DLT Interoperability and More ⛓️#6 ⛓️ — SPECIAL EDITION: a peek into BUNGEE
This is a special edition covering a preview of our next paper, BUNGEE.
➡️ Title: Is My Perspective Better Than Yours? Blockchain Interoperability with Views
➡️ Authors: Rafael Belchior, Limaris Torres, Jonas Pfannschmid, André Vasconcelos, Miguel Correia
In this edition, we will explain what are blockchain views and why they are important.
➡️ Background:
- “A problem naturally emerges from a multi-chain ecosystem: since participants might have different views of a chain, how to have a consistent view of it, from the perspective of a third-party blockchain, that we want to interoperate with?”
There’s a lot to digest here. Some blockchains, namely private, offer partial consistency, to provide privacy-preserving functionalities. For this reason, its users may have different views of the same ledger. In the cross-chain scenario, even if ledgers A and B are public, the joint vision of A+B do not necessarily have a single view. The blockchain view is the artifact that allows us to reason about this composability and heterogeneity.
- “Creating and integrating views allows interesting applications, such as stakeholder-centric snapshots for audits, cross-chain analysis, blockchain migration, and data analytics.”
There are multiple applications of a view. As stated in our paper,
“building and analyzing views is important to accurately understand each stakeholder’s view of each DLT at every moment (including in public blockchains [44]) as a tool for business intelligence (e.g., understanding better a certain protocol) and auditing (e.g., monitoring a protocol). Views directly support blockchain interoperability since it would now be easier to share the perspectives of all participants across heterogeneous DLTs [6, 42], allowing for a better representation of the business ecosystem. This could enable complex orchestration of cross-blockchain services and support the new research areas of DLT interoperability, including blockchain gateway-based interoperability [11, 8, 13, 26].”
We provide a practical example of the utility of views in section 3.1 of the paper. A peak:
In this use case, there are multiple non-trusting entities that have different views on a single ledger. The auditor has a global view of the ledger (domains 1+2+3). Other stakeholders are paired (e.g., Wholesaler and Retailer share view on domain 2).
This scenario originates five views (plus the auditor), that could translate into the following:
➡️ Contributions:
- “This paper introduces BUNGEE (Blockchain UNifier view GEnErator), the first DLT view generator, to allow capturing DLT snapshots, constructing views, and performing arbitrary operations on those, such as integrating views.”
➡️ What is a blockchain view:
We mathematically formalize all concepts surrounding the blockchain view in Section 3.2 Blockchain View Integration Framework.
Formally, a
“A DLT view corresponds to the sets of different values referring to the same key that a participant can access”, and, in practice, is created from a snapshot. It is basically a filtered (projection function) snapshot of the ledger, according to the perspective of a stakeholder, that comes with a set of proofs (e.g., validator signatures over the state).
➡️ How to generate a view:
We provide a system design, which we call BUNGEE, that generates and merges views.
First, BUNGEE creates a snapshot according to a participant.
After that, the view is created. While all the details are in the paper, we specify an algorithm to build a view:
Optionally, views can be merged. There are different ways to merge views. One of the presented ones is merging and pruning. Merging views are based on yet another concept, the extended state. An extended state holds all values for each state key, for all views to be merged. We then iterate all the extended states and apply a bespoke algorithm, in this case, the prunning:
We finalize the paper with a discussion on key questions:
- Do views provide integrity and accountability guarantees?
2. Can one prove facts with a blockchain view?
3. Are views suitable for representing on-chain data from different sources?
We believe that this work paves the way for a more robust representation of the cross-chain state.
🚀 How does it relate to our work at Técnico Lisboa, INESC-ID, and Blockdaemon? (views are my own and do not necessarily reflect the opinions of my employer)
- Enabling secure, scalable interoperability is an important part of what we do at Blockdaemon. Studying formalisms to represent cross-chain state is a basilar step to advance current knowledge.
🚀 What are the implications for our work?
- Given an efficient, standardized to represent the cross-chain state, cross-chain use cases such as visualization of the thecross-chain state, blockchain migration, and asset transfers will be simplified.
