[jkhdigital]

> tons of code that only increases the attack surface

Ethereum itself was the original sin in this respect. In Bitcoin, for the most part, the economic value at stake in a block is (1) the block reward, (2) the transaction fees, and (3) the amount transacted. (1) + (2) + (3) is therefore the maximum amount that can theoretically be "moved" if a block is invalidated through an adversarial fork, since someone else can claim the miner rewards and every transaction could possibly be double-spent. Anyone who wants to be assured of transaction immutability can, at the very least, calculate the total value that is at stake and wait until at least this much PoW has been layered on top of the transaction.

Not so with smart contracts. A paper I read yesterday showed how smart contracts can be used to execute trustless, crowd-sourced DoS attacks on rival blockchains (https://eprint.iacr.org/2019/775.pdf). It's ridiculous. Once you open the Pandora's Box of smart contracts, you can throw your consensus guarantees out the window.

[lostmsu]

> smart contracts can be used to execute trustless, crowd-sourced DoS attacks on rival blockchains

And that's a good thing. We needed an established way to attack rival blockchains that use the same PoW mechanism. They are insecure by design of PoW, and until they are all torn down by attacks, unsuspecting people can loose a lot of money.