Introduction

Neuromorphic computing is a cutting-edge approach to designing and building computer systems that draw inspiration from the structure and function of the human brain. The term “neuromorphic” comes from combining “neuro,” which refers to neurons and neural networks, and “morphic,” which means resembling or imitating. The goal of neuromorphic computing is to create hardware and software systems that can perform complex cognitive tasks with high efficiency and low power consumption, similar to how the brain operates.

Neuromorphic computers are inspired by the human brain. Just like the brain, neuromorphic computers are made up of many small units called neurons. Neurons are connected to each other by synapses. When a neuron fires, it sends a signal to the neurons that are connected to it. Imagine lots of tiny workers called “neurons” that talk to each other with special messages. These workers help the computer learn and remember things, like how we learn new stuff. The special thing is, this computer uses very little power and can do many things at once, just like how we can play and talk at the same time. It’s like having a mini-brain inside a computer that can do cool stuff!

How does neuromorphic computing work?

Here is a simplified overview of how neuromorphic computing works:

1. An input signal is received by an artificial neuron.
2. The artificial neuron amplifies the input signal and then passes it on to its synapses.
3. The synapses then pass the signal on to other artificial neurons.
4. The signal is propagated through the network of artificial neurons until it reaches an output neuron.
5. The output neuron then produces an output signal.

What are the benefits of neuromorphic computing?

Neuromorphic computing has several benefits over traditional computing. These benefits include:

• Low power consumption: Neuromorphic computing systems can consume significantly less power than traditional computing systems. This is because they can mimic the way that the human brain works, which is very energy efficient.
• High scalability: Neuromorphic computing systems can be scaled up to much larger sizes than traditional computing systems. This is because they can use the same basic building blocks (artificial neurons and synapses) to create larger and more complex systems.
• Ability to learn and think: Neuromorphic computing systems can learn and think in a way that is similar to humans. This is because they can mimic the way that the human brain learns and thinks.

What are the challenges of neuromorphic computing?

Neuromorphic computing is still a relatively new field, and there are several challenges that need to be addressed before it can be widely adopted. These challenges include:

• Designing artificial neurons and synapses: Designing artificial neurons and synapses that can accurately mimic the behaviour of their biological counterparts is a complex challenge. Artificial neurons and synapses need to be made from high-performance materials that can withstand the high currents and voltages that are used in neuromorphic computing systems. These materials are often expensive and difficult to work with.
• Creating large-scale neuromorphic computing systems: Creating large-scale neuromorphic computing systems that can perform complex tasks is another major challenge. Biological brains are incredibly complex structures. They contain billions of neurons and trillions of synapses. Also, neuromorphic computing systems need to be manufactured in a scalable way. This is necessary in order to produce large-scale neuromorphic computing systems that are affordable. With the current advancements, we can have much less artificial neurons for manufactured in a scalable way.
• Software development: Developing software for neuromorphic computing systems is also a challenge. This is because neuromorphic computing systems operate very differently from traditional computing systems. There are currently no standard software development tools and techniques for neuromorphic computing systems. Researchers at the Massachusetts Institute of Technology have developed a new software development framework called the Neural Engineering Framework (NEF). This framework provides a high-level abstraction for developing software for neuromorphic computing systems.

Which companies are working on neuromorphic computing?

There are several companies that are developing neuromorphic computers. Some of the most notable companies include:

• Intel: Intel has been developing neuromorphic computing technology for many years. The company is working on the Loihi neuromorphic chip, designed to mimic how the human brain works.

• IBM: IBM has also been working on neuromorphic computing technology for many years.

• BrainChip Holdings: BrainChip Holdings is an Australian company that develops neuromorphic computing technology. The company’s flagship product is the Akida neuromorphic chip, which is designed to be used in a wide range of applications, such as autonomous driving and medical imaging.

• Qualcomm: Qualcomm is a leading developer of mobile processors. The company has also been investing in neuromorphic computing technology. In 2020, Qualcomm announced the Neuromorphic Computing Platform, which is designed to help developers build neuromorphic applications. Recently, they announced partnership with Prophesee to bring neuromorphic computer vision to mobile phones

Conclusion

Neuromorphic computing is a promising new field with the potential to revolutionize the way that we compute. It has the potential to solve many of the challenges that are facing traditional computing, and it is being explored for applications in a wide range of fields.

Further Reading

1. Processing Units
2. A Review of Graphene-Based Memristive Neuromorphic Devices and Circuits
3. A Programming Framework for Neuromorphic Systems with Emerging Technologies
4. Neuromorphic Computing Guide
5. Opportunities for neuromorphic computing algorithms and applications.
6. Introduction to Neuromorphic Computing
7. Neuromorphic Computing and Engineering
8. Perspectives on Neuromorphic Computing

About the Author

Arpit is a seasoned CTO with vast experience in leading large cross-functional and cross-geography teams. Arpit is interested in early-stage investments in startups in sustainable fashion, clean energy, people tech(health & wellness, leisure, jobs & personal growth), web3, media tech/content creation, legal tech and ed-tech.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter

A Primer on Neuromorphic Computing was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

Introduction

The composable architecture(TCA) is a design pattern that emphasises the separation of concerns and allows developers to create complex systems by composing smaller, simpler components. The idea is to break down a complex system into smaller, self-contained parts that can be easily combined and recombined to create a larger system. This makes it easier to understand, test, and maintain the code.

Why is Composable Architecture Important?

Composable architecture is important because it allows one to create systems that are flexible and scalable. When we build a system using composable architecture, we can easily change or replace one component without affecting the rest of the system. This makes it easier to maintain your code over time, and it also makes it easier to add new features to your system.

