Skip to content

Comment on A lost decade chasing distributed architectures for data analytics?

Comments

This is really a question of economics. The biggest organizations with the most ability to hire engineers have need for technologies that can solve their existing problems in incremental ways, and thus we end up with horrible technologies like Hadoop and Iceberg. They end up hiring talented engineers to work on niche problems, and a lot of the technical discourse ends up revolving around technologies that don't apply to the majority of organizations, but still cause FOMO amongst them. I, for one, am extremely happy to see technologies like DuckDB come along to serve the long tail.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.