7.7 KiB

Raw Blame History

Understanding the Logos Blockchain Whitepaper

Title: Understanding the Logos Blockchain Whitepaper | A High-Level Conceptual Overview/Review of the Logos Blockchain

Purpose: This document reflects an approach of whitepaper review that combines a rough-draft whitepaper review containing comments/reflections/concerns with a further simplification of concepts to make everything more digestible to more people outside of the internal Status infra team for communication purposes. In addition, this review will provide a gauge of difficulty of understanding involved with various different terms/concepts.

Tags: #learning #whitepaper #Logos

**"We build everything on top of consensus." ** Dr. Corey Petty

How-To (Get Cookin')

There are some descriptions below which detail the usage of visual guidance metrics available in the form of "cookbook-like" instructions for how to consume this document:

Gauges	Description	Visualization
Difficulty	1-10 (Ticks/Pips)	[\|\|\|\|\|\|\|\|\|\|]
Time	In Minutes	X Minutes
Resources	Article/Video Links w/ Visual Content	Conceptual & Pictoral "Ingredients List" of Links

See below 'Recipe' and modify accordingly to above in Figma.

Operating Definitions: Easiness/Difficulty is defined as the ease of which the concepts can be generally understood.

Time/Scope is in reference to the size of information necessary to learn to gain a decent level of comprehension.

Resources is for links relevant to helping people understand the concepts.

Necessary Terminology by Layer

The Six (6) Technology Layers of Logos as a Blockchain

Terminology Template: here

Consensus

Node Reputation

Ikingut (Reputation Algorithm)
- Important Conceptual Goals
  - Simple
  - Lightweight
  - Pluggable
  - Adaptive
  - Dynamic
  - Robust
  - No Transitive Trust
  - Reasonable Bootstrap Time
- Reputation Polling (Polling Dynamics)
  - Indirect Request
  - Direct Request
    - Unirep (an example)
    - Qualities Necessary
      - Requires Verifiability
      - Requires Non-Repudiation
      - Privacy Preserving (preserve origins of score, emit opinions without a way to trace back the origins - MPC)
  - Local Heuristic
- Algorithm
  - Each Iteration
    1. Voting
    2. Agent Action (Post-Consensus Decision Finality)
  - Min-Multiplicative Reputation Punishment
    - Multiplicative
    - Linear
- Experimental Research
  - Starting Point
    - Trust Wisdom per Node (requires further elaboration)
    - Adversary Types and Effects (requires further elaboration)
  - Current Stage of Testing/Challenges
    - Silent-Omniscient Adversaries
    - Modulating Punishment/No Punishment Impacts to Conditions
    - Adding Multiplicative-min Punishment
    - Attacks
      - Con-Artist Attack
      - The On-Off Attack
    - Effects on Glacier Consensus
    - Limitations
      - No immediate defense against coordinated attacks
      - Reputation does not add to security
  - Future Work
    - Circumstantial Impact of Reputation on Consensus
    - Long-running Simulation
    - Sudden changes in Collective Byzantine Behavior
    - Pending Questions
      1. Interaction of Stake-based and Reputation-based selection
        
        Stake simulations are necessary for exploring options
        
        Relevant in the incentives discussion?
      2. How much (and if) does reputation really help in a coordinated attack? (Assuming patient con-artist attack)
        
        Complex interactions here, this model would particularly benefit having a prototype/PoC
Node Challenges
Design
Eigentrust
- Transitive Trust
- Understanding the Math
- Trust Decay
  - Malicious Clusters (describe nuances better | trusted nodes cannot overlap the malicious collective)
- Trust for Consensus
- Separation of Reputation
- Confidant
- XRep
- P-Grid
- R2Trust
Generic Taxonomy
- Dimensions
  - #Single
  - Multiple
- Time computation
- Aggregation
  - Deterministic sum of positive and negative ratings
  - Probabilistic
- Logic
  - Local vs gathered data
  - Age of data
  - Frequency of data
  - Weight of multiple dimensions
- Value Control
  - External
  - Internal
- Data Aging
  - None
  - Decay
  - Death of old/selected
- Selection
  - Ranking-based
  - Threshold (trusted/untrusted)
  - Probabilistic selection

Network Layer and Mempool

Node Discovery
Subnetworks
- Design
- Implementation
- Ideas
Mempool
- Challenges
- Approach
- Mempool Design

Staking and Multi-DAG

Factory of DAGs
- Concept
- Algorithm
Sub-DAG
Cross-DAG
Bridge (Intermediary) Nodes
Gravity
Application-Level User-Driven
Direct Communication Channel
Direct Communication VGER
Challenges
- Probabilistic finality and cross-network communication
- Network partitioning, forks and cross-network communication
- Weak Subjectivity - Reputation of nodes matters to clients. Nodes are clients of other networks.
- Staking: Verifying Weights
- Glacier Algorithm Quiesces
Approach
1. Deterministic Finality
2. Secondary Consensus Protocol (form local opinion on external DAG)
3. Intermediary DAG (Intersection of nodes participating mutually in common structures)
Vertex Sealing
XSub (Cross-sub-DAG communication)
XDAG (Cross-DAG communication)
XDAG Fees
Communication Patterns
- Direct
- User-Coordinated Channel
- Sub-DAG-Coordinated Channel
Stake Sub-DAG

Data Model and Concurrency

Reference Work
- eUTXO
  - References (Cardano foundational paper)
  - Direct extension of the basic UTXO model
  - Validator
  - Datum
  - Redeemer
  - Ledger Description
- CKB
  - Key feature: state space ownership
  - Cell description
  - Transaction description
- Zexe
  - Key Feature: Privacy
  - Transaction Description
  - Record Description
Approach
Data Model Design
- Challenges
- Concept
- Data Structures and Constraints
Concurrent Execution (Threads)
Execution Patterns
- Useful patterns
  - Example of EVM execution on this model
  - Explore and describe
- Anti-patterns
  - High Contention
  - Low Affinity

7.7 KiB Raw Blame History