John Mitchell

You know what the biggest problem with pushing all-things-AI is? Wrong direction. I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do my laundry and dishes.

@neuranne @kadavy calls this the "Barf Draft": a brain dump of words onto paper/computer, to immediately be ignored, but the *brain* helps get focused onto the next draft.

When you’re stuck, ask yourself: what’s the simplest, worst version of this I can whip out in 15 minutes? • Free-flow shitty first draft of a blog post • 5-slide presentation with no design flourish • Messy project mindmap in your notebook Put a timer on and give it a try :)

Testing matters! Hiring competent people matters! // Formula 1 chief appalled to find team using Excel to manage 20,000 car parts | Ars Technica arstechnica.com/cars/2024/03/f…

One Second PM: "SORT ACTIONS BY DOWNSTREAM CONSEQUENCES" - e.g. everyone fixes CI/CD or team issues! // Build time is a collective responsibility yoyo-code.com/build-time-is-…

Finding critical bugs in 4 minutes vs NOT finding it in 40k HOURS! This is great! // Deterministic Simulation Testing for Our Entire SaaS - WarpStream - Stream More, Manage Less warpstream.com/blog/determini…

Linting is great for feedback: very fast! Generally narrow scope // API Linting Levels | LornaJane lornajane.net/posts/2024/api…

Monoliths have a *lot* of advantages, like coherent testing // What Is a Modular Monolith? milanjovanovic.tech/blog/what-is-a…

Excellent and useful Git commands, including `sparse-checkout`, great for speeding up CICD! // Modern Git Commands and Features You Should Be Using | Martin Heinz | Personal Website & Blog martinheinz.dev/blog/109

Cookiecutter has been very successful in Python world. Now there's a DevOps equivalent! <3 // Introducing Boilerplate. An open source, cross-platform project… | by Yevgeniy Brikman | Mar, 2024 | Gruntwork blog.gruntwork.io/introducing-bo…

Years of combing through player feedback for RimWorld made a huge difference. No intermediaries; we just read the forum posts and Discord messages directly. See enough data points and you can discern the trends pretty clearly without formal metrics.

@mathsppblog 100% of my projects for 10+ years use Makefiles. It's a no-brainer way to raise the level of abstraction. I can think in high level "run tests" vs low level "lint and then pytest but exclude these directories" kind of thinking!

Do you ever use Makefiles when working with Python projects?

An online Caddy playground, how cool is that 🙃

Alex gives tons of real-world ways to have fast, secure, efficient builds! Running two months of tests gives lots of insight :) // The state of Arm CI for the CNCF actuated.dev/blog/cncf-arm-…

getting container resources right is valuable but tricky. Here's a great solution! // Right sizing VMs for GitHub Actions actuated.dev/blog/right-siz…

The Joel Test 2022 – The Coding Craftsman codingcraftsman.wordpress.com/2022/03/04/the…

OMG you can interactively control and *debug* Lambdas! <3 // Debugging AWS Lambda Functions with a Reverse Shell · Bits and Cloud bitsand.cloud/posts/reverse-…

What are the business consequnces of things breaking in your cluster? Invest! // Learnings From Our 8 Years Of Kubernetes In Production| by Anders Jönsson | Feb, 2024 | Medium medium.com/@.anders/learn…

Tests (and all types of Feedback) are an *investment*. If the business doesn't believe they'll pay off, then that's fine. Tests aren't free. // Too much of a good thing: the trade-off we make with tests | nicole@web ntietz.com/blog/too-much-…

@generativelight gorgeous! <3

finally! a little breakthrough on texture for these generative mountain ranges