[Cryptography] Proof of Work is the worst way to do a BlockChain

Ray Dillinger bear at sonic.net
Wed Feb 7 22:18:15 EST 2018



Notaries, Bankers, Lawyers, and Accountants, are very literally
professionals who people hire because they don't trust each other.  They
are in the trust business.

If they are widely enough untrusted, their businesses fail. So I don't
really have much of a problem with having them act in their usual
professional role here.  They have a sound motive not to fail.

Speaking of trust, there is the idea of a bond, and it's applicable too.

A thing allowed by the *idea* of proof-of-stake is that the miner
forming a block could use their stake to act as guarantor of the
transactions in the block.  This means the miner accepts some risk, and
gets paid for it, and it allows formation of blocks with limited
information.

And that means you could have parallel formation of blocks rather than
just serial formation of blocks - a block mesh instead of a block chain.
 Conflicting transactions might be entered.  When detected, the 'stake'
put up as guarantee would be used to make good immediately, and further
transactions making fiscal corrections then entered in later blocks.

Which is a whole lot messier than the nice clean elegant world of "that
was inconsistent so it didn't really happen" of classical block chains,
but allows forming blocks in parallel because now the position at which
the next block 'attaches' to the mesh doesn't have to hash the immediate
previous block.  And with parallel block formation you can go quite a
ways toward alleviating the block bandwidth problem.

	Bear

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20180207/a380876c/attachment.sig>


More information about the cryptography mailing list