How Does Composable Architecture Work?

Composable architecture works by breaking down a complex system into smaller, self-contained components. Each component has a specific responsibility, and it communicates with other components through a well-defined interface. This makes it easy to change one component without affecting the rest of the system.

Here’s a simple example of how composable architecture works. Let’s say you’re building an app that allows users to view and edit their profile. You could break down the app into two components: a profile view component and a profile editor component. The profile view component is responsible for displaying the user’s profile information, and the profile editor component is responsible for allowing the user to edit their profile information.

These two components can be combined to create the complete app. The profile view component communicates with the profile editor component through a well-defined interface, and the two components work together to provide the complete user experience.

More Complex Examples

Now let’s look at a more complex example to see how composable architecture can be applied in real-world scenarios.

E-Commerce:

Consider a complex e-commerce system that includes a catalog of products, a shopping cart, and a payment system. Each of these components can be broken down into smaller, self-contained components. For example, the catalog component could be broken down into a product search component, a product list component, and a product detail component. The shopping cart component could be broken down into a cart list component and a cart checkout component. And the payment system component could be broken down into a payment gateway component and a payment confirmation component.

Each of these smaller components can be combined and recombined to create the complete e-commerce system. The product search component communicates with the product list component, the product detail component communicates with the shopping cart component, and the payment gateway component communicates with the payment confirmation component. In this way, the system can be built by composing smaller, simpler components into a larger, more complex system.

In the context of e-commerce, this architecture pattern is also called composable e-commerce. There is another term PBC used in context of e-commerce applications. PBC stands for Packaged Business Capabilities — for instance, shopping cart and checkout, promotions are some PBCs.

In the software architecture terminology, there is another related term used — MACH Architecture which stands for Microservices, API-First, Cloud-Native, and Headless. We can say MACH is a cloud based composable architecture. In terms of PBCs, one or more microservices can be used to create a PBC.

App Development:

TCA can be implemented on the app development side too. In general, there will be five main components

• State: A type that describes the data your feature needs to perform its logic and render its UI.
• Action: A type that represents all of the actions that can happen in your feature, such as user actions, notifications, event sources and more.
• Reducer: A function that describes how to evolve the current state of the app to the next state given an action.
• Store: A place that UI observes for changes and where actions are sent. Based on these actions, reducers are executed.
• Environment: A type wrapping all dependencies of app/feature.

TCA and state design pattern: The ideas behind TCA framework are very close to the state design pattern. The MVI pattern in Android for building UI is also based on state machine. MVI (Model-View-Intent) — is a UDF architecture from the MV* family.

Drawbacks, Potential Solutions and Best Practices

Composable architecture has some potential drawbacks, including:

1.Complexity: Composable architectures can be more complex to design and implement than monolithic architectures, which can lead to increased development time and cost.

To manage complexity, it’s important to have clear standards for communication and APIs between services, as well as good documentation and testing to ensure that services are working correctly together. Additionally, using microservices frameworks and tools can help to abstract away some of the complexity of building and deploying services.

2. Performance Overhead: Composable architectures can introduce performance overhead due to the need to communicate between services and coordinate their interactions.

To mitigate performance overhead, one can use techniques like caching and asynchronous communication between services to reduce the number of synchronous calls. Additionally, one can use load balancing and horizontal scaling to distribute traffic across multiple instances of a service.

3. Integration challenges: As services in a composable architecture are developed and maintained by different teams, integrating them can be challenging, especially if there are changes to the APIs or the way services interact with each other.

To manage integration challenges, it’s important to have clear API contracts and versioning policies, as well as good communication between teams. Additionally, using automated testing and deployment pipelines can help to catch integration issues early in the development process.

4. Operational complexity: Managing and monitoring a composable architecture can be more complex than a monolithic architecture, as there are more components to manage and monitor.

To manage operational complexity, one can use tools like container orchestration and service mesh to manage and monitor services at scale. Additionally, implementing logging and monitoring across all services can help to identify and troubleshoot issues more easily.

5. Debugging challenges: Debugging issues in a composable architecture can be more challenging, as issues may be spread across multiple services, and tracing the root cause can require deep understanding of the entire architecture.

To manage debugging challenges, it’s important to have good observability across all services, including tracing, logging, and monitoring. Additionally, using distributed tracing tools can help to identify the root cause of issues across multiple services

6. Security risks: Composable architectures can introduce security risks if services are not properly secured and authenticated, or if vulnerabilities in one service can affect the security of other services.

To mitigate security risks, it’s important to implement proper authentication and authorization mechanisms across all services. Additionally, implementing security testing and monitoring across all services can help to identify and address security vulnerabilities early on in the development process

Conclusion

Composable architecture is a powerful design pattern that makes it easier to build and maintain complex systems. By breaking down a system into smaller, self-contained components, you can create flexible and scalable systems that are easy to understand, test, and maintain. Whether you’re building a new app or working on an existing codebase, composable architecture can help you create better, more maintainable software.

Disclosure: This was an AI assisted article generated using Open.ai’s ChatGPT.

References & Further Reading/Viewing

1. Sitecore — Composable Ecommerce
2. Origins of Modern Commerce
3. Collections from Point Free
4. Pointfree TCA example (Source Code Examples)
5. Kotlin example of TCA
6. Jetpack Compose
7. Jetpack Compose (Source Code Examples)
8. Create a Component Based Architecture in Android
9. Botomation — Declaratively composing bots
10. Monolith to composable architecture — Magento

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit is interested in early-stage investments in startups in sustainable fashion, clean energy, people tech(health & wellness, leisure, jobs & personal growth), web3, media tech/content creation, legal tech and ed-tech.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter. If you are looking for any help or guidance or investment, please drop a mail to arpitg.smp18 [at] mdpalumni.iimcal.ac.in

What is the Composable Architecture (TCA)? was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

If you landed up here because someone told you that you should implement DORA metrics in your team, you have landed on the right page :) First a bit of history, DORA (DevOps Research and Assessment)was acquired by Google in 2018.

