Source Network Concept Explainer: How Modular Deployment Can Bring DefraDB to (Any) Infra Near You

Our DefraDB-focused Source Network Concept Explainers have journeyed into the core of our database to help you understand how our technology makes data management more flexible and scalable, dynamic and secure, consistent and conflict-free, user-centric and globally available, and seamlessly connected. Together, these components solve countless problems around traditional data management — centralization, fragmentation, a lack of privacy and security — via a database that provides power over where application and user data exists, who can access it, and how it's secured.

Why is our approach so powerful — and essential? Just think about all the data you engage with daily across your devices. Whether you're sending a text on your iPhone, posting a photo on Instagram, collaborating in a Google Document, or streaming a song on Spotify, all of the data you're engaging with is being stored somewhere — most likely in the cloud.

Data in the cloud might mean a rainy day for users

Perhaps you’re thinking: what’s wrong with the cloud, which I pay for each month, and have a password to access and can even download data from to my local device? Sure, you’re not wrong. But if you read between the lines — and code — you’ll begin to understand that while you might be able to access, change, and download this content; this cloud-based data is managed on centralized servers (Google Cloud Provider, Azure, Amazon’s AWS, etc.). While we often sign away rights to our data via the terms and service documents sneakily buried within an applications legal page, our data is still often at risk of unexpected manipulation or being lost via a security breach if things go wrong (which they often do). When this happens, companies are often quick to cite updated changes to terms and services, which can happen unknowingly to the user.

While many put a great deal of faith in the cloud and enjoy these services due to their ability to enable real-time access and collaboration, by centralizing data processing and storage servers, these cloud-based applications remove ownership and control from users (some even refer to the cloud as a prison of data). If a cloud server goes down (which could happen via a technical issue or something as simple as a developer not paying their bills or going against the provider’s terms and conditions), all the data can be lost, which can mean game over for applications, which may stop working completely. For users, this loss of connection can be disastrous.

Luckily, cloud-based servers aren’t the end-all, be-all of data processing and storage; other, more secure, intuitive options exist. Let’s learn what they are.

Change is in the air via local-first software

Enter local-first software, devices, and modular deployment, which can help users truly own their data, even with the cloud hovering above and around us and our devices. Popularized in a paper by Martin Kleppmann, local-first software is a "set of principles for software that enables both collaboration and ownership for users," allowing for the ability to work offline and collaborate across multiple devices while maintaining "the security, privacy, long-term preservation, and user control of data."

This should sound pretty familiar if you've been reading our concept explainers (especially our Merkle CRDTs blog). This concept packs all of the accessible and synchronous punch of cloud-based technology but with software that is "local,"(phone, computer, smartwatch, etc) which means that it uses your personal device hardware to process and store the data of your local application. Local-first software enables you to experience the potential of running web applications across your devices, including with collaborators, while protecting your data against conflicts that might occur.

Case in point: your device goes offline. In this scenario, an offline replica of your data is accessible from the cloud or decentralized infra, giving you the best of both worlds (remote and local-first software). This drastically different data storage and sharing approach enables updating collaborative data in real-time without conflict — even if your device is offline.

DefraDB would keep a replica of your local state of the data while you are offline, and once you come back online, it would talk to the other user's nodes. If they have different states for the same data, it will sync it to arrive at an agreed state between the different local apps.
For developers and applications, local-first software frees up risk around backends, servers, latency (a key cost and sustainability issue among next-generation, data-intensive applications across AI and machine learning), and pricey cloud computing fees, which can be detrimental to scale, speed, and the security of products — and users. This software can be connected to the cloud (or decentralized infrastructure) to sync data whenever and wherever it’s needed. And If you lose connection, everything still works because it's local to your device.

Across the growing landscape of web3-powered applications and other next-generation technologies that reimagine the relationship between data, applications, and users (edge computing, internet of things, AI, etc.), local-first software can work in unison with other forms of innovative modern computing. Our database, DefraDB, is one of the only few.