Futurelock: A subtle risk in async Rust(rfd.shared.oxide.computer)

407 points by bcantrill 11 months ago | 218 comments

Oxide Computer RFD 609 by David Pacheco describing "futurelock," a subtle deadlock risk in async Rust found in the Oxide control plane.

Transparency in Hardware/Software Interfaces(rfd.shared.oxide.computer)

22 points by bcantrill 1 year ago | 3 comments

Why Oxide Chose Illumos(rfd.shared.oxide.computer)

208 points by kblissett 2 years ago | 138 comments

Oxide Hiring Process(rfd.shared.oxide.computer)

49 points by jryb 3 years ago | 47 comments

Has anybody seen my keys? A key-hierarchy strategy for rack-level security(rfd.shared.oxide.computer)

45 points by cyb0rg0 1 month ago | 3 comments

CharlotteOS – An Experimental Modern Operating System(github.com)

156 points by ementally 11 months ago | 81 comments

Seeing like a software company(seangoedecke.com)

327 points by praptak 1 year ago | 96 comments

Sean Goedecke essay applying James C. Scott's Seeing Like a State to software companies, contrasting legible tracked work with illegible essential work.

We have used too many levels of abstractions(unixsheikh.com)

674 points by riidom 3 years ago | 427 comments

An opinion essay on unixsheikh.com arguing that the software industry relies on too many layers of abstraction, leaving few people who understand how the underlying technology works.

Training LLMs from ground zero as a startup(yitay.net)

360 points by swyx 2 years ago | 92 comments

Yi Tay's blog post on the experience of training large language models from scratch at a startup (Reka), covering compute, infrastructure, and lessons learned.

Ideas on modernizing the open-source desktop(lwn.net)

363 points by signa11 1 month ago | 449 comments

LWN.net article reporting on Scott Jenson's Akademy 2026 talk urging open-source projects like KDE to move desktop design beyond the traditional WIMP model.

The Lost Atomic Update on Loongson CPU(jia.je)

120 points by jiegec 1 month ago | 6 comments

A blog post investigating a hardware erratum on the Loongson LA664 CPU where an atomic instruction loses updates, causing an infinite loop while packaging normaliz for Debian.

GLM-5.2 is a step change for open agents(interconnects.ai)

344 points by vantareed 4 months ago | 197 comments

Nathan Lambert's Interconnects post arguing that the GLM-5.2 model marks a capability threshold for open-weight agentic AI.