There is a famous saying — “You can’t improve what you don’t measure”, often attributed to management thinker Peter Drucker. This is why we have all the talks around KPIs.

So, how can you decide how effective is your software development process? Or how do your measure your overall software delivery performance? The answer is DORA metrics. If you take up the survey created by DORA team based on their research, you will end up in one of the categories — low, medium, high, elite. Here’s a sample for medium performer

The tool will also show your performance overall and in your industry

Based on the performance rank, the tool also suggests some immediate next steps that might be relevant.

Though, DORA was acquired by Google, the research is publicly available and the KPIs can be added on any cloud — Azure, AWS or custom. DORA metrics, essentially measures software delivery performance of teams using following

1. Deployment Frequency — How often is the dev team deploying to production successfully?
2. Lead Time for Changes — This is average time it takes for a commit to move to production.
3. Change Failure Rate — What percentage of deployments failed?
4. Time to Restore Services — How long does it take to recover from failure in production?

Four keys project provides a vanilla set up to measure these parameters in Google cloud. Here’s how it works

For any cloud platform, similar architecture can be used for creating a platform to measure the above four metrics. There are many paid tools (https://oobeya.io/, https://www.propelo.ai/, https://www.klera.io/ etc) which can be integrated with existing cloud/code repositories to generate all data needed to measure DORA metrics.

DORA metrics should be seen in conjunction with operational performance of the application and not as a standalone metrics. State of Dev Ops report 2022 mentions some important contextual elements that should be considered. It says — “High software delivery performance is only beneficial to organizational performance when operational performance is also high”. So an additional metric to measure operational performance based on reliability (availability, latency, performance, scalability, SLAs) is also important.

DORA Research also suggests that capabilities across different key areas also impact the performance of software delivery teams. Two key areas common to all organizations:

1. Technical: This covers decisions around version control, CI/CD, continuous testing, architecture, deployment automation, test data management, cloud infrastructure, trunk based development, security, database change management etc
2. Measurement: This covers Monitoring & Observability, proactive failure notification, visual management capabilities etc

References & Further Reading

1. DORA Research
2. SRE Books
3. Continuous Data

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit is interested in early-stage investments in startups in sustainable fashion, clean energy, people tech(health & wellness, leisure, jobs & personal growth), web3, media tech/content creation, legal tech and ed-tech.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter

What is DORA metrics and How can it be implemented? was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

While many people argue that VCs and their LPs control Web3 and not the users, many others believe that the blockchain platform will ultimately lead to decentralized Web3.

Decentralized apps or DApps are the apps built-in Web3. They either run on blockchains or decentralized networks or a hybrid of the two. In the Web3 ecosystem, for a DApp if the control over the network, protocol and data is with the users or governed by a set of rules driven by a governance body, we can say that its more decentralized as compared to the Dapps where these controls are with limited people or entities.

In the Web2 world, most companies exploit user data for ad revenue. In most cases, users don’t have control over their data and how it’s stored. We use a product for free and hence become a product. Web2 apps generally have hooks to collect our data including personal choices and behaviours. This data is then monetized in several ways. There is no free lunch in a capitalist economy. GDPR and other similar regulations have brought some control to users over their data. However, users don’t have 100% control yet. More recently, Google announce that they will do away with third-party cookies in Chrome which is a step in the right direction.

Let us now explore some of the key aspects of Web3 with some relative comparison to the world of Web2.

1. User Identity — Web2 apps generally use OAuth or similar tools or password/OTP based authentication. In the majority of cases, users are required to share some personal information. So you cannot use Web2 websites if you are not sharing personal information in some form. Some websites allow anonymous login for partial features. There is only a handful websites that allows full anonymous access. In the world of Web3, authentication is generally tied to a wallet address. So unless the users want to share publicly the wallets they hold, they can use utopian Web3 apps without sharing personal information. In cases where users plan to use the same wallet across different apps, they can transfer their identity. Ceramic and IDX are two tools that allow implementing user-centric authentication. “Secret Network is the first blockchain with customizable privacy. You get to choose what you share, with whom, and how. This protects users, and empowers developers to build a better Web3.”

2. Token Economy — As of now companies raise funds using private equity or debt or public sale of equity. VCs or the investors who own majority stake controls. the board and hence all major decisions of the company. Web3 provides another way of building the company. Entrepreneurs can give tokens to all stakeholders, tokens with voting rights. So, in a utopian Web3 world it’s possible that all stakeholders take key decisions and contribute to long term strategies instead of a few people (more decentralization). Also in this case the wealth created from the success of the company will be distributed to all stakeholders. In reality, VCs dominate this new way of building the company as the percentage of public participation is very less and the risk of companies failing is very high. Radicle has 50% of the token supply allocated for its community members. Gitcoin allows funding of Web3 based digital public goods projects that serve everyone, and solve our most immediate problems. Uniswap community members will get 60% of UNI (Uniswap protocol’s token). The Web3 goal of token economics is that all stakeholders enjoy benefits instead of a few centralized entities.

