We're extremely excited to announce the release of a bunch of innovations we've been working on for the past two years!
More than a decade after Satoshi's major breakthrough in 2008, blockchain's main limitations and issues—scalability, interoperability, and privacy—remain largely unresolved and are keeping the industry adoption of the technology yet to be a reality.
Starting from the concept of a distributed platform for storing and exchanging value represented by a digital currency, numerous evolutions have shaped the blockchain that we know today:
Smart contracts paved the way for web3 (dapps).
Divide and conquer strategy (sharding) enabled greater scalability.
Novel Sybil control mechanisms such as PoS allowed for eco-friendly blockchains.
Cryptographic breakthrough enabled greater user privacy.
Yet, despite the energy and interest the technology is clearly capable of inducing in the global public community, the web that we use today still resides in vast majority within the realm of Web2.0 tech stacks, i.e., *client-server.* Most private companies are still serving their services and data from resources hosted on a handful of distributed yet very centralized cloud vendors.
So what is still holding blockchain's adoption back in late 2021?
The scalability problem differs whether we're considering permissionless or permissioned blockchains.
Permissionless blockchains are victim of their own success. More/better features make platforms more open and attractive to both developers and consumers. In a few years and after multiple major growths in attraction from the community, existing solutions have shown us how Web3.0 platforms end up greatly restrained by high message concurrency or high entry cost.
Permissioned blockchains have often been the answer for private companies. Research has shown that most business entities would rather create their own network and be in charge of their own fate than join an existing network. This eventually leads to islands of incompatible blockchain networks. Performance and control at the cost of closure you might say. This obviously is not a viable long-term solution in the industry world where business stems from the extensive use of APIs interfacing internal services with external ones.
Permissionless blockchains are trying to fix the scalability problem with different techniques based on *sharding*: a central blockchain network delegates non-core features to a set of secondary chains which have their own network and participating nodes. The way communication between secondary chains is conducted goes a long way to defining a key property of such multi-chain systems but we'll come to that in the next section.
Permissioned blockchains—as we've seen—need to rely on interfaces and APIs to allow for compatibility and connectivity with one another, as well as with legacy systems.
What both domains need is interoperability.
Permissioned blockchains need to rely on interfaces and APIs to allow for compatibility and connectivity with one another, as well as with legacy systems.
Interoperability is a complex challenge to tackle. We've mentioned the role of APIs as drivers of technological changes in our world: standards like Representational State Transfer (REST) have greatly helped reaching a global consensus on how such APIs should expose their endpoints and how data is to be retrieved, created, updated, and deleted.
The definition of such a standard for blockchain networks is now more than ever a requirement for the seamless interoperability that public and private blockchain networks need.
Today, multi-chain public networks are facing today different limitations:
Networks like Polkadot allow for trustless cross-chain communication by relying on a strong primitive like classic BFT algorithms on the primary layer. This leads to a number of secondary chains that is intrinsically bounded by protocol specifications thus it eventually delays the scalability problem but does not solve it.
Other networks such as Cosmos have opted for simpler cross-chain mechanisms at the cost of the trust that secondary networks must have in one another. This greatly deters private organizations from entering such systems to interface with other networks.
No current solution simply allows a blockchain network to remain unconstrained while choosing its underlying structure and to be interoperable with other networks in a trustless and truly scalable way.
Beyond the need for trustless interoperability, private entities often need to keep their data private. Looking back again at existing trustless multi-chain systems, the validity of a cross-chain transaction is always pegged to the transparency that these chains need to provide. The central blockchain—the primary layer—validates state transitions of secondary networks by having access to their entire state and history.
No private company is willing today to open their data layer in such a way. Thus only blockchain enthusiasts are building on top of such systems while other organizations continue working with their isolated blockchains or legacy systems.
The Topos solution
At Toposware, we've been working on a solution that makes no compromises on scalability, interoperability, and privacy.
We believe the future of blockchain lies in its adoption in the industry, hence are directly addressing this problem.
Today we are presenting Topos, a consensusless, trustless, privacy-enhancing interoperability protocol for permissionless and permissioned blockchains to seamlessly interact with one another in a trustless fashion that remains privacy-focused and whose scalability is unbounded by design.
Moving away from traditional consensus solutions, our ecosystem builds on top of a weaker primitive: the Topos reliable broadcast algorithm. Subnets—networks participating in the Topos ecosystem, e.g. public and private blockchains and legacy systems—exchange data and transactions via the Transmission Control Engine (TCE) which leverages the reliable broadcast primitive. The overall consistency of message passing in the Topos ecosystem is ensured by the TCE. The state transitions of subnets remain private and their validity is guaranteed by zkSTARKs proofs. Subnets expose relevant data by means of the Universal Certification Interface (UCI), a common interface that defines the core abstraction subnets need to abide by.
The end result is a fully asynchronous, more robust and more efficient transmission layer, than its consensus-based counterparts, supporting trustless and privacy-enhanced cross-subnet communication.
As part of our software DevKits, we are currently working on augmenting Parity's Substrate framework to allow new public initiatives and private companies on the verge of their transition to blockchain to quickly and surely start developing their own Topos-ready blockchain network.
What we release today
Today, we're proud to release the following:
Our brand new company website
Topos Docs, our technical documentation website
Our first academic paper: ICE-FROST, our contribution to the FROST threshold signatures scheme
Multiple open-sourced repositories:
frost: Our modification of the original FROST protocol to support additional Identifiable Cheating Entity property and Static group keys. Originally forked from frost-dalek.
cheetah: Our custom STARK-friendly elliptic curve constructed over a sextic extension of a small prime field with high two-adicity.
winterfell: A fork of novifinancial, modified for our specific needs.
certificate-stark: A custom AIR program for efficient Topos state transition verification.
hash: A start of a collection of algebraic hash functions for zero-knowledge proofs, currently offering two instantiations of Rescue-Prime over Cheetah’s small prime field.
schnorr-sig: A custom implementation of the Schnorr signature protocol, for efficient verification inside a STARK statement.
September 7th, 2022
Industry figureheads from Meta, Monex Group, and CEA-List laboratory will advise the company as it shapes new ...
March 16th, 2022
For Topos’s trustless and privacy-preserving cross-subnet communication, Toposware requires subnets to prove the validity of...
December 21st, 2021
In this post, we will give a brief introduction to reliable broadcast and how we use it in Topos. The first thing to mention about reliable...