The World Wide Web, restored to 800 pixels of glory.
GeezerVison is showing htdocs.dev — Posts | htdocs.dev
All the articles I've posted.
Building a curated list of Agentic Development Environments meant verifying a hundred repositories by hand — and discovering that the usual way of spotting what is rising on GitHub has quietly stopped working
macOS virtualization stopped being about running Windows. Apple container, OrbStack, Lima, Tart, UTM, QEMU, and why the VM became an allocatable unit again.
Frontier models are converging into commodities; your accumulated context is the one input nobody else has. The personal knowledge stack now exists — format, capture, sync, protocol — except for the one layer nobody has an incentive to standardize: governance. It's the last moat left, and whoever owns it owns the model of you.
the agent harness is the new browser, MCP is the new HTTP, and the hard problems have moved from the loop to the tool layer, context management, and federated capability.
The architecture behind the most important leak of AI history, and probably the inflexion point for agent orchestration
The complete landscape of multi-agent coding orchestration — from Claude Code Agent Teams to Oh My OpenAgent, the Ralph Loop pattern, and how to choose the right tier for your workflow
"The analogy from the ecosystem that resonated most: Claude Code is "Rails" — opinionated, structured, with best practices baked in. Pi is "Arch Linux" — giving you the smallest kernel and the greatest freedom. "
A practical guide to OpenClaw — the open-source AI agent framework that connects LLMs to your operating system, messaging apps, and tools. Understand its architecture and how it can automate your workflow.
A practical guide to Claude Cowork — Anthropic's desktop AI agent for knowledge workers. Learn how it automates document processing, file management, and multi-step workflows without writing a single line of code.