# E189: Why Your Backup Platform Should Be Open Source with Plakar Page: https://stenobird.com/podcast/open-source-startup-podcast/e189-why-your-backup-platform-should-be-open-source-with-plakar Text version: https://stenobird.com/podcast/open-source-startup-podcast/e189-why-your-backup-platform-should-be-open-source-with-plakar.md Podcast: [Open Source Startup Podcast](https://stenobird.com/podcast/open-source-startup-podcast) Published: 2026-01-15T19:16:59+00:00 Episode link: https://podcasters.spotify.com/pod/show/ossstartuppodcast/episodes/E189-Why-Your-Backup-Platform-Should-Be-Open-Source-with-Plakar-e3dn706 Audio file: https://anchor.fm/s/3eab794c/podcast/play/114055622/https%3A%2F%2Fd3ctxlq1ktw2nl.cloudfront.net%2Fstaging%2F2026-0-15%2F34116dca-f424-46e9-2a88-2ab6eee8ca87.mp3 Processing state: processed JSON: https://stenobird.com/v1/public/podcasts/open-source-startup-podcast/episodes/e189-why-your-backup-platform-should-be-open-source-with-plakar Duration seconds: 2394 ## Resource Cloud providers and SaaS platforms do not guarantee data backups, leaving organizations vulnerable to data loss and ransomware. Plakar addresses this by providing an open-source, end-to-end encrypted backup platform built on an immutable storage engine. ## Highlights - Main idea: True data security requires end-to-end encryption that allows for deduplication without compromising privacy - Practical takeaway: Do not assume SaaS providers like Google Workspace or Office 365 handle your backups; the responsibility lies with the user - Failure mode: Relying on cloud-only, proprietary backup solutions can lead to vendor lock-in and a lack of verifiable data integrity - Technical insight: Using an immutable storage engine like Kloset enables secure, versioned snapshots that protect against accidental deletion or ransomware - Strategic advice: Open-source founders must be transparent about their monetization models to maintain community trust and avoid sudden shifts to closed-source ## Topics Open Source Software, Data Backup, End-to-end Encryption, Immutable Storage, Cloud Security, SaaS Resilience, Deduplication, Ransomware Protection ## Chapters - 1:00 — The Origins of Plakar: The technical foundation of Plakar, rooted in advanced chunking and deduplication algorithms developed by experienced open-source contributors. - 3:55 — The Kloset Storage Engine: An exploration of Kloset, the immutable, encrypted, and deduplicated container that serves as the core of the Plakar platform. - 6:45 — Solving Modern Data Scale: Why legacy backup designs fail when faced with the massive, rapidly growing datasets of the modern era. - 9:35 — The Plakar Protocol: How the protocol enables secure data transmission to service providers while maintaining user control over encryption and resilience. - 12:45 — The Challenge of SaaS Integrations: The high cost and technical difficulty of maintaining up-to-date connectors for frequently changing SaaS APIs. - 15:40 — Diverse Use Cases: From personal photo backups to enterprise-grade data protection, examining the variety of users currently adopting Plakar. - 18:40 — The Mission of Open Source: The importance of educating the market on the 'shared responsibility model' and the dangers of assuming cloud-native safety. - 21:35 — Lessons from Data Loss: Personal experiences with catastrophic data loss and how they shaped the necessity for a resilient, transparent backup system. ## Actions - request_transcript: `POST https://stenobird.com/v1/public/podcasts/open-source-startup-podcast/episodes/e189-why-your-backup-platform-should-be-open-source-with-plakar/transcription-requests` — Idempotently request low-priority transcript generation for this episode. - read_markdown: `GET https://stenobird.com/podcast/open-source-startup-podcast/e189-why-your-backup-platform-should-be-open-source-with-plakar.md` — Read the agent-friendly Markdown representation of this episode resource. A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed. ## Transcript Full transcripts are not published on public pages unless there is a clear rights basis.