3. Application/Content Discovery — In the world of Web2, search engines are the primary source of application discovery. Indexing and curation are very important in the world of Web3 also. The Graph is a protocol for organizing blockchain data and making it easily accessible. What Google does for search, The Graph does for blockchains. It is also based on token economics. The graph protocol will challenge the monopoly of cloud providers and search engines if more and more small contributors implement it. Note that almost everything in space is still very much early stage and it’s expected to evolve in the next few years. There will be many providers and just like good SEO helps you get on top of search results, we will have something similar for Web3 discovery.

4. DAOs — DAOs stands for Decentralized Autonomous Organizations. As defined in Wikipedia — “A decentralized autonomous organization (DAO), sometimes called a decentralized autonomous corporation (DAC),[a] is an organization represented by rules encoded as a computer program that is transparent, controlled by the organization members”.

“DAO governance is coordinated using tokens or NFTs that grant voting powers. Admission to a DAO is limited to people who have confirmed ownership of these governance tokens in a cryptocurrency wallet, and membership may be exchanged. Governance is conducted through a series of proposals that members vote on through the blockchain, and the possession of more governance tokens often translates to greater voting power.”

DAOs are not legal yet in many countries and this space is also evolving. In the coming years, we will most likely have hybrid DAOs that adhere to governance rules of the country/state in which operate but also use DAO based rules.

5. Interoperability — This is a nightmare in the world of Web2. If you want to move from one platform to the other, your data will not go with you. Some of the recent policies and standards have enforced that the users can at least export their data. Blockchain interoperability has become an important feature of Web3 as a decentralized web cannot happen without this. Interoperability will allow the transfer of data and value across different networks. Web3 suffers from similar issues as Web2 as the majority of the projects are siloed as of now.

More and more Web3 enthusiasts are realising the importance of interoperability and as the space evolves, we can expect that some good solutions will come up. And we will see the adoption of protocols that enable interoperability at large in the Web3 ecosystem.

6. Payments — As reported in ET, “DeFi is a bottom-up innovation that takes the component of centralised finance and replaces human trust with math-based trust, paperwork with smart contracts, legal enforcement with cryptographic enforcement, and third party audit with open source code and public ledger.”

DeFi eliminates the fees that banks and other financial companies charge for using their services, saves money in a secured digital wallet, and allows anyone with an internet connection can transfer the funds in minutes/seconds.

Cryptocurrencies are the equivalent of fiat money in the Web3 payment world. Since many cryptocurrencies are highly volatile, stablecoins are many times used for transactions. Fast processing and low transaction fees make stablecoins a good choice for peer to peer or cross border transactions.

7. File/Media storage— In the world of Web2, HTTP protocol is used for the transfer of files between client and server. In the world of Web3, IPFS is a protocol for peer to peer transfer. As stated on ipfs.io, “IPFS is a peer-to-peer hypermedia protocol designed to preserve and grow humanity’s knowledge by making the web upgradeable, resilient, and more open”. When a file is added to IPFS, it is split into smaller chunks, cryptographically hashed, and given a unique fingerprint called a content identifier (CID).

Files stored on IPFS are resistant to tampering and censorship — any changes to a file don’t overwrite the original.

Web3 represents the next phase of the internet and is evolving everyday. It is built on open, trustless and permissionless networks. Web3 enthusiasts envision that the Web3 technology stack will be open-source where anyone can collaborate. Open-source protocols and decentralized blockchains in Web3 have already taken co-creation to a new scale. They have opened up ways of creating truly decentralized apps where the users own their data, the creators can leverage public funding to build what they actually want and the community/stakeholders can govern and can also get paid for their contributions.

References & Further Reading

1. Why is interoperability important for blockchain — Gemini Blog
2. Web3 Interoperability — The Hard Parts
3. Oracles, Web3 Interoperability, & the Emergence of IoT
4. Where will interoperability take Web3 — Axelar blog
5. DAO — Investopedia
6. DAO — Ethereum
7. How ‘Web3’ could evolve from a trendy buzzword to a better internet
8. Web2 vs Web3 — Ethereum
9. Why DeFi is the biggest thing in the history of finance — Economic Times
10. Stablecoins — Investopedia
11. Programmable Money for the internet.
12. The Architecture of Web 3 application.
13. Decentralized alternatives — Peertube, Activity Pub
14. The meaning of decentralization
15. Why Decentralization Matters
16. What is Web3 and Why It Matters?
17. What is the Decentralized Web?
18. Modeling Cryptoeconomic Protocols as Complex Systems

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit also consults clients on competitive market analysis, defining MVPs, product ideation, product monetisation and go live strategies.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter

ABC. Always be clappin’.

What are some of the key aspects of Web3? How different are those from Web2? was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

Blockchain and Web3 are the most popular technology trends. If your are a software developer or just a technology enthusiast looking at how to get started with blockchain/web 3, here are some great free learning resources(in no particular order)

1. Nader Dabit Youtube Channel: Has The Complete Guide to Full Stack Ethereum Development, a playlist for web3 that includes How to Build a Full Stack NFT Marketplace on Ethereum with Polygon and Next.js, The Complete Guide to Full Stack Web3 Development and many more.

2. Patrick Collins Youtube Channel: Has The Undeniable Value of Blockchain and Smart Contracts, Intro to Foundry | The FASTEST Smart Contract Framework, How to build an on-chain DAO | Typescript & Solidity and many more.

3. Smart Contract Programmer Youtube Channel: Has Playlists for Tools, Solidity and many more topics.

4. Moralis Web3 Platform: Youtube channel and website for Moralis Web3 cover how to use Moralis Web3 Products to quickly develop dapps.

