andresen on October 08, 2014, 05:36:07 PM

Would 40% initial size and growth make you support the proposal?

I dont think the distinction between 50% and 40% is that much, the issue may be whether there should be permanent exponential growth in the blocksize limit or some other model. Nothing grows exponentially forever, consumer bandwidth speeds wont and neither will demand for Bitcoin transactions, therefore why does one require permanent exponential growth? Why not consider a model where say the blocksize grows by a fixed percentage each year, however the rate of increase falls by 50% for example when the block reward drops?

Consider the example below:

Year	Blocksize limit MB	Growth Rate
2015	1.0	100%
2016	2.0	100%
2017	3.0	50%
2018	4.5	50%
2019	6.8	50%
2020	10.1	50%
2021	12.7	25%

Quote from: Gavin Andresen on October 08, 2014, 05:36:07 PM

Quote from: theymos on October 08, 2014, 05:37:37 AM

Determining the best/safest way to choose the max block size isn't really a technical problem; it has more to do with economics and game theory. I'd really like to see some research/opinions on this issue from economists and other people who specialize in this sort of problem.

Anybody know economists who specialize in this sort of problem? Judging by what I know about economics and economists, I suspect if we ask eleven of them we'll get seven different opinions for the best thing to do. Five of which will miss the point of Bitcoin entirely. ("...elect a Board of Blocksize Governors that decides on an Optimal Size based on market supply and demand conditions as measured by an independent Bureau of Blocksize Research....")

Theymos, I agree that the maximum blocksize is also an economic/game theory problem, thats why a drew those supply and demand curves to try to analyse this using an economic framework. If the blocksize limit increases and transaction fees fall, this can increase the velocity of money, boost inflation and stimulate the economy. In contrast if the blocksize limit falls, the velocity of money can fall, causing deflation and an economic slowdown. I agree with Gavin that having a Blocksize policy committee to manage the economy, is not very consistent with Bitcoins values. There should not be a problem here as all the economic variables (volume of transactions, transaction fee data) are in the blockchain and therefore economic policy could be automated. Perhaps a simple fixed 50% increase per year is the best solution or maybe a more complicated economic formula is required. However, obviously consumer bandwidth speeds cannot be obtained from blockchain data, therefore it could be important to act with caution and keep the blocksize growth somewhat restricted. Permanent exponential growth should be avoided for this reason.

I like the increase plan on this but as it seems is the one thing we can all agree on is that it NEEDS to be have a growth, and I think it will need a big name like Gavin to make the change.