Best method of changing the maximum block size

[markm]

Would a "sorry this code is out of date" shutdown for nodes that fail to update in time be "a trainwreck" or just get such nodes out of the way so those that do upgrade in time can smoothly synch in whatever the new system turns out, by then, to be?

Maybe there is starting to even be a case for a Big Bitcoin launch that will allow anyone interested in massive blocks to destroy coins on the old chain in return for being minted coins on a new chain that uses massive blocks.

The new, massive-block chain would still be denominated in bitcoin, without inflating the total number of bitcoins in (spendable) existence.

The new massive chain would be the one to attempt to muscle in on the retail sales market or other markets where handling vast numbers of transactions fast is key.

We could even call the old chain "Bitcoin savings accounts" and the new one "Bitcoin current accounts".

Make it big enough and maybe it will show us the hard way whether blockchains are just too hard/expensive to scale.

Make it a secondary chain in merged-mining, so classic bitcoin need not bother with it at at all but miners huge enough to handle it (let it be really really huge) should be able to easily handle the classic chain as primary to merge with so they don't lose the benefit of all the hashing power being directed at the classic chain.

Maybe the Bitcoin plus Ripple thread I am about to read offers a better idea though...

-MarkM-

[ArticMine]

Would a "sorry this code is out of date" shutdown for nodes that fail to update in time be "a trainwreck" or just get such nodes out of the way so those that do upgrade in time can smoothly synch in whatever the new system turns out, by then, to be?

...
-MarkM-

This depends on the time-frame. One a six month time-frame a major this could lead to serious problems. One a two year time-frame it would likely not be an issue at all. Was there not recently a time-out on a delayed change that had been implemented by Satoshi that made certain old clients obsolete.

The key here is that delay could cause very serious problems because we will need to implement this with at least a 12 month delay or even better an 18 month delay or longer.

[solex]

Would a "sorry this code is out of date" shutdown for nodes that fail to update in time be "a trainwreck" or just get such nodes out of the way so those that do upgrade in time can smoothly synch in whatever the new system turns out, by then, to be?

...
-MarkM-

This depends on the time-frame. One a six month time-frame a major this could lead to serious problems. One a two year time-frame it would likely not be an issue at all. Was there not recently a time-out on a delayed change that had been implemented by Satoshi that made certain old clients obsolete.

The key here is that delay could cause very serious problems because we will need to implement this with at least a 12 month delay or even better an 18 month delay or longer.

When the blue line touches 345,000 the cow-catcher on Bitcoin's train connects with the buffers...

https://blockchain.info/charts/n-transactions?showDataPoints=false&show_header=true&daysAverageString=7&timespan=all&scale=1&address=

[markm]

When the blue line touches 345,000 the cow-catcher on Bitcoin's train connects with the buffers...

https://blockchain.info/charts/n-transactions?showDataPoints=false&show_header=true&daysAverageString=7&timespan=all&scale=1&address=

So not until then will we see whether transactions are so worthless that the only application that can afford to pay for them is an application that amounts to a "throw away money" game? No legitimate use can possibly make better use of the space we already have than Satoshi Dice does?

If that is the case we seem to have been throwing away money all along on building the worlds most useless, albeit most powerful, distributed system, thus maybe should rather shut the experiment down than throw even more money at providing unwanted blockchain space that transactors (other than those in the business of throwing away other people's money) do not want.

-MarkM-