Software that allows several users to collaboratively edit a document, such as Google Docs, has traditionally been ignored by the databases community. This is surprising, because managing the edits to a text document, spreadsheet, vector graphics file, etc. is very much a data management problem, albeit with a data model that is very different from that supported by most databases. Collaboration software needs replication, concurrency control, and data layouts for efficient storage and retrieval.

In our work on Automerge, a CRDT library for building collaborative software, we have been taking concepts from databases and applying them to the domain of collaborative editing. In this talk I will show how column-oriented data formats, multi-version concurrency control, log-structured storage, and other ideas from databases can be used to make collaboration software better.

Bio:
Martin Kleppmann is a researcher in distributed systems, currently at the Technical University of Munich; in January 2024 he will become an Associate Professor at the University of Cambridge. He works with the Ink & Switch research lab on local-first collaboration software, and advises the decentralised social network Bluesky. In a previous life he was a Silicon Valley software engineer and entrepreneur, cofounding and selling two startups and working on large-scale data infrastructure at LinkedIn. He is the author of the best-selling O’Reilly book Designing Data-Intensive Applications.