Case in point: a number of years ago I knew a kid who was smart enough to flash Tomato on his router, enable SSH and even install a bunch of Entware packages. But he wasn't intelligent enough to change the SSH port from 22 or leave the remote access disabled.

Fast forward a month or two and his ISP tells him that they traced some pretty serious botnet shenanigans to his IP.

Just because someone is smart enough to use a device doesn't necessarily mean they're intelligent enough to use it safely.

That’s just good password security and reasonable.

Yes, that's my point, you don't need to disable it by default.

See that qualifying word there? “Most”? That’s why they force SSH to be disabled and password changes. If you PERSONALLY can guarantee that no one will EVER put a freshly imaged RPi directly on the internet backed by a 10 million dollar/pound/euro guarantee per incident it still doesn’t matter; there’s still a need to change these defaults. I’ve seen the RPi’s deployed in a business environment and I 10000% know that vendors are fscking stupid and would leave default permissions enabled because they’re the lowest bidder.

There are those things called licenses and liability liability waivers that are signed specially for those cases. The people doing deployments on business environment should know how to change password / use SSH keys and whatnot, if they don't that's not the Pi's problem.

It’s people like you why we have massive botnets due to default security measures being ignored by major manufacturers.

By enabling people who shouldn't be configuring Pi boards in the first place you're are the one creating botnets. They might be saved by the fact that it doesn't have SSH enabled by default just to be hacked later on when they decide to run a sudo wget ... | sh.

Making things easier has this downside, you protect people so much, they don't ever learn and then things go bad they can't handle it and the damage is way way worse.