Interaction State Machine

ISM — Bringing Contextual Compute to the World

9 min readJun 7, 2023
Abstract idea of a Conceptual Compute Machine

Emergence of the Digital Society & it’s needs

The emerging digitally interacting world is evident all around us. We no longer use the internet to simply send emails or search for information. We are beginning to use digital networks to sate the most fundamental of Maslow’s hierarchy of needs: learning, eating, socializing, business processes, regulatory governance, money management — all are conducted within a digitally interacting world. More than simply connected by technology, we now operate by technology.

In this new model, our digital interactions go beyond simply sharing information. We now share a wide range of personal values in a more human-like manner. However, transferring values is different from sharing information. It involves being adaptable to various factors like trust, assets, storage, computing power, and user control, all tailored to the specific interaction and participant preferences.

However, the decentralized protocols of today are inadequate in meeting this requirement, as they rely on a systemic and programmatic approach for state management that is disconnected from participants and their preferences. This approach has resulted in a complex multi-layered state management solution with an incoherent end-state that is leading to a trade-off between security, scalability, and sustainability in web3 networks.

To foster sustainable, practical, fair, and fulfilling interactions in the emerging digitally interacting society, it is crucial for Web3 to undergo a fundamental transformation that enables human-like digital interactions. This can only be accomplished through a state management technology that is fundamentally centered around participants and operates in a fully peer-to-peer manner.

In order to facilitate this transformation of Web3, a groundbreaking technology called the Interaction State Machine (ISM) has been developed. ISM introduces the concept of contextual compute, enabling implementation of digital interactions that closely resemble human interactions. In this article, we will delve into the core principles and applications of ISM, and examine its potential to reshape the future of decentralized networks and digital interactions.

State of today’s p2p networks

Over the past decade, the emergence of p2p networks, exemplified by Bitcoin, has started to impact how we communicate and conduct activities in the digital world. While Bitcoin demonstrated the power of decentralization through a p2p value transfer model, it is not general purpose (restricted to one crypto currency) and the network state agreement is linear and non-concurrent. Over time, this has given rise to significant flexibility, scalability, and sustainability challenges.

Since Bitcoin, several efforts have been made to develop distributed technologies capable of supporting multiple cryptocurrencies and assets in a general-purpose fashion on the same network. The most popular of these is the smart-contract model. Using a programmable state management paradigm, the smart-contract model has made great strides in this area, but it has also introduced significant inefficiencies and compromises. Specifically, the smart-contract based app-centric state management paradigm has shifted away from the true p2p value transfer model of Bitcoin, the fundamental premise of decentralization, leading to new security and implementation challenges.

Furthermore, existing distributed systems still rely on a global, linear, and non-concurrent network state agreement model, posing significant scalability issues. The answer to-date to this challenge has been to reduce the problem set and create smaller and smaller networks. While this engineering fix temporarily addresses the challenge, multiple layers of complex technology is starting to become unwieldy for adoption and more importantly compromises security and decentralization. Under the current model, as digitally interacting society progresses, the problem of scaling and securing billions of heterogeneous p2p value transfers will only become exponentially worse.

ISM addresses this challenge by embracing a fundamentally new computing paradigm that places the participant at the core of value transfers.

Interaction State Machine (ISM)

ISM is a groundbreaking state machine replication technology purpose-built for p2p networks, with a focus on supporting the needs of our digitally interacting society. With the concept of contextual compute, ISM introduces participants into the realm of computing, enabling human-like digital interactions in a truly peer-to-peer fashion.

ISM is a new state machine replication technology for distributed systems where all aspects of a value transfer — data management, execution, and consensus — are designed to be peer-to-peer.

Unlike today’s distributed systems, ISM associates the state of an object with a participant and their state management preferences across the entire network. This facilitates true peer-to-peer and interaction-centric computing in digital networks.

Components of Interaction State Machine

With ISM, we will be able to build flat million-node distributed networks with hyper-scalability and security. With ISM, we will be able deliver a true general-purpose p2p value transfer system that is simple and sustainable for a digitally interacting world.

Simply put, ISM is a break-through technology to fulfill the original promise of Web3 — an Internet of Value that is truly decentralized and participant-centric.

Overall, ISM represents a significant shift in how computing is done, emphasizing the importance of participants and context in delivering a more user-friendly, efficient, and secure digital experience.

Context — A new computational dimension

The fundamental principle of ISM is based on the concept of “Context (π)” . Context is a groundbreaking computing element that puts the participant at the center of computation in ISM. Essentially, Context serves as the participant’s representative in the network. It adapts and expands as the participant engages with the network and with other participants.

Context is fundamentally different from Identity in today’s networks. Identity just enables recognition of a participant and is an attribute of computation. But Context allows an understanding of the participant through their behavior, preferences, and values and it enables the computation.

Identity is like your name, Context is you!

Context in real world

Context of a participant in an interaction can be

Network Behavior Context derived from activity of the participant in the network generated by the temporal entropy (e) of the participant which represents their change in network preferences and behavior over time.

Personalized Interaction Context setup through type of digital asset, interaction trust factor, multi-dimensional value algorithm, complexity of consensus, and computational/cryptographic/storage preferences.

Context Object (πo)

ContextObject is an autonomous object that establishes the relationship between a content and its context in an ISM network. It is comprised of two components: Content, which represents the value that must be managed in the network, and Context, which refers to the most recent context associated with the content. ContextObject serves as the core state management component in ISM.

