To me the chain with the lower operational cost (all other things being equal) is better. And that chain is the one that is more efficient.
Amazing how people STILL, after 10 pages of the thread, are missing the point of
the article in the OP.
There is no "lower" operational costs regardless of the security model!
Security costs will always rise to the level of the rewards being
given, due to competition.
erm.. Actually no.. How do security costs rise in a pos system when rewards are tx fees? Did security costs rise for nxt when the asset exchange came out and transactional fees doubled or quadrupled and in turn rewards doubled or quadrupled? No they didn't so your theory that securty costs rise as rewards rise is wrong..
Because people will game the system -- they will create alternate chains to try to earn more fees.
Why WOULDN'T they, as long as there's money to be made?
well why haven't we seen any increase of multiple nxt forks or any at all for that matter during the period of increased reward? Why? You say why wouldn't they.. But yet no one has done that.. Pretty much of proves your statement wrong..
And even if someone forged on multiple chains, only one is valid so they only get the fees from the valid chain because the rest won't be accepted. they won't earn any extra than if they only forge on the valid chain.. Your theory is pure tripe.
And your response does not address my disagreement with your statement that security costs rise as rewards rise regardless of security model.
Did security costs rise for nxt when fees per block increased? And if your answer is yes, where did the extra cost come from?
Someone forging on a fork does not get included in the security cost for securing the main chain of nxt simply because they are not securing the main chain. They are securing a fork. So even if your claim that people will game the system was true, that still doesn't increase the cost of securing nxt. Your trying to make an argument but it just doesn't stand under further scrutiny.