People accept the risk of the theoretical >50% attack when they use PoW networks.
At the moment there is a similar theoretical risk of a 'Nothing at Stake' attack on a PoS network.
People accept the risk of what a >50% attacker could do to a PoW network if they were a non-rational(economically speaking) bad actor. So why do people dismiss PoS simply because it currently has the possibility of a theoretical non-rational attack?
Seems very contradictory to me.
Except it's not theoretical, and it's not non-rational. Had you bothered reading the links I posted you would have learned that Peercoin was successfully attacked this way.
You don't even need to actually own any coins to attack a PoS coin. You can use coins you don't own anymore, meaning you can't even use the argument "Why would a stakeholder want to destroy the value of his own stake?".
Peercoin isn't relevant at this point and PoW coins have been attacked before too so the same logic applies.
From what I understand from the V. Buterin and Come-From-Beyond discussion on the NXT forum there is a solution for this problem anyway.
Also, if you think that someone attacking the a PoS network is significantly more likely than someone mounting a >50% attack on a PoW network then that's fine as well and I'm not making an argument otherwise.
I just wanted to point out the apparent double standard of some people who consider the chance of a >50% attack not sufficient to warrant the dismissal of PoW as valid method while dismissing PoS due to a similar weakness at this time.