Network-State (NS) and Network-Object (NO)

Using the idea of participants and context, ISM introduces two new distributed state management primitives: Network-State (NS) and Network-Object (NO). Network-State represents the network truth of an object that programs can manage across the entire network in a deterministic and agreeable way. ContextObjects manage Network-State, making them Network-Objects. This approach enables the network itself to become a stateful resource for objects managed by ISM, allowing for personalized and scalable interactions.


Balance -> Information

Alice’s Balance -> Program State

// recognized in a program object

Alice’s Network agreed Balance –> Network-State (NS)

// recognized in a Network-Object (NO) in ISM with Context

Alice’s perceived use of the Network agreed Balance in an Interaction — Value

// for Alice

Context Super State

ISM utilizes Context Super State as a composite state vector for handling state management in distributed networks. It represents the participant’s most recent state in ISM and includes all the context objects managed by the participant, as well as the systemic context objects possessed or owned by the participant.

KRAMA — Participant-Centric Ordering and Agreement Mechanism

KRAMA is a participant-centric ordering and agreement mechanism built on context. It serves as the foundational state agreement mechanism of ISM and is based on personal context clocks, eliminating the need for a network-wide system clock in a distributed system. This enables hyper-concurrency and infinite scalability in ISM based networks.

KRAMA: Participant centric ordering mechanism for distributed networks

How ISM works: Simple Steps

1. In ISM, Context embeds Network-State (NS) into Program objects, enabling ISM to create and manage programmable ContextObjects (πo).

2. Each ContextObject is recognized by a unique identifier, such as πBal for the Balance object, representing (Bal, Context of Bal).

For example, to retrieve Alice’s balance from the network:

3. In traditional distributed networks, the query would be {“Give me Alice’s Balance,”} which may not guarantee the latest synchronized state.

4. In ISM-based networks, the query transforms into {“Give me Alice’s Network Agreed Balance”} providing a precise answer in a single step due to the network-state embedded in ContextObjects.

ISM utilizes trust (Network Intelligence) in each context object, allowing it to work at a network level across multiple nodes, providing global finality as the default state management action.

Differences between traditional distributed networks and ISM based networks

Traditional distributed networks:

  • Trust is intermediated by the system, as represented by the nodes of the network
  • Objects are local program-level entities with no deterministic network state information associated with them.
  • Network optimization is not a primary goal.
  • Limited scalability and throughput

ISM based Networks:

  • Trust is intermediated by the participants of an interaction, as represented by their context
  • Objects are intrinsically network-level entities with embedded network state.
  • Network optimization is a goal, enabling infinite throughput and linear scalability.
  • Global finality is the default action, thanks to cryptographically verifiable consensus built into each object.

The creation of Network-Objects and Network-States using Context is a fundamental differentiator of ISM. Through ISM, we gain the ability to manipulate objects at the network level, leveraging the network itself as a fundamental resource for computation. This unique approach empowers us to achieve infinite scalability and hyper-concurrency in distributed state management, unlocking new possibilities for efficient and concurrent processing within p2p networks.

Usage: IFP/CP Protocol Suite

ISM is implemented and brought to life by IFP/CP, a new networking protocol suite specifically built for p2p value management in distributed networks. IFP/CP delivers flexibility, scalability, and interoperability to distributed value management, similar to how TCP/IP revolutionized information management.

Stateful Internet: Infrastructure for the Digitally Interacting world

In the IFP/CP protocol suite, IFP (Interaction Finality Protocol) and CP (Context Protocol) are two essential protocols that work together to enable stateful communication over networks, including the internet.

Context Protocol (CP):

CP is responsible for state connectivity in the IFP/CP model. It provides context addressability and the necessary information for transmitting Network-State across interconnected ISM networks. CP assigns and manages unique Context (π) for peers and Network-State (NS) for objects, allowing deterministic identification and access to their network intelligence. CP is analogous to the network layer in TCP/IP, but for distributed state management.

Interaction Finality Protocol (IFP):

IFP operates as the state management layer in the IFP/CP model. It provides deterministic finality for peer-to-peer value transfers with reliability and cryptographic verifiability. IFP uses relevant context for values, files, and logic in a granular fashion based on the object and interaction preferences to achieve true peer-to-peer consensus. It also updates the network-state of interaction participants (and objects) using context addressability, ensuring reliable delivery of state changes across the context, and guaranteeing interaction finality.

In summary, CP is responsible for the context addressability across ISM networks, whereas IFP ensures the reliable state and context management during a p2p Interaction. CP handles the network layer functions, whereas IFP operates at the value transfer layer and builds on the services provided by CP.


ISM’s innovative approach is a significant step forward in the field of distributed systems and has the potential to revolutionize various industries, including finance, healthcare, and supply chain, by delivering human-centric networks that are simple, secure, and personal.

With the integration of participants and context into computing, ISM has made the stateful Internet and global computer a practical reality.

With ISM, the vision of an equitable and sustainable digitally interacting world can be realized, transforming the way we engage with technology and each other.

MOI, the world’s first context-aware blockchain protocol, is based on the ISM technology. You can check it out at




Founder of MOI Protocol & CEO Sarva Labs Inc | Building a digitally co-operative world. Using technology to deliver shared prosperity .