UPI is touted as a public good, and open infrastructure, but the reality is that UPI is open to certain select corporations that can partner with banks. The UPI network is not directly exposed to the customer, but access is proxied via third-party proprietary applications which come with their own restrictions, limitations, and usage tracking.

As a FOSS enthusiast, what would it take to build a FOSS UPI app from the ground up? We set our goals (legal, functional, secure, cross-platforms), and look at the various approaches that can get us there.

The rest of the talk will cover a multi-step plan for what can get us to hosting a FOSS UPI Application on F-Droid and other platforms.

This talk is based on the original discussion hosted at FSCI Online Forum and private discussions over email with friends.

Kovid Goyal shares his experience with FOSS.

Personal website: kovidgoyal.net

Excalidraw (github.com/excalidraw/excalidraw) is a free, virtual, open-source editor for sketching hand-drawn like diagrams. Its built in React + typescript.

This talk covers:

Opening Note: IndiaFOSS 2.0 | Rushabh Mehta | Krutika Thakkannavar | Vrinda Menon | FOSS United

In this talk we look at that collaboration that takes places between storage-devices and open-source system-software.

The talk covers few important storage/software advancements that has taken place in recent times:

The short talk covers the followings:

https://github.com/reactplay

Application testing is one of the biggest barriers to achieving truly Continuous Deployments because it's use-case specific. Developers often avoid writing test cases because its time consuming, needs to be maintained for every change and metrics like coverage don't necessarily guarantee quality. In this session, we'll talk about how we could capture test cases from traffic data, how all infrastructure can be mocked automatically and how application writes could be safely replayed.

https://github.com/keploy/keploy

Keploy is an open-source no-code API Testing Platform that generates test cases and data mocks from API calls.

We'll walk through examples of how keploy can work alongside existing testing frameworks and capture test cases quickly and mock infrastructure without needing to write Unit API test cases. We'll also cover how these test cases would evolve as the application grows. The core contributors to Keploy will provide an overview of its features and capabilities, and how it is used at scale covering use-cases across microservices across various programming languages.

Outline/Structure of the Demonstration

Learning Outcome

Chaos Genius is an open-source ML-powered analytics engine for outlier detection and root cause analysis. It can be used to monitor and analyze high dimensionality business, data and system metrics at scale. Chaos Genius is used in over 25+ countries in domains ranging from IoT, gaming, fintech & retail.

Advanced augmented analytics that Chaos Genius offers has been the domain of select heavily funded & closed source companies like Sisu Data, Inference.io & Anodot. Despite real need with ever increasing data scale, this was not accessible to most orgs due to exorbitant prices.

In this talk we would cover the motivation & journey behind building & open sourcing Chaos Genius. We will focus on how open source helped us build something fast & deep along with the technical challenges we encountered.

The talk will focus on:

Product walkthrough available at Data Talks OSS Spotlight: https://www.youtube.com/watch?v=AJQwtNipFaA

https://github.com/chaos-genius/chaos_genius

NeST (Network Stack Tester) is an open-source Python package that simplifies the process of performing networking experiments by using Linux network namespaces. It provides easy-to-use APIs to build a virtual network topology, set up routing, run different types of traffic, capture packets, and collect and visualize statistics in different data/graphical formats. NeST provides a cost-effective, reproducible, and scalable alternative to physical testbeds to perform network experiments. NeST can be used for multiple purposes:

This talk will cover the following topics:

https://gitlab.com/nitk-nest/nest

What is OCaml?

OCaml is an industrial strength functional programming language with an additional bonus of imperative and object-oriented features. OCaml has been around for more than 25 years now, with more and more exciting developments.

Why OCaml?

Amongst other thigs, OCaml offers -

We'll have a quick tour of how they make life as a developer more pleasant.

Wait, do you use it in real life?

Of course! We shall see some examples of applications written in OCaml. To name a few: Coq - theorem prover, Hack - programming language, Irmin - a git like data store and the list goes on.

But but, where is Multicore support?

It's here!! OCaml 5, the first official release of Multicore OCaml that adds native support for parallelism and concurrency. OCaml is the first mainstream programming language to support Algebraic Effects. In this section, I'll summarise the many-year journey of Multicore OCaml and the efforts it took to culminate into OCaml 5.

https://github.com/Sudha247