5. EatTheBlocks Youtube Channel: Covers Ethereum, Solidity, Smart Contracts, Defi and Dapps.

6. Dapp University: Has tutorials/videos on Blockchain, Dapps, ICO and many more topics.

7. Ethereum Development Course: Blockchain at Berkeley — This course on Udemy is free for those who are not interested in a certificate of completion or in interacting with the instructor but just want to learn.

8. Free Basics Course: Enterprise Blockchain Fundamentals is very basic 5 days crash course on 101 Blockchains for anyone who wants to get started. Coursera also has a course that covers the basics. Edx also has a free version that covers basics.

9. Cryptocurrent & Blockchain: DCXLearn website has courses on blockchain and cryptocurrency. If you are interested in trading in crypto, this courses are a good start. CoinDCX is a platform for trading cryptocurrencies.

Udemy also has a free course on this topic.

10. CoinMarketCap: Coin Market Cap has a Learn Crypto, Earn Crypto program. It has partnered with trusted & emerging projects to offer an easy way to learn about cryptocurrency, earning cryptoassets as a reward.

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit also consults clients on competitive market analysis, defining MVPs, product ideation, product monetisation and go live strategies.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter

ABC. Always be clappin’.

Top 10 Learning sources(Free) for Blockchain and Web 3 was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

Social creatures working as a unified entity can perform much better in terms of solving a problem or making a decision as compared to a vast majority of the entity working individually. Many minds are better than one.

The above image shows a flock of birds moving together. You might have seen bees and fishes doing something similar. In general, these groups are collectively called swarm.

As defined on Wikipedia — “Swarm intelligence (SI) is the collective behaviour of decentralized, self-organized systems, natural or artificial.” Examples of swarm intelligence in natural systems include ant colonies, bee colonies, bird flocking, hawks hunting, animal herding, bacterial growth, fish schooling and microbial intelligence. The below TEDx talk video describes a swarm as brain of brains.

Swarm Intelligence has many use cases.

Some of them are listed below:

1. Internet of Things: Powerful decentralized algorithms supported by SI are used to logically control operations of complex IoT systems. By applying swarm intelligence algorithms for IoT devices, we can provide major advantages for energy saving in IoT devices.
2. Human Swarms: Humans don’t have the natural ability to form a Swarm Intelligence. Forming such intelligence requires tight feedback loops among members. Some companies like unanimous.ai are leveraging high-speed networking to form real-time systems that enable “human swarms” to converge online, combining the knowledge, wisdom, insights, and intuitions of diverse groups into a single emergent intelligence.
3. Healthcare: Researchers have used ‘swarm learning’ to detect blood cancer, lung diseases and COVID-19 in data stored in a decentralized fashion. ASCA (Ant System-based Clustering Algorithm) has been used in image segmentation producing better results than 1D-SOM, k-Means, FCM and PFCM algorithms in the detection of small, atypical regions of the image.
4. Pervasive Computing: When the computational capability is embedded into everyday objects such that they can effectively communicate and perform useful tasks without end-user interaction, it’s a form of pervasive computing. Distributed perceptive networks which are used to implement pervasive computing utilize Swarm Intelligence Algorithms. Multi-hop networking, optical network optimizations, and insect drones are some examples.
5. Mobile Ad hoc NETwork (MANET): MANET is a decentralized network with the ability to scale to thousands of connections. They self-learn and organize as they operate and adapt as new and old nodes enter and exit the network under dynamic conditions. Swarm Autonomous Routing Algorithm (SARA) protocol can manage node to node communications in a MANET. SI can also be used to implement a distributed location service in a MANET
6. Data Science & ML: By applying swarm intelligence in parallel we can achieve better results in Hydrological Forecasting. SI can optimize the parameters of an artificial neural network. SI has many other applications in data science and machine learning.
7. Swarm Robotics: Swarm robotics is a field of multi-robotics in which many robots can coordinate in a distributed and decentralized way.

How Swarm Intelligence Works?

In general, swarm systems work in the following way :

1. There are be a large number of distributed agents who work/act in-parallel
2. The agents transmit/ receive signals from the other agents. It is through these signals that all interactions happen.
3. Each agent individually processes information. There is no central command. In that sense, the agents are autonomous. But the agents are strongly influenced by what others in the system do.
4. As the interaction evolves, a swarm intelligence emerges, and the group’s behaviour self organizes. The entire group decides on something that each agent might not have been able to do.
5. If some of the agents are destroyed, others will quickly adapt and the system will continue to function as earlier.

In general, the following principles can be used to describe behaviour that leads to swarm intelligence

1. Proximity Principle: The basic units of a swarm should be capable of simple computation related to its surrounding environment.
2. Quality Principle: A swarm should be able to respond to quality factors such as determining the safety of a location.
3. Principle of diverse response: The distribution is designed so that each agent will be maximally protected facing environmental fluctuations.
4. Principle of stability: The group should not change their mode of behaviour every time the environment changes
5. Principle of adaptability: The swarm is sensitive to the changes in the environment that result in different swarm behaviour

References & Further Reading

1. Swarm Intelligence — Wikipedia
2. What is Swarm AI Technology — Unanimous.ai
3. Swarm Intelligence — Science Direct
4. Swarm Intelligence — Research Gate, Queen’s University, School of Computing Technical Reports
5. Swarm Intelligence — Tech Ferry
6. Building an Identification Model Using Swarm Intelligence and Its Applications
7. Swarm Intelligence — Singularity Group Blog
8. The biological principles of swarm intelligence
9. Swarm Intelligence — Yichen Hu

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit also consults clients on competitive market analysis, defining MVPs, product ideation, product monetisation and go live strategies.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute toward the same goals.

You can follow Arpit on Linkedin and Twitter

ABC. Always be clappin’.

