Skip to content
Permalink
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
116 lines (87 sloc) 4.29 KB

Towards a User Interface for Selectively Revealing Fine-Grained Personal Information to Transaction Counterparties

by Ryan Grant <rgrant@contract.design>, for Web-of-Trust Design Workshop (fall2016)

In a world where identity is carried with dignity by the participants themselves, built through a stored history of social interactions, revealed (or proven) with the fine granularity necessary to respect privacy and the security only it can afford, choices multiply. In this rich sea of choices, we can now ask how to interface a human with the marketplace, engaged to find the best deal for the least security risk.

Domains to Consider

The user-interface questions involved may lead us through domains as diverse as product logistics, medical assistance, the ongoing governance of human movement, education and caretaking, financial deals, billing for infrastructure, insurance, peripheral device key management, task and sharing contracts, the seemingly simple transmission of messages, and more.

The Three Standards

Along the way to enabling fine-grained identity capabilities in client software, standardization will be required in three key areas:

  • feature negotiation;
  • domain-specific namespaces; and
  • transaction dependencies (including templates).

This is in addition to architectural components less central to the concern of user interface, such as issuers, repositories, inspectors, holders, public ledgers, and root storage (terms mostly borrowed from the list in "Self-Sovereign Identity Architecture" by Manu Sporny and David Longley).

To successfully deliver products enabling this vision, implementations must also develop:

  • a protocol of questions, usually asked by a computer to an authorizing human; and
  • clear test scenarios.

Two Test Scenarios in fully collateralized lending

Of the many domains listed above, here are two financial contracts with a slight difference, intended to expose some user interface issues and spark discussion.

Collateralized Zero Coupon Bond #1

Description of terms:

A zero coupon bond, with values measured in USD but transacted in BTC. Final payment in-kind (as BTC). borrower will lock up collateral of 3x BTC into multisig contract. Price oracle required as one signing party, along with at least one of the other parties. Borrower receives BTC worth "discounted price" at bond initiation, and pays about 9% interest. Deal initiates as an atomic transaction.

    Face Value           : $10,000
    Discounted Price     : $8,000
    Term in Months       : 24
    Value of Collateral  : $24,000

Wallet user interface requirements (borrower and lender):

  • generate address for multisig payout
  • hold complete transaction until bond matures
  • alert user when bond matures
  • review price oracle's reputation
  • request counterparty help revoke price oracle authorization if it fails
  • sign complete transaction, with price oracle

Collateralized Zero Coupon Bond #2

Description of terms:

Exact same numbers and terms as above, except borrower will receive USD at bond initiation, and lender will send fiat money through a standard escrow house. Once the bond is initiated, the work of the escrow house is completed. The bond's collateral is held on the Bitcoin blockchain as in the prior example.

Additional wallet user interface requirements (borrower and lender):

  • store terms of contract with escrow house
  • sign escrow contract
  • review escrow house AML/KYC requests
  • send AML/KYC information to escrow house
  • track bond's bitcoin collateral through escrow stages
  • review escrow house determination that USD has transferred
  • pay invoice to escrow house

Notes on Direction

Comparing the two scenarios above, we observe one consequence of building out Self-Sovereign Identity tools is increased awareness about the price of security. Participants will be able to quickly reject deals containing too much risk of harm from misused identity, for too little reward.

The examples we develop further can help guide discussion about how to solve the three standardization challenges:

  • feature negotiation;
  • domain-specific namespaces; and
  • transaction dependencies.
You can’t perform that action at this time.