Quote from: Peter R on November 14, 2015, 05:44:55 PM

...
Regarding "longest chain" versus "longest valid chain," I am using the language from Satoshi's white paper. He refers to it simply as "the longest chain" and only refers to transactions as being valid or invalid. From page 3:

- Nodes accept the block only if all transactions in it are valid and not already spent.

- Nodes express their acceptance of the block by working on creating the next block in the chain, using the hash of the accepted block as the previous hash.

- Nodes always consider the longest chain to be the correct one and will keep working on extending it.

Nice try, but the whitepaper was an early description focused on the basics of the system Satoshi was creating.

Here, half a decade later, we are discussing attempts to subvert the system by strong-arm tactics of breaking the protocol and manipulating total work ratios to accomplish it. That was well beyond the scope of Satoshi's whitepaper.

Quote from: Peter R on November 14, 2015, 05:44:55 PM

In other words, "the longest chain" means the chain with the greatest cumulative work that is composed exclusively of valid transactions.

There's that 'valid' word again.

To say the truth, I've always expected attackers to achieve an advantage in the potential for 'cumulative work' eventually. I'm surprised it has not happened yet, and very heartened because sha256 mining has lived long enough for a subordinate chains ecosystem to be viable.

This means several things:

- The owners of the hash hardware are under risk of losing their investment should the nuclear option be undertaken (that is, modification of the current POW hashing methods.) They will be most conducive to mounting a majority attack only if their investments are insured against this eventuality (which is quite possible) but it would not be terminal for Bitcoin anyway.

- A mostly successful cumulative work attack suddenly turns into something with is maybe not even half bad.

- Hodlers might be able to capitalize in it since funds would flow in to make it work

- Freeloaders would be drawn to the GoogBook fork and thus, away from core.

- A lot more focus would be put on hardening Bitcoin against a variety of attack types and much more focus would be put into exploring the network attack failure modes and how to thwart them.