Re: Open Timestamps

good point "waiting next blocks"


2016년 9월 30일 (금) 오전 10:17, Peter Todd <pete@petertodd.org>님이 작성:

> On Thu, Sep 29, 2016 at 08:48:19PM -0400, Wayne Vaughan wrote:
> > I'm with Peter.
> >
> > With Chainpoint, we purposefully avoid using timestamping as part of our
> > vocabulary.  We consider Chainpoint a proof protocol. You are proving
> that
> > your data is cryptographically linked to a set of external sources.  If
> one
> > of those external sources provides a reliable source of time, then you
> can
> > use the proof as a timestamp. The key constraint to recognize is that you
> > inherit time from the anchor sources.
>
> Eh, I have no issues with saying that OpenTimestamps does timestamping. The
> trick is simply making sure you communicate what is being proved correctly.
>
> Forward dating timestamps appropriately seems find by me. Basically, as
> applied
> to a git-commit timestamped by multiple notaries that might look like:
>
>     $ git log --show-signature 6a1c61a8dfbbcca672858500fcd11ecfeb34d8cc
>     commit 6a1c61a8dfbbcca672858500fcd11ecfeb34d8cc
>     ots: Success! Bitcoin attests commit existed as of Mon Sep 26 05:25:10
> 2016 EDT (99% prob)
>     ots: Success! Ethereum attests commit existed as of Mon Sep 26
> 06:12:01 2016 EDT (99% prob)
>     ots: Success! Litecoin attests commit existed as of Mon Sep 26
> 04:45:23 2016 EDT (99% prob)
>     ots: Success! Google (via Roughtime) attests commit existed as of Mon
> Sep 26 01:28:00 2016 EDT
>     ots: Good timestamp
>     gpg: Signature made Mon 26 Sep 2016 01:28:08 AM EDT
>     gpg:                using RSA key 6399011044E8AFB2
>     gpg: Good signature from "Peter Todd <pete@petertodd.org>"
>     gpg:                 aka "[jpeg image of size 5220]"
>     Author: Peter Todd <pete@petertodd.org>
>     Date:   Mon Sep 26 01:27:51 2016 -0400
>
>         Improve error message when ./ots stamp encounters IO errors
>
> They're all clearly statements that some notary attests the commit was in
> existance as of some time - not attestations as to when the commit was
> created.
> Equally, the fact that the Roughtime attestation is a few seconds prior to
> the
> actual Git commit isn't unexpected: we shouldn't be surprised if a user's
> inaccurate clock means the PGP signature was created a few seconds off.
>
>
> Anyway, all this discussion about Bitcoin's innaccuracy misses another
> point:
> you can't get a timestamp committed by Bitcoin instantly anyway! You're
> always
> going to end up waiting until the next block, and that takes a few minutes
> most
> of the time, often an hour.
>
> --
> https://petertodd.org 'peter'[:-1]@petertodd.org
>