What is Swarm Intelligence and How it works? was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

Ethereum is by far the most popular blockchain for developing DApps. However, building directly on top of Ethereum can result in huge gas fees.

Here’s a note on the gas fee from the Ethereum website.

“Gas refers to the unit that measures the amount of computational effort required to execute specific operations on the Ethereum network. Since each Ethereum transaction requires computational resources to execute, each transaction requires a fee. Gas refers to the fee required to conduct a transaction on Ethereum successfully. Gas fees are paid in Ethereum’s native currency, ether (ETH). Gas prices are denoted in gwei, which itself is a denomination of ETH — each gwei is equal to 0.000000001 ETH (10–9 ETH).”

Gas prices fluctuate and are higher during high activity periods. This is true for all blockchains. If you are trying to mint something on any blockchain, the impact of the gas fee should be considered. It could be one of the primary attributes for selecting a particular blockchain for any project.

In general, building on any blockchain(layer 1) directly will result in a higher fee as compared to doing something off-chain(layer 2). Though most blockchains are trying to introduce mechanisms that can result in lower fees, working on a layer on top of the blockchain can save substantial fees. How? We will review that soon.

The main blockchains are Layer 1 like Ethereum, Tezos, Solana etc. Layer 2 is a scaling mechanism for the main blockchain. There could be multiple implementations for scaling the main blockchain such as rollups (that performs transactions off-chain and rollup those onto the main blockchain), sidechains (which links to the main chain via bridge), plasma chains, multichains, state channels (where a channel is opened among participants).

Layer 2 generally offers better speed for transactions and lower gas fees. We can consider it as a secondary framework or protocol that is built on top of the main blockchain. Generally speaking, we can move a lot of computation heavy work/complex use cases to Layer 2 and leverage Layer 1 for its security and other core blockchain-based features. Let us consider how some of the Layer 2 solutions work:

1. Lightning Network: The Lightning Network can facilitate fast peer to peer transactions on top of a blockchain. It has its nodes and software which allows people to transact on the network without the need of pushing everything to the main blockchain. This allows for quick processing as there is no block wait time. Blocks on a blockchain are a scarce resource. So if we want our transaction to be written to the block quickly, we might have to pay a higher fee for the same. At peak load, this could mean a transaction of USD 50 also. So for small peer to peer transfers, Layer 1 blockchains can be very expensive. Why would anyone pay a fee of USD 50 or even USD 5 for that matter, to pay someone USD 5? When doing volume transactions between 2 peers, this fee is reduced drastically as the user will have to pay one for 2 transactions on the blockchain — open and close of the channel. When you start a peer to peer transaction, record it in the blockchain. Do any number of transactions and then record the final state. Intermediate states can also be recorded if needed.
2. Ethereum Plasma: This allows the creation of a framework of secondary chains to reduce interaction with the main chain. Plasma chains allow the creation of hierarchically arranged numerous small chains on top of the main chain, operated as a blockchain tree. Plasma structure is built using smart contracts and Merkle trees.
3. Roll-Ups: This mechanism enables transactions at layer 2 (reduced gas feed). Data and proof of transaction reside on layer 1. A smart contract at layer 1 is used to enforce proper transaction execution at layer 2 using the data stored on the main chain(layer 1). Roll-ups can be optimistic such as Arbitrum or zero-knowledge such as Polygon Hermez
4. Sidechains: Sidechains are separate blockchains that run in parallel to the main blockchain and can have their operating mechanism (own consensus algorithm ) and can also use the main blockchain for validation. Liquid, RSK are some of the bitcoin pegged sidechains. Polygon is Ethereum based sidechain.
5. Multichain: Multichain networks have their own consensus algorithm and network architecture. We can consider these to be made up of more than one blockchain. For instance, SKALE has some machines in Ethereum main chain and some on SKALE network.

In conclusion, Layer 2 chains allow us to solve several complex use cases and also help us scale Layer 1 chains. As NFTs have become popular, we now have ImmutableX which allows high-performance minting of NFTs. More and more Layer 2 chains will evolve in the next few years and Dapps development and execution is only set to become more robust and economically viable (near-zero transaction fees).

References & Further Reading

1. Layer 2 Blockchain — PC Mag
2. Layer-1 and Layer-2 Blockchain Scaling Solutions
3. A Beginner’s Guide to Bitcoin’s Lightning Network — Binance Academy
4. What Is Ethereum Plasma? — Binance Academy
5. Layer 2 solutions — One37PM
6. Sidechains — Ethereum Website
7. Blockchain Sidechain — Komodo Platform
8. Sidechains and Plasma — Polygon

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit also consults clients on competitive market analysis, defining MVPs, product ideation, product monetisation and go live strategies.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter

ABC. Always be clappin’.

L1 and L2 chains in Blockchain Ecosystem was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

NFT = Non Fungible Token. What does that mean? In order to understand, let's try to learn what fungible is first. A fungible item is one that can be interchanged with another such item for all practical purposes. For example the dollar bill or any other currency. One twenty dollar bill can be exchanged with two 10 dollar bills or other twenty-dollar bills. So the dollar is fungible. Commodities(same grade), common shares(same company) etc are all fungible items.

Let's try to understand non-fungile now with the example of a piece of land. One piece of land cannot be replaced with another. Two owners can swap land of the same value but that will be more of barter of two different assets. Two land plots will differ in one or more attributes. For example, a corner plot of land of the same size is generally of a higher value than a regular plot. This is not true for fungible items, it does not matter if the dollar bill is old or new — both will have the same value. So in the case of non-fungible assets, each unit(land) has unique characteristics(for land consider — location, fertility, size and so on) that add or subtract value.

