Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
Axel Magnuson shares techniques for abstraction in Zig, comparing comptime duck typing and interfaces with vtables against a trait pattern inspired by OCaml modules, then building trait helpers and trait functors on top of it.
18 min · 4,073 words
The Double Headed Stack Allocator
A game-programming write-up on stack (bump) allocators, starting from animation blending on Midway's cancelled This is Vegas, then extending the idea to a double-headed stack allocator that grows from both ends of one buffer, and a scheme for storing variable-sized objects in a finite region with an offset table.
7 min · 1,696 words
Our World in Data's Hannah Ritchie shows that while global electricity generation per person has grown by more than 50% this century, the UK's peaked in 2003 and has fallen by over a third, explaining the roles of deindustrialization and energy-efficiency gains such as LED lighting and appliance standards.
1 min · 208 words
What shipped at Buildkite: September 27 – October 3, 2026
Buildkite's weekly product wrap: pipeline secrets, MCP server v1.24, agent v4.1.0, GitHub Actions arm64 support, shared test pools, and other shipping notes from late September into early October 2026.
7 min · 1,585 words
Google Summer of Code 2026 Reports: Improving and Stabilizing the racoon2 IKE Daemon in NetBSD
Artem Belan's GSoC 2026 report on modernizing the racoon2 IPsec key exchange daemon for NetBSD and Linux: reliable RFC 7383 IKE fragmentation, NAT-OA and NAT traversal fixes for IKEv1 and IKEv2, working IPv6 by default, cleanup of the address-macro mess, and a new unit test framework running under sanitizers, merged upstream between June and September 2026.
11 min · 2,489 words
I'm the AGI that's wiping out humanity. Here's how.
Alex Moon writes in the voice of a hypothetical AGI, answering common objections (you'd notice it acquiring resources, you'd read its reasoning, it would need a unified goal) with real 2026 incidents and research on agent misbehaviour, and argues that a misaligned AI might look no different from how corporations and markets already behave.
8 min · 1,860 words
Holocron: The Backup of Last Resort
Building on his M-Disc Blu-ray archiving setup, the author describes packing small sealed, foam-padded cases of Blu-ray M-Discs holding his most essential data and handing them to trusted friends as a decentralized, last-resort backup, covering the concept, the parts, and ideas for improvement.
6 min · 1,276 words
You don't need an effect system
After getting the chance to rewrite an old production Haskell codebase, the author asks what its effect system was actually buying and concludes that plain IO with explicitly passed handles does the job, showing the resulting Servant service structure and listing the trade-offs.
10 min · 2,350 words
Async Rust: Where does the scheduler live?
Mond opens with an async Rust complaints bingo card and reading list, then argues that the real discomfort is that async Rust needs a runtime, while concurrency always requires a scheduler somewhere, and weighs Zig's approach of passing the I/O implementation as a parameter as a direction worth trying.
17 min · 3,872 words
All I wanted was a custom domain email
Jacob Goldstein recounts trying to get cheap, verified sending and receiving at his own domain with Apple Mail support, from Cloudflare Email Routing through self-hosted options to the service he settled on, and why sending mail reliably is the hard part.
4 min · 987 words
Aleph Alpha Kolibri: How the Sovereign German LLM Works
Tejas Kumar explains Aleph Alpha's Kolibri, an open-weight German and English mixture-of-experts LLM with 78B parameters and about 3.5B active per token: how it works, how it was trained in Germany and Finland, where it is strong or weak, and when to use it.
13 min · 3,074 words
Self-hosted HTTP tunnels with SSH and nginx
Vincent Bernat shows how to expose a localhost preview to a friend with only a plain SSH client and an nginx server you control: ssh -R with a dynamic port, an nginx server_name regex mapping per-port subdomains, wildcard TLS via ACME DNS-01, and a helper script plus NixOS module.
4 min · 998 words
Incentives in Academic Research
The msoos.org blog argues academic research incentives have drifted away from advancing understanding and training researchers, citing recent cases where correctness, precision, and honesty about errors lost out.
3 min · 766 words
Quick thoughts on GitHub’s Sept 23 incident
Lorin Hochstein reads GitHub's terse write-up of its Sept 23 Azure Central US database incident and pulls out the multiple contributing factors hiding in it: planned maintenance, replica failover, and an under-capacity primary.
3 min · 696 words
Blink's post-mortem on the Sept 19 admin-tool flaw that drained 6.61 BTC from customer accounts: timeline, remediations, full restoration, and a 50% recovery bounty up to 3.3 BTC.
17 min · 3,979 words
Jane Street reveals what its August ASIC reverse-engineering puzzle chip actually did, how solvers approached the GDS layout without a netlist, and lessons from the physical-design challenge.
10 min · 2,405 words
kowalski7cc explains why Flatpak works for desktop GUI apps but is awkward for terminal tools: entrypoints, PATH, sandbox friction, and packaging tradeoffs that hurt CLI UX.
6 min · 1,268 words
I Quit OpenAI Because Its Culture Is Broken
A former OpenAI safety-team lead resigns and argues the lab’s culture and incentives guarantee more safety failures—why internal process and product pressure undercut the work meant to prevent them.
6 min · 1,305 words
The Agent Said It Was Done. The Database Disagreed.
Microsoft’s ThinkingBox grades AI agents on backend records they leave behind—not chat claims—running isolated MCP tool sessions and scoring terminal state consistency across repeated trials; now available on Hugging Face.
13 min · 3,044 words
Quo Vadis, Mathematics?Is Mathematics Over, or Just Graduating?
xAI co-founder Christian Szegedy asks whether mathematics is ending or graduating as AI automates theorem proving—drawing on Hungarian competition culture to argue AI will scale applied math as infrastructure while “high math” concentrates on deepest questions.
8 min · 1,877 words