[Cryptography] Time(keeping) and Crypto-Economics

[Ray Dillinger]

On 07/09/2016 08:57 AM, g3upa9y4t9 at snkmail.com wrote:
> An comments will be helpful.
> 
> Time(keeping) and Crypto-Economics
> draft

FWIW, I've already published code that keeps block height
(multiplied by target block interval) roughly synchronized
with consensus time (time elapsed since genesis block as
measured by consensus time established by the timestamps
of recent blocks).

This would allow any script (in a script language able to
detect block height or query intervals in block height) to
calculate things like interest or contract performance
deadlines etc, and would keep events like block subsidy
reductions happening on a more predictable schedule
(ie, predictable with much greater precision than by
block height in a block chain without such regulation)
over the long term.

It is, obviously, just an approximation. IIRC, the numbers run
against simulation were something like "spends 99.9%+ of its
time within a week of clock-correct and altering difficulty
by 5% or less," and "spends 50%+ of its time within 4 hours
of clock-correct and altering difficulty by 0.22% or less."

It speeds or retards block creation by making small adjustments
to block difficulty (with adjustments capped at 10% for any
block height more than 2 weeks out of sync with the consensus
time).

It does this while simultaneously regulating difficulty to
keep block intervals consistent despite immediate addition
or reduction of hashing power (because it was developed for
altcoin chains, which have much higher hashing rate variance
than the bitcoin chain).

With the expectation of block-height-to-time adjustable for
the immediate periods (people know if the current block height
is off-sync and can calculate beginnings of contract periods
starting from it) and expectations of time as deadlines
approach visible in final periods (people know the actual
block height so they have warning of the rough variance in
deadlines), I think that kind of "light-touch" regulation
addresses most of the practical problems with block height
as a proxy for time.

Or at least it serves adequately for purposes of contracts
between humans who accept its roughness as part of their terms.
It also ought to be acceptable for purposes of events whose
precise timing is only of minor significance as opposed to
their effect as measured in months or years (such as regulation
of coin creation).

Code publicly available, usable without fee, modifiable and
redistributable under the terms of the MIT free-software
license.

http://dillingers.com/blog/2015/04/21/altcoin-difficulty-adjustment-with-midas/

It may not do *exactly* what you want but I think it's a
way to make block height into a decent if rough proxy for
wall (calendar) time.

				Bear





-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20160710/61dd12e2/attachment.sig>