In summary, fungible items are defined by their value and non-fungible items by their unique properties (and hence cannot be exchanged).

Technically, NFT is a non-interchangeable unit of data stored on a blockchain. It can be sold and traded but it cannot be exchanged or replaced because of its unique cryptographic signature. Here’s a video from Finematics that explains some basics about NFTs and DeFi

DeFi is Decentralized Finance. In the world of web3, everything is decentralized so any finance based application in the world of web3 is a DeFi app. If you are wondering what is web3 — consider it to be a blockchain-based version of the internet for simplicity. An internet where everything is decentralized and built on top of a blockchain! One can say that web3 is the base of the decentralized future — DAOs(Decentralized Autonomous Organizations), DeFis, GameFi, DApps(Decentralized Applications), Metaverse, creator economy and so on.

Let's go back to NFTs as web3 is out of scope for this story. NFTs can be implemented on any blockchain that supports smart contracts. The most common ones are Ethereum based — ERC 721 and ERC 1155.

Tezos, Polkadot, Cosmos, Binance Smart Chain, Solana, Matic, Cardano and so on also support the creation of NFTs.

In order to understand what use cases NFTs can solve, we must first understand some of the unique properties/mechanisms of the physical world that can be replicated in the digital world using NFTs.

1. Scarcity- “Scarcity refers to a basic economics problem — the gap between limited resources and theoretically limitless wants.” The primary causes of economic scarcity are demand-induced, supply-induced, and structural. Any NFT that has this unique property is likely to do well in future. So limited editions of anything, popular art and creator tokens for fans can all do well theoretically.
2. Uniqueness/Originality — Famous paintings, custom-built homes/cars etc all have this attribute.
3. Proof of Ownership — How does one prove that an asset belongs to them? Physical control does not prove ownership. One can have physical control of a piece of land or art even though they do not have the ownership. Depending on what the asset is there are different ways to prove ownership. NFTs provide a way for the same in the digital world.
4. Exchange — A movie ticket bought can be transferred to any other person. The same is true for passes for some events. The benefits that come with the ticket/pass are transferred to another person with the exchange. This can be replicated in the digital world too.
5. Royalties — In the most simple terms, royalties are payments to the owner of the asset for using that asset. An author will get royalties on the books sold, a singer on the copies sold and so on. In most cases, there is a mediator between the creator and the person who uses the asset.

Over the last few years, NFTs have been used in many areas, some of which are mentioned below. There are many more use cases that can be implemented using NFTs.

1. Digital Art /Collectibles — Any digital art can be minted as an NFT and it then becomes a unique digital art as the hash associated with each NFT is unique. Many traditional art auction houses like Christie’s have entered the NFT world in the last few years. NFTs are designed to give ownership of work only and do not restrict making any copies. Just like we can find print copies of The Last Supper but there is only one original, there can be only one original NFT based digital art. Many individual artists have also sold their digital art as NFTs. NFTs have been used to create Digital Collectibles by many well-known brands and franchises. Many digital collectibles based new communities have also started in the last 2 to 3 years. Digital collectibles have also been used in games.
2. Digital version of real-world items — Many real-world items can be replicated in the digital world using NFTs. For instance, tickets to events, legal documents/deeds and so on
3. Gaming — In addition to providing in-game items /collectibles, NFTs can be used to create new revenue models in-game economy. Imagine trading a farm created in-game like Farmville or selling a particular level that you have reached in a game to someone else who wants to play that. New avenues of generating money by doing something you already like. Collectibles can be resold to new players generating revenue for both gamers and game creators.
4. NFT Domains— As of now, IP addresses are mapped with domain names to find any website. In the web3 world, we have equivalent services like Ethereum name service which provide alternate website names powered by NFTs. NFT domains are new web extensions that are linked to the blockchain via smart contracts. NFT domains map hexadecimal wallet addresses into human-readable form and enables censorship-resistant websites.
5. Authentication — Since NFTs are unique and cannot be reproduced, they can be used for verification and authentication also. Everledger uses NFTs to solve this use case in the supply chain. Identity credentials like driver licenses, course certifications etc can also be issued as NFT in future.
6. Proof of ownership — Cryptographic proof of ownership can be used as a key in future to unlock cars, homes or any other physical assets. NFTs can be used to tokenise ownership of physical assets giving way to fractional ownership of real estate etc. As the ownership is settled by NFTs, they can also be used in decentralized loan applications also.
7. Royalties — Creators can earn some commission on each future sale by using NFTs. Creators can have a smart contract that triggers an auto payment to the original creator whenever the NFT change hand. So a music video or digital audio will earn recurring revenue if it’s transferred. NFTs will also allow creators to directly sell their content to a mass audience without any intermediatory.

There are no limits to what the creative human mind can up with! The use of NFT is not limited to the above uses cases only but these are the common ones.

References & Further Reading

1. Non Fungible Token — Wikipedia
2. Fungible Goods — Investopedia
3. Non Fungible Token — Ethereum Site
4. Scarcity — Investopedia
5. NFTs a question of ownership.
6. Crypto Kitties sold for USD 172,000
7. Twitter Thread on Ether Rock

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit also consults clients on competitive market analysis, defining MVPs, product ideation, product monetisation and go live strategies.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter

ABC. Always be clappin’.

The Many Use Cases of NFTs was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.

Most of us understand the word debt — essentially an amount of money that you owe somebody. For a debt transaction, we need a borrower and a lender. The lender will lend money to the borrower and the borrower will pay some interest and return back the principal on a mutually agreed timeline.

