Why OpenTimestamps does not "linearize" its transactions [combined summary]



Individual post summaries: Click here to read the original discussion on the lightning-dev mailing list

Published on: 2022-06-14T11:39:39+00:00


Summary:

The Opentimestamps system has been designed to be broken, but there is room for improvement. Jeremy Rubin proposes a model for timestamping services that he believes is necessary for reliability. However, Peter Todd disagrees with the linearization approach and argues that timestamps only prove the existence of a message prior to a certain time.The OpenTimestamps service uses commitment operations on an initial message to create a message known to have been created at some point in time, such as a Bitcoin block header. The technical details of this process are not relevant to the validity of the proof. Todd suggests that random beacons can provide a solution for dual-sided bounds on when messages were created.Trusted alternatives like the NIST random beacon or Roughtime can be used for this purpose. Todd also notes that OpenTimestamps could offer a trustless relative random beacon service using per-second commitments in a merkle mountain range. However, involving transactions in any random beacon approach would be pointless.Overall, while the Opentimestamps system is experienced at defending itself, there is potential for improving timestamping services by considering alternative approaches such as random beacons.


Updated on: 2023-08-01T00:33:01.235078+00:00