> Proof of Work isn't the only consensus algorithm, there are ones with lower computational costs, such as Proof of Stake
PoS still doesn't exist in the real world, as far as I know.
> See IPFS, and there are other distributed storage solutions
IPFS has nothing to do with blockchains per se, though for better or worse (IMO for worse), the team working on it is also heavily involved with the cryptocurrency world. You might be thinking of Filecoin, which looks reasonable and had probably the only ICO so far that isn't total bullshit, but while I'm hopeful, we still need to see if and when it actually gets deployed.
> the main benefit is that smart contracts can eliminate the need for 3rd parties, but still provide trustworthy transaction settlements
Let's not forget about the most important drawback of "trustless" systems - machines know no mercy. You'd better be sure you've covered all potential problems in the present and the future, because if you make a single mistake in the code and lose all your money, there will be - by design - literally no one to help you.
(There's this idea of replacing laws with smart contracts that occasionally pops up on the Internet; it's a stupid and dangerous idea precisely because of that.)
* PoS was just an example - there are a range of consensus algorithms that are not proof of work, which was the very first one. Ripple is a "real world" blockchain that does not use proof of work: https://ripple.com/build/xrp-ledger-consensus-process/
> because if you make a single mistake in the code and lose all your money, there will be - by design - literally no one to help you.
Actually there are, by design, roll-back mechanisms that can be deployed if the majority of the network agrees to carry it out. I believe the current controversy in the Bitcoin network is partially over the removal of such a mechanism. Various other blockchains and blockchain-like networks have correction mechanisms. Also note, its not always about storing money - thats just one use.
PoS still doesn't exist in the real world, as far as I know.
> See IPFS, and there are other distributed storage solutions
IPFS has nothing to do with blockchains per se, though for better or worse (IMO for worse), the team working on it is also heavily involved with the cryptocurrency world. You might be thinking of Filecoin, which looks reasonable and had probably the only ICO so far that isn't total bullshit, but while I'm hopeful, we still need to see if and when it actually gets deployed.
> the main benefit is that smart contracts can eliminate the need for 3rd parties, but still provide trustworthy transaction settlements
Let's not forget about the most important drawback of "trustless" systems - machines know no mercy. You'd better be sure you've covered all potential problems in the present and the future, because if you make a single mistake in the code and lose all your money, there will be - by design - literally no one to help you.
(There's this idea of replacing laws with smart contracts that occasionally pops up on the Internet; it's a stupid and dangerous idea precisely because of that.)