[chclt]

Meh, being able to synchronise an append-only tamper proof linked list and control who gets to put what on it in a decentralized way is pretty cool. Not to mention it's far easier to implement with limited means, than 5nm chips, or large tansformer models.

(Also implementing money on it does seem pretty intelligent -- its a scarcity device -- peoble just seem to do stupid stuff with it)

[ninkendo]

Git (and any merkle tree) already implements a tamper proof linked list, there’s no reason to burn electricity in a proof of work algorithm to accomplish this.

Controlling who gets to put things in it in a decentralized way isn’t as important as many people think, at the end of the day almost all applications (other than currency) can just use a central authority to publish hashes.

[chclt]

> Git (and any merkle tree) already implements a tamper proof linked list, there’s no reason to burn electricity in a proof of work algorithm to accomplish this.

Thats why the word "and" is there in my sentence

[midmagico]

Pardon me while I git push -f..

[ninkendo]

You should probably reread git’s documentation if you think git push -f rewrites any commits. It doesn’t. It makes new commits and re-assigns the named branch to track the new commits you just pushed. It rewrites branch names, not commits.

It is impossible (without finding a SHA collision) to rewrite a commit in git.

And before you say “yeah but how do you know which SHA sums are the right one if you don’t trust branch names”, please reread the final paragraph of the post you replied to.