DINRG Meeting at IETF-103

Tuesday, November 6, 2018, 9:00 - 11:00

• Agenda: https://datatracker.ietf.org/meeting/103/materials/agenda-103-dinrg-00
• Material: https://datatracker.ietf.org/meeting/103/session/dinrg
• Note take: Marie-Jose Montpetit

Welcome, agenda bashing

• https://datatracker.ietf.org/meeting/103/materials/slides-103-dinrg-agenda-and-chairs-slides-00
• DIN workshop at NDSS-2019 - talk to chairs if interested (and see below for workshop links)

Distributed Authenticated Mapping update (Jean-Luc Watson et al.)

• https://datatracker.ietf.org/doc/draft-watson-dinrg-delmap/
• https://datatracker.ietf.org/meeting/103/materials/slides-103-dinrg-distributed-delegated-mappings-update-00
• Update on the draft with new features.

Questions/comments:

• (Dave Oran): Is this based on voting majority
  • Answer (J.L.W): No
• (Melinda Shore): Where would you like that to go in the co
• (J.L.W.): We would like to implement this and if someone wants to work on it, it would be great.

Stellar Consensus Protocol (SCP) update (David Mazières)

• https://datatracker.ietf.org/doc/draft-mazieres-dinrg-scp/
• https://datatracker.ietf.org/meeting/103/materials/slides-103-dinrg-scp-update-00

• Presentation on the main features of the draft and:

  • a detailed description of the federated voting algorithm
  • the protocol phases

  Questions/comments:

  • (Dirk K.):Version 4 final?
    • (Answer): Not really ; need to fix a few things
  • (Melinda Shore): Is the multicast issue resolved?
    • (Answer): Not sure maybe some IETF work is relevant here;
    • if anyone knows something please contact the author ; want a byzantine
  • (Dino Farinacci): Need a reliable multicast?
    • (Answer): Not really but need to be reliable against eclipse attacks
    • (Dino Farinacci): UDP OK?
    • (Answer): Yes
• (Dirk): Multicast discussion is good to follow up on the mailing list

New Results on Liveness properties of SCP (Giuliano Losa)

• (relates to) https://datatracker.ietf.org/doc/draft-mazieres-dinrg-scp/
• https://datatracker.ietf.org/meeting/103/materials/slides-103-dinrg-on-the-liveness-properties-of-the-stellar-consensus-protocol-00
• Goal is to have a new consensus algorithm.

Questions/comments:

• (Melinda Shore): How is this standing? One or more drafts
  • (Answer - David Mazières): Map one protocol on the other; maybe an easier solution than SCP is better
  • Separate drafts? A better way of explaining SCP
• (Melinda Shore): Are you going to do modeling?
  • (Answer): Nothing ready to publish but yes.
• (Aanchal Malhotra): How is the leader nominated?
  • (Answer): Probabilistic algorithm; can describe off-line
• (Dave Oran): If you have a BFT agreement why do you need the rest if the BFT protocol could scale; Usually BFT does not scale; what is the concept here?
  • (Answer): They are synonyms for us.
  • (Dave Oran): Consensus is one semantic; Byzantine agreement protocol is not consensus
  • (David Mazières): Consensus is a bigger word; byzantine agreement is one solution
  • (Discussion to be continued off line)

The Decentralized Identifier in the DNS (Alexander Mayrhofer)

• https://datatracker.ietf.org/doc/draft-mayrhofer-did-dns/
• https://datatracker.ietf.org/meeting/103/materials/slides-103-dinrg-the-decentralized-identifier-did-in-the-dns-00
• Related work in W3C Credentials.

Questions/comments:

• (Melinda Shore) Status of the work in DNSops (?)
  • (Answer): Related
• (Dino Farinacci): Interoperability: blockchains are just fine
• (Dino Farinacci): Naming useful but dangerous as the opacity is a feature; should be application based not global
  • (Answer): Depends on the use case
• (Paul Hoffman): This is not a payment system - you would want it found from the Internet it should be found
  • (Answer): This is IPR related: it should be open not patented
• (Melinda Shore): Distributed applications: why new scheme and not part of the URI
  • (Answer): Not the right person; could be good; ask the W3C community

Decentralized Identity (Nathan Aw) (remote from Singapore)

• • no draft - see presentation
• https://datatracker.ietf.org/meeting/103/materials/slides-103-dinrg-decentralized-identity-01
• Review of the aspects of decentralized identity mainly from a banking point of view

Questions/comments:

• (Shigeya Suzuki) Last statement how to use TIemLock Contract
  • (Answer) In Singapore Central Bank system we are exploring hash time lock for transfer across platforms; could also help connote the different planforms together; single sign on across multiple islands
• (Dirk): Follow up on the mailing list

Economics of Decentralized Identity (Alastair Berg) (remote from Melbourne, Australia)

• • no draft - see presentation
• https://datatracker.ietf.org/meeting/103/materials/slides-103-dinrg-the-economics-of-self-sovereign-identity-00
• Follows from IAB discussions in September
• Decentralized identity from an economics point of you: identity as an economic good

Questions/comments:

• (Dragen Bogdanovic): Slide on self-sovereign identity: nice goals but how to extract the value when you are self governing
  • (Answer) Million dollar question - FB, iLinkedin, etc. are riding identity value
  • (Dragon: Governments also
  • (Answer) Yes - do not have a clean answer; as more and ore self-identity and regulations appear some solutions may emerge

Wrap-up

• 3 Topics to now:
  1. concepts of identity
  2. Infrastructure (SCP)
  3. application ideas
• Feedback?
  • please discuss on mailing list
• NDSS workshop information coming
  • CFP: https://www.ndss-symposium.org/
  • submissions: https://din19.hotcrp.com