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

[Ray Dillinger]

I think I agree with Mr. Baker that proof-of-work is problematic.

Aside from the issues he points out, Bitcoin in particular has an
additional problem in that its difficulty adjustment takes place in big
chunks and lags by two weeks of estimated time as measured in blocks.

In the worst case, if a high price has been sustained for two weeks,
the difficulty adjustment has been adapted for it, and then a price
crash of this magnitude cuts the price to 1/3 of its former level right
after a difficulty adjustment....

Then its already-inadequate transaction rate is cut from about 3 per
second down to about 1 per second, and stays there for six weeks.

I think his idea of a Notary Digest chain is not a bad one, and can be
used to resolve the short-term vulnerability of some scheme such as
TaPoS which might be more acceptable to the partisans. TaPoS is
completely Trustless and in the long run secure, but it works on a
cumulative total that can be manipulated in the short run, so it
otherwise doesn't work fast enough to pick between the very most recent
blocks.

				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/20180206/f773194f/attachment.sig>