Triplox
Open-source database inspired by Datomic, running on S3-backed storage with incremental Datalog queries and clients in multiple languages.
🔗 Visit Triplox
Description
Datomic pioneered a distinctive way of thinking about databases: instead of overwriting data, every change is recorded as a fact over time, and you query that history with Datalog, a more expressive query language than SQL for certain relationship-heavy questions. Datomic itself is a commercial product with a specific storage backend requirement; Triplox is an attempt to bring that same data model to teams that want it open source and built directly on cheap, durable object storage like S3.
Triplox is an open-source database built on SlateDB, designed around Datomic's transaction semantics and data model, but optimized for OLTP (transactional) workloads with Datalog queries that can be incremental — subscribing to a query and getting updates as new facts arrive, rather than re-running the whole query each time. It uses a single-writer, multiple-reader architecture and is built with a client/server design so client libraries can eventually exist across Rust, Clojure, Java, Go, Python, and TypeScript (some planned rather than shipped yet). It's licensed Apache 2.0 and free.
💬 Our review
The short version: Triplox is a serious, technically ambitious attempt to give Datomic's data model (time-aware facts, Datalog queries) an open-source home on object storage — genuinely interesting for developers who've wanted Datomic's approach without its commercial licensing, but it's early-stage infrastructure, not a drop-in production database yet.
Datomic itself remains the mature, battle-tested option if you want this data model today and can accept its commercial terms and storage requirements. DuckDB solves a different problem (fast local analytical SQL, not transactional Datalog) and is mentioned mainly as a storage-engine comparison point rather than a functional competitor. Mentat was Mozilla's attempt at an open Datomic-like store and was eventually discontinued — a useful reminder that this exact idea has been tried before and is hard to sustain long-term; Triplox will need to prove it avoids that fate. The incremental/subscribable Datalog queries are a genuinely useful feature beyond what Datomic itself emphasizes. Worth watching and experimenting with for teams drawn to the Datomic model who want an open, S3-native alternative; not yet a safe bet for production-critical data given its early stage and multi-language clients still partly planned.
💰 Pricing
📊 Global score
🤖 AI-enriched data
Apache License 2.0, gratuit
Pros
Modèle de données Datomic (faits temporels, Datalog) en open source
Construit sur object storage (S3) via SlateDB, coût d'infra réduit
Requêtes Datalog incrémentales/souscriptibles
Architecture client/serveur multi-langage prévue
Cons
Projet précoce, pas encore prêt pour de la production critique
Clients multi-langages encore partiellement à l'état de plan
Précédent historique (Mentat de Mozilla) abandonné sur une idée similaire