The Fifth Elephant 2019

Gathering of 1000+ practitioners from the data ecosystem

Tickets

Verifiable Logs and DLT: A recipe for smashing UCC using hashing

Submitted by Raksha M P (@rakshamp) on Tuesday, 16 April 2019


Preview video

Session type: Lecture Session type: Short talk of 20 mins

Abstract

An “Unsolicited Commercial Communication”(UCC) means a commercial communication which a Subscriber opts not to receive.
For a long time, Telecom Regulatory Authority of India (TRAI) has been a centralized regulating body for any Commercial Communication and now has a mandated for a DLT based solution to mainly overcome problems of cost of regulation apart from establishing proofs

A DLT though helps in creating a decentralized distributed system of trust, doesn’t focus on privacy concerns of the emerging world. The Gartner Survey claims - “By 2021, 75 percent of public blockchains will suffer “privacy poisoning” - inserted personal data that renders the blockchain noncompliant with privacy laws”. [1] and Justice Srikrishna committee report on data protection claims that “Protection of personal data holds the key to empowerment, progress, and innovation.” [2].

Keeping the above claims in consideration, we are presenting a DLT based solution combined with the verifiable logs to regulate the current eco-system of Commercial Communication . The subscriber preferences and consents etc. are stored privately by the ecosystem participants as Merkle tree and Merkle proofs are published to the DLT to help in auditing and addressing complaints automatically . The solution implements the concepts of Signed Root Hashes of Merkle Trees and Sparse Merkle Trees, a distributed append-only data structure, to efficiently store the Consumer Preferences and Consent privately but still provides verifiability for the ecosystem participants.

Additionally, the solution implements APIs and services which are designed to provide an auditing mechanism for validating user preference stored/not stored(Proof of Inclusion and Exclusion) and message delivered, over a time frame along with a Reputation model, to implement an incentivization scheme which ensures discipline in Commercial Communication.

References:
[1] https://www.gartner.com/en/newsroom/press-releases/2018-10-16-gartner-unveils-top-predictions-for-it-organizations-and-users-in-2019-and-beyond
[2] http://meity.gov.in/writereaddata/files/Data_Protection_Committee_Report.pdf

Outline

  1. Overview of the problem and limitations in the solution in the current scenario.
  2. Approach for the solution incorporating Merkle tree as the data structure for validating information providing Merkle path hashes for audit and consistency proofs.
  3. Proposal for distributed append-only logs for tracking history on account of signed root hashes to validate the inclusion at an index in the tree and the entry of the root in the blockchain.
  4. Architecture specifying various components used as Independent registrars for handling various entries, AI for reputation model for the entities, Microservice based architecture.
  5. An On-demand mechanism for auditing.

Requirements

Basic Understanding of Decentralized applications and Blockchain.

Speaker bio

This session will be shared between Harihara Vinayakaram Natarajan, Raksha M P and Prachi Gupta working for Wipro Limited.
Harihara Vinayakaram Natarajan is a Distinguished member of technical staff of Wipro Limited and is a Software Professional with extensive experience(20+ years) with a strong recent focus on Distributed Systems and Machine Learning ( Distributed systems include Blockchain, Cloud, Automation (DevOps) , Log Analytics, Scalable Architectures ) and Big Data. Architected solutions for various domains including Retail, BFSI and Media. Raksha M P(Myself) and Prachi Gupta are blockchain developers working for Wipro CTO Office.

Links

Slides

https://www.slideshare.net/secret/lTYu6Ln1M1wIwA

Preview video

https://youtu.be/X003RCv63s8

Comments

  • Abhishek Balaji (@booleanbalaji) Reviewer 6 months ago

    Feedback from Rehearsal 1:

    • Initial slides are too dense. Make a decision about what would be suitable for the fifthel audience.
    • Add block diagrams and clear headings to explain the flow. Show detail when explaining a particular block.
    • Clarify some of the terms used or use simpler definitions.
    • Problems are not explained correctly - privacy, centralized solutions, etc.
    • Proof of inclusion is not explained clearly. Highlight the details and the flow, as it’s difficult to follow currently.
    • Justify with an example of the fake complaints problem
    • GDPR section needs to be expanded.
    • Explain why the current solution does not work, how has the legal system evolved abroad, how’s solution evolved with these developing constraints?
    • Explain the reputation part for content providers - How would their behavior be impacted based on the consent mechanisms?
    • Explain more of the conflict between service providers. - land grabbing
    • The work is very relatable for the audience, especially from a data engineering perspective, but needs work in organizing the flow of the presentation.
    • Define the prerequisites clearly so the audience knows what to expect.
    • What’s the story? Summarize the entire solution in 10 sentences and then structure your presentation.

    Next steps: incorporate the above feedback to weave a story with your talk – Use an example to run through the problem. You may choose to incorporate all or some feedback. The revised presentation, which has to be more detailed and fine tuned, needs to be uploaded to your proposal page in 10 days from now, i.e. 24 May. Based on the deck you upload, we will do an evaluation, and decide on whether you proposal is accepted to the conference.

    • Raksha M P (@rakshamp) Proposer 6 months ago

      Thank you for the feedback. Will update the revised presentation as early as possible.

  • Raksha M P (@rakshamp) Proposer 6 months ago

    Hello Abhishek. I have uploaded the updated slides for presentation as specified. Please find the attached.

  • Zainab Bawa (@zainabbawa) Reviewer 7 months ago

    Hello Raksha, The Fifth Elephant’s policy is one speaker per session. This is non-negotiable. You have to decided that if this proposal is shortlisted, who will present it for the rehearsal and review (as well as for the final session).

    • Raksha M P (@rakshamp) Proposer 7 months ago

      Hello Zainab, I(Raksha M P) will be presenting this session.

Login with Twitter or Google to leave a comment