This is one of those moments in history when there are more distractions than at any other time in my career. Today, I ended up installing an app called “Refocus.” Yes, we have a lot of cool tech coming to us, but it’s too easy to get distracted by so many shiny things. It’s also

"I am the bottleneck now" Few more thoughts

Safari Mac "tab group" implementation is clearly superior? I don't get why Chrome / Atlas doesn't adapt it

every week I modify how I code due to Codex updates crazy pace

GPT 5.3 Codex Spark is filling the last gap in manual edits today, agents write most of the code, but the models aren’t perfect (yet), so manual edits are still required. with Codex Spark I can do the "manual" edits through Codex

Reviews for an app with a one time payment. Everyone asks for no subscriptions until they realize that incentives are not aligned.

I couldn't get Codex worktrees to properly handoff to local... ...so I created a skill, type "handoff to local" from the worktree and it works great every time ✅🤷‍♂️

Codex Web UI seems more optimal than Codex App UI? Codex Web prioritizes the idea of a "task to be done" while Codex App is more about "bucketing threads in repos." Tasks-to-be-done feels more correct as the agents are starting to be so good that it can just do "tasks."

makes me rethink my spreadsheets in fact, I’ve been migrating, I just didn’t realize that until now

Guillermo Rauch True

I created a Codex skill that utilizes OpenAI's "PLANS.md" or "ExecPlan" ...and it is making Codex work longer with no effort on my part. no idea if results are better (yet) developers.openai.com/cookbook/artic…

many gold nuggets in there

I kept thinking about this problem every time Ben Thompson mentioned how consumers don’t pay for productivity tools (businesses do)

TIL: when I use Codex Cloud (ex. from iOS ChatGPT app), the "task" appears as a thread in Codex macOS app

Terence Tao spent a year at the Institute for Advanced Study - no teaching, no random events of committees, just unlimited time to think. But after a few months, he ran out of ideas. Terence thinks that mathematicians and scientists need a certain level of randomness and

nothing soothes a dev quite like hearing that delaying marketing is totally fine