Keynote: Does Open Data Science Lead to the End of the World?
summary
Argues that open data science drove rapid adoption by lowering barriers, cites partnerships with Snowflake Microsoft Azure and Oracle, and names Pi script (a WebAssembly Python runtime he links at pisgrip.com) as a built demo that runs in-browser on billions of devices. Claims code generation via LLMs "mostly works" and will vastly increase crappy software written by non-developers, flags attribution and licensing as the hard unresolved problem, and asserts AGI is possible and likely unless values and economic attribution change.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:27Keynote introduction
- 3:34Python and open data science growth
- 7:49Meaning of 'open' in data science
- 10:55Software as community
- 13:24Open source as economic oddity
- 17:18Conway’s law and culture
- 19:45Python’s community and design
- 22:54Python’s education focus
- 25:43Coding literacy limits and thoughtware
- 27:27Pi Script introduction
- 30:54ChatGPT and accessibility
- 31:55State of AI and LLMs hot takes
- 36:53Inevitability of AGI and values
- 40:45Who decides tech values
related talks
17mKeynote: From Artisans to AI: The Evolution of the Open Source Ecosystem
12mKeynote: The Community of Intelligence
41mKeynote
16mKeynote: Open Source for AI and AI for Open Source
15mKeynote: Building an Open Agentic Future
13mKeynote: Introductory Remarks
52mKeynote: State of the Union
13mThe Good, The Bad And The Ugly Of Open Source AI