wouldn't it be a whole lot easier for the script being curled to just detect it was running as part of a `curl | bash` pipeline and bail out with a message?

wouldn't it be a whole lot easier for the script being curled to just detect it was running as part of a `curl | bash` pipeline and bail out with a message? if [ ! -t 0 ]; then echo "script must be run interactively, etc."; exit 1; fi of course that's equally useless because as someone else said, the root problem with doing `curl | bash` in the first place is that you are putting the trust in someone else's server, so you can't fix that "server side". what the article is proposing is a whole lot of magic detect `curl | bash` from his safe server, where's it's already safe.