data-oriented programming

Brian Will makes the case for procedural programming over object-oriented programming, arguing that OOP's promised encapsulation and decoupling are largely illusory. He analyzes real code examples and the problems of tying data to behavior in classes.

Brian Will follows up his earlier anti-OOP video by walking through four short code examples, including Java code in the style of Uncle Bob, to show how object-oriented design complicates simple programs. It argues that OOP hides and scatters state and adds needless indirection.

A short clip of Casey Muratori from Handmade Hero explaining why he rejects the object-oriented mindset and how to move away from it. It is part rant, part programming philosophy.

Tantacrul examines sonification, the practice of turning data into sound or music, and critiques how scientists and musicians, including NASA and CERN projects, use it well and poorly.

Linus Tech Tips documents a major RAID failure on its Whonnock server and the attempt to recover the lost data with a data recovery firm.

Casey Muratori shows how typical 'clean code' object-oriented patterns can cost large factors in performance. He benchmarks them against simpler data-oriented code.

Tom Scott recounts the worst typo he ever made, a subtle mistake in code with destructive consequences at a previous job. It is a cautionary tale about irreversible operations.

Physicist Angela Collier rants for 42 minutes about why violin plots are a poor way to show data, citing papers on violin and bean plots. It is part lecture on how to read distributions and part opinionated critique.

Jonathan Blow's hour-long livestreamed response to a RustConf 2018 closing keynote about entity systems in Rust. He argues that index-based entity systems amount to manual memory management that sidesteps the borrow checker, and he talks about friction for game developers.

Patrick Boyle examines why birth rates are falling in nearly every country and argues the standard explanations do not hold up. He presents data suggesting fewer couples forming is a major driver.

A long video essay arguing that Koko the gorilla probably could not actually use sign language. It examines the cherry-picked data and the lack of transparency behind the claims, drawing on published critiques.

CityNerd lays out a data-driven method for deciding where high-speed rail should be built in North America and identifies 56 city pairs that justify it, based on population and distance.

Not Just Bikes argues that North American transit struggles to attract riders because of the land use around it, not the transit itself. It uses examples such as a London, Ontario line with unreachable stations.