Fixed Deposits are the most common form of debt where we are giving a loan to the bank and the bank pays us the interest. This is where many people invest their money as it's considered a safe investment. If the interest rate of FD is less than inflation, you are actually taking a hit at your buying power for the money deposited in the bank. With equity and other investments, you might lose the money so FD is what most people in India prefer even if the interest rates are low.

Just like banks, there are many other borrowers in the market — governments, companies, NBFCs and so on. All of them raise money by floating debt instruments — bonds/mortgages etc. Debt instruments do not allow for pre natural withdrawals generally, unlike FDs. But they can be traded in the secondary markets just like equities.

Debt instruments can be short term( maturity < 1 yr) or long term(maturity > 1 yr). Treasury Bills, Certificate of Deposits, Commercial Papers etc are short term debt instruments. Government Bonds/G-Secs, Corporate Bonds etc are long term debt instruments.

Debt mutual funds will invest in short term/ long term debt instruments. Just like for equities we have NSE/BSE, debt funds are traded on the debt market. Debt funds are ideal for investors who aim for regular income but are risk-averse as compared to equities. If you have some amount that you want to invest for the short term, you can go for short term debt funds. If you are ok with locking in the amount for a longer duration, you can go with longe term debt funds.

Depending on how the funds are structured, debt funds do carry interest rate risk(that the interest rate may go down) and credit risk (delay in payment/defaults). This is why bond ratings/ company ratings are important. All government bonds are safest as governments will not default. Companies with higher ratings will have lower yield rates on the floated debt instruments and vice versa.

Many debt mutual funds will take debt instruments from different rating buckets and the investors can invest in those instead of investing in individual debt instruments. This is similar to equity-based mutual funds where there will be securities with different market capitalization.

As the interest rate goes up in markets, the existing price(face value) of the bonds will decrease. So long term bonds have a higher interest rate risk as compared to short terms bonds as the chances of the interest rate going up is higher for a longer duration. When interest rates are low then long term bonds can give double-digit returns also.

Debt funds cannot beat returns of equity but have a lower risk but give better returns than the fixed deposits for the same duration of the investment, especially if you are investing for 2–3 years.

In general, for diversification, it is recommended to distribute the portfolio across equity, debt, gold and other assets. Depending on your age and risk profile, the % allocation can be decided.

In terms of taxation, there is some difference between equity and debt. For long term bonds, the tax rate is 20% after adjustment for indexation(as compared to 10% in equity) and for the short term, it will depend on the income tax slab you are in.

Further Reading

1. Different between Debt & Equity Markets
2. Debt Instrument
3. Money Market Instruments

What Are Debt Funds? was originally published in Notes From Arpit on Medium, where people are continuing the conversation by highlighting and responding to this story.

Modern-day computing tasks require us to design systems that are spread across multiple computers/servers. Distributed systems are built around the following key features

Scalability — the capability to add additional processing units as the workload increase

Availability — reduce chances of failure by having multiple nodes

Concurrency — the ability of the sub-components to run asynchronously.

Replication — the ability to share information across different nodes for consistency

The CAP theorem (Iron Triangle of Data) states that any distributed data store can provide only two of the following three guarantees

C:Consistency — every read will receive the most recent write

A:Availability — all requests will receive a response which may or may not be the most recent write

P: Fault tolerance of next work partitions — system will work even if one or more of its components will fail.

CAP theorem has been used to decide trade-offs in designing distributed systems. However, in most systems, strong consistency is preferred over availability. And hence the CAP theorem might not be the right tool for assessing trade-offs while designing a distributed system.

So, what other tools are available?

1. Delay Sensitivity — provides tools for reasoning about trade-offs between consistency and robustness to network faults.
2. PIE Theorem — The pie theorem(Iron Triangle of Purpose) states that for a data store we can choose two out of the following three purpose-based features

P: Pattern Flexibility — the database allows random access and ad hoc queries — can we ask new, unanticipated questions?

I: Infinite Scale — the database can gracefully increase size and throughput without any practical limits — data can grow as big as we want.

E: Efficiency — the database will deliver the required query latency for the workload all the time.

Instead of thinking of trade-offs based on the CAP theorem, we should ask what is the purpose? And this is where the PIE theorem is relevant — in the context of purpose-built systems.

PE System: When we need pattern flexibility and efficiency. Example — relational databases.

IE System: When we need infinite scale and efficiency. Example — NoSQL databases like Mongo, Cassandra, HBase, DynamoDB

PI System: When we need pattern flexibility and infinite scale. Example — Amazon Redshift and Snowflake.

References & Further Reading

1. Splunk Blog — What are distributed systems.
2. CAP Theorem — Wikipedia
3. Don’t settle for eventual consistency.
4. Stop Calling Databases CP or AP
5. Video — Matching the Database to the Workload
6. Applying AWS Purpose-Built Database Strategy
7. A one size fits all database doesn’t fit anymore

About the Author

Arpit is a seasoned technologist with vast experience in leading large cross-functional and cross-geography teams. Arpit also consults clients on competitive market analysis, defining MVPs, product ideation, product monetisation and go live strategies.

Arpit is also interested in early-stage investments in startups in sustainable fashion, clean energy, people tech(health & wellness, leisure, jobs & personal growth), web3, media tech/content creation, legal tech and ed-tech.

Arpit believes we should all contribute back to society. He has set his goals for social work in five broad areas. You can read more about the same in his blog post “Do Good, Together” on Tumblr. Arpit is interested in working with people who want to contribute towards the same goals.

You can follow Arpit on Linkedin and Twitter

ABC. Always be clappin’.

What is PIE Theorem and Where it’s Relevant was originally published in Tecnología on Medium, where people are continuing the conversation by highlighting and responding to this story.