Eve,

 

Interestingly in the IRM WG we are on this topic.  We are about to publish design principles and the conclusion is about how to merge the work.  Maybe we could have a join group call to discuss?  I can send around a final draft of the doc in the next week.

 

Sal

 

From: dg-bsc-bounces@kantarainitiative.org [mailto:dg-bsc-bounces@kantarainitiative.org] On Behalf Of Eve Maler
Sent: Tuesday, November 29, 2016 11:36 AM
To: dg-bsc@kantarainitiative.org
Subject: [DG-BSC] Notes from BSC telecon Tuesday November 29

 

http://kantarainitiative.org/confluence/display/BSC/2016-11+%28November+2016%29+Meetings#id-2016-11(November2016)Meetings-Tuesday,November29

Agenda:

  • Discuss end-stage recommendations

Attending: Thomas, Eve, John W, Eve, John M, Jim

Next steps: What would a follow-on group work on? DG or WG? You'd normally expect a WG to deliver technical specs. Existing group (such as the IRM WG) or new group?

Blockchain has a challenge in that the "center is floating" but it still has one (a silo?). While the name of smart contracts is odd, the concept of joining identities in a legal sense to the code is valuable.

Jim suggests that if we have an identity (that is, start from the identity-centric view), then we necessarily have semantics. The legal layer of transactional semantics is "prose" (as R3 calls it), which allows incomplete statements that can gain clarity and interpretation over time. "Code", on the other hand, must be complete in some sense immediately (modulo machine learning inference stuff). We're saying "an identity" meaning a single digital identity; all identities meet in the meat! (See this...) In legal-land, think of "persons".

(sync (relationship (identity (person (meat)))))

------------------------------------------------

(sync (identities (relationship)))

(See the much nicer diagram JohnW sent to the list!)

Do we think the lowest-hanging fruit around our recommendations would be around the identity/smart contracts elements? Could the relationship elements be captured in a smart contract itself? The incompleteness of any contract means there are assumptions that lean on a legal framework. ("Can smart contracts be legally binding contracts?" In short, yes, if you reference both prose and code.)

Use cases that are real for many today:

  • Smart home
    • Entities with identities, and their identities
      • Homeowner (primary), family member, AirBnB renters
      • Smart home devices
    • Contract conditions
      • Renters contracted through AirBnB are allowed to access the specified functions of the smart home devices, and only for a limited time
      • Family members can access the specified functions of the smart home devices until access is revoked
  • Medical/telematics
    • Entities with identities, and their identities
      • Patient (primary), doctors, nurses (care team), administrative staff
      • Insulin pump, other devices
    • Contract conditions
      • Whoever is a member of the care team can control and access the insulin pump settings
        • Substitute nurses etc.
  • Proxy powers
    • Entities with identities, and their identities
      • Mother, toddler, nanny
    • Contract conditions
      • The nanny can legally pick up the toddler from daycare and drive him around town

Looking at PSD2, the bank account is becoming the fundamental unit of identity. Actually, it sort of always was. The notion of an "identity account" came from more monetary notions of accounts.

Proposal: Add identity to the prose+code recommendations already extant for smart contracts. Jim sent this link to a CommonAccord contract.

AI: Jim: Send the Norton Rose paper to the list so we can bash it.

Eve Maler
ForgeRock Office of the CTO | VP Innovation & Emerging Technology
Cell +1 425.345.6756 | Skype: xmlgrrl | Twitter: @xmlgrrl