if (blocknumber > 115000)
maxblocksize = largerlimit
In the original white paper, the goal is to have a consensus protocol for tamper resistant record keeping and accounting. Satoshi's proof of work using SHA-256 has now proven to be vulnerable to ASIC attacks and heavy centralized miner operations and mining pools.The arguments of semantics really miss the point of the intension to offer egalitarian access to currency production and account. Keeping the transaction rate low on purpose is counter productive by all measure, but several of the core devs work for blockstream which would stand to profit by selling their own sidechains to enterprises, counter to the first sentence of how Bitcoin is defined in the white paper.
A purely peer-to-peer version of electronic cash would
allow online payments to be sent directly from one party to
another without going through a financial institution.Breaking the ability of the network to follow a valid majority chain is breaking backward compatibility. It's disingenuous to claim otherwise simply because the forked version eventually ended up with the longest chain.
From https://github.com/bitcoin/bips/blob/master/bip-0050.mediawi...
What happened was the behavior became probabilistic based on the state of the machine it was running on and the how much block history was being rewritten due to a recent reorg, etc. Usually if the node was restarted, it synced and ran just fine.
The recovery didn't change the bitcoin consensus rules as seen by existing clients. There was just a temporary extra limit put in place (a soft-fork) to keep blocks small while the issue was sorted out, then lifted.
In the case of Bitcoin, it's not clear yet who it's going to keep the name at the next fork. You can see Coinbase blog post about this https://blog.coinbase.com/clarification-on-the-upcoming-segw...
Brands are words that need to be trademarked, but "a cow" does not need to be, because everyone knows what a cow is, and if I sell you what I say is a cow but is actually a cat and then explain that that's now the established brand - you'd say it's cat, not a cow.
The test for Bitcoin is very simple - fire up an old version of the Bitcoin software, and if it validates the blockchain, then the blockchain you have is Bitcoin. If it doesn't, it's not Bitcoin.
And if/when Bitcoin Core ever changes the software such that it is a hard, backwards incompatible fork, then on that day they should stop referring to it as Bitcoin, because it is not the Bitcoin that Satoshi originally started.
If you mean starting an older version of Bitcoin Core after the on-disk storage has been converted by a newer version, then yes, of course. The way data is stored has changed, not the least of changes being Leveldb instead of Berkleydb.
But at the network level newer nodes will communicate with older pre-Segwit nodes just fine.