lectures

Lichess founder Thibault Duplessis talks at Big Techday 10 in Munich about how Lichess works and how he built it. He covers the project's origins and technical challenges including cheating detection.

David MacKay, scientific advisor to the UK energy department, gives a Harvard talk on sustainable energy at macro and micro scale, drawn from his book Sustainable Energy Without the Hot Air. It stresses the land area renewables require.

Jon Purdy's Stanford EE380 seminar giving an overview of concatenative programming, from function composition theory and history down to low-level implementation and correctness benefits.

Tom LaGatta's talk on category theory at a 2014 LispNYC meetup, covering objects, morphisms and the basic ideas in an informal, conversational style. It runs about 97 minutes.

Jon Gjengset's 2020 PhD dissertation presentation at MIT on partial state in dataflow-based materialized views, the idea behind the Noria database. The Q&A is cut out.

Salman Khan explains why he created Khan Academy, a structured library of video lessons with interactive exercises. He calls on teachers to flip the classroom, with lectures watched at home and homework done in class with teacher help.

Bret Victor's DBX 2013 talk, delivered in the guise of a 1973 IBM engineer with an overhead projector. It surveys forgotten ideas from 1960s and early 70s computing and warns against assuming you know what you're doing.

Alan Kay's 2009 UIUC talk arguing that computing is more a pop culture than a science and that ignoring history and fundamentals leads to inferior solutions. He touches on the history of interactive environments and what the web should have been.

James Mickens's humorous USENIX Security keynote argues that improving security and trusting opaque technology like machine learning is far harder than the industry admits.

Andrew Ng presents unsupervised feature learning and deep learning, showing how algorithms can learn representations from unlabeled data instead of hand-designed features. He gives examples in vision and audio, including visualizations of features learned by successive layers.

Michael Steil and Christian Hessmann's 34C3 talk explaining the Apollo Guidance Computer end to end: hardware design, operating system, and mission software.

David Graeber's Talks at Google presentation on his book Debt: The First 5,000 Years, covering the history of debt, the myth of barter and the Jubilee tradition.

Alan Kay's 2015 talk on the power of simplicity, arguing that systems become simpler by building on more sophisticated fundamental building blocks and that big research horizons drove invention at Xerox PARC. It ranges across computing, tools and scaling software.

Andrej Karpathy, then Tesla's head of AI, presents at ScaledML 2020 on how Tesla builds full self-driving using vision-based neural nets and fleet data collection. He covers training custom detectors and neural nets gradually replacing traditional code.

Andrej Karpathy gives a one-hour general-audience introduction to large language models, covering what they are, how they are trained, comparisons to operating systems, and security challenges such as jailbreaks.

Shashi Tharoor's Oxford Union speech arguing that Britain owes reparations to its former colonies, focused on the effects of British rule in India. It lists economic and political damage done by colonialism.

Valve's John McDonald presents at GDC 2018 on how Valve uses deep learning (VACnet) to detect cheaters in Counter-Strike: Global Offensive. He covers training on human-reviewed replays and how model output is confirmed by manual review.

Geoffrey Hinton's Google Tech Talk linking machine learning to biology, explaining dropout as a form of model averaging and why sexual reproduction and stochastic spiking neurons may prevent overfitting.

Steve Blank tells how World War II, Stanford's Frederick Terman and military and intelligence agencies shaped the rise of Silicon Valley. The talk centers on the Cold War electronic warfare and signals intelligence work that seeded the region's tech industry.

Jason Scott's DerbyCon 2012 talk on rescuing the Prince of Persia source code from old floppy disks, working with the game's author Jordan Mechner. It is a story of digital archaeology and recovering data from ancient media.

Dylan Beattie's entertaining NDC talk imagines an alternate history of the web, asking what if different decisions had been made on operating systems, browsers and standards. It blends real history with humor.

Luis von Ahn, creator of CAPTCHA and reCAPTCHA, explains how small contributions from many people online can digitize books. He then introduces Duolingo as a way to translate the web while people learn languages.

A GDC 2018 talk by Valve's John McDonald on VACNet, a deep learning system that analyzes CS:GO gameplay to detect cheaters. It covers how the model is trained on human-reviewed Overwatch cases and how its output feeds human moderation.

Psychologist Barry Schwartz argues in this TED talk that abundant choice makes people more paralyzed and dissatisfied rather than freer and happier.

Keith Packard traces the 35-year history of the X Window System at linux.conf.au 2020. He uses X to examine licensing, corporate interests and community dynamics in free software, including Unix workstation vendors and Stallman's influence.

Conal Elliott presents his paper on automatic differentiation, deriving a simple, generalized AD algorithm from a natural specification. Varying the representation of derivatives yields forward and reverse mode algorithms that avoid the complexity and statefulness of backpropagation.

Erlang co-creator Joe Armstrong's Code BEAM SF 2018 keynote revisits good ideas from the history of computing that were dropped, and asks which should be revived. It also touches on Erlang and Elixir.

David Beazley's PyData Global 2021 keynote, an oral history of early scientific Python. He tells how Python ended up running on a Connection Machine supercomputer at Los Alamos.

Physicist Geoffrey West speaks at an FQXi conference on time about complexity and scaling laws. He covers how metabolism, growth and lifespan scale across organisms and how the same mathematics applies to cities and companies.

Mike Rowe, host of Dirty Jobs, shares stories from messy real-world jobs in this TED talk. He argues that hard work has been unjustly degraded in society.

The 1965 Cambridge Union debate between James Baldwin and William F. Buckley on race, justice and history in America. Baldwin argues that the American dream came at the expense of Black Americans.

Sandy Pentland of MIT presents his book on social physics, which uses big-data studies of how people interact to explain how ideas spread. He describes findings on collective intelligence and social influence.

Ali Rahimi's NIPS 2017 Test-of-Time award talk, in which he argues that modern machine learning has become alchemy, working without being rigorously understood. He calls for more rigor and understanding in the field.

Locksport historian Schuyler Towne's RVASec 2012 talk on why people lock their doors, tracing the history of locks and the assumptions about security behind them.

Todd Becker's Ancestral Health Symposium talk arguing that myopia is driven by near-work and can be reversed, covering the biology and epidemiology and experimental evidence for reversal by using weaker lenses.

Jason Scott of Archive Team describes how a loose band of volunteers rushes to copy sites like Geocities, Friendster and Yahoo Video before their owners delete them. The talk covers the tactics they use and the argument that this data is history worth saving.

Max Tegmark's MIT CBMM talk on why deep learning works so well. He argues that the laws of physics and the symmetries in physical data make it compressible into small neural networks, tying physics to network structure.

George Hotz of Comma AI gives a 2017 conference talk on lessons from building a self-driving system. He argues about how driving should be defined and why the common approach to autonomous vehicles is flawed.

Retired Boeing Technical Fellow Doug McLean walks through common misconceptions in aerodynamics, drawing on his book Understanding Aerodynamics. He critiques popular explanations of wing lift and rocket thrust and argues for arguing from the real physics.

Andrej Karpathy walks through how GPT assistants are trained, from tokenization and pretraining to supervised finetuning and RLHF, and gives practical prompting strategies and a view of the tooling ecosystem.