Day 2,789 of not building a single page app.

I would like a refund.

You KNEW I was going to post this tonight. The San Antonio national anthem as sung by Augie Meyers. youtu.be/1b6z4Z3vSnU?... RIP Augie!

architecture:

I built another javascript framework for the UI. signals, effects, computed but with the philosophy of "build just a little bit on top of the existing browser API and DOM".

Building a new client side framework. It has signals because AI is coding it for me. I remain skeptical.

I building an app that embodies my system for accomplishing missions (oplog.today). What it's teaching me is that it is not the place to keep a list of books that I want to read. So I'm going to use bsky to keep #mabooklist.

#mabooklist You're not listening – www.journalistkatemurphy.com I don't remember :( where I heard about this book. But I think a lot of people are just not listening.

#mabooklist Play Nice: The rise, fall and future of Blizzard Entertainment – www.hachettebookgroup.com/titles/jason... This one comes by way of zachbeastly1.

#mabooklist Team of Rivals: THE POLITICAL GENIUS OF ABRAHAM LINCOLN – doriskearnsgoodwin.com/books/ Paula recommended this one. The context was a startup.

#mabooklist Project Hail Marry – andyweirauthor.com My friend Beau recommended this one.

Sure, 10x your productivity at work, but please don't carry that corporate mindset into your real life. It's ok to slow down and build meaningful relationships.

This is a cross post (from LinkedIn). Tell me if it's stupid to do that. Anyways, #TIL I REALLY need to work harder to control scope when coding with AI. I mean, I need to do that regardless, but the AI can make really large change sets all in one go that it's super easy to let the scope run.

You can say the moon is the sun all night long, but that don't make it so.

Bitch. I. said what I said. www.youtube.com/watch?v=m4_9...

There is a common pattern when teams get under pressure to deliver. Managers abandon task management and project planning tools, pull up a document, list what needs to be done today, and use it to walk the team through what everyone is actually working on. oplog.today Would love your feedback.

Personal operational log oplog.today

`nom install` yourself a chat app. github.com/hubot-friend...

Implementing commands in the evolution plan drafted 9 years ago for Hubot. Just boring automation power. github.com/hubotio/hubot

I'm spiking out API performance benchmarking tools – k6, jmeter, locust. I thought I was going to favor k6, but I'm favoring locust. The default report shows the p50/90/95 for each individual endpoint call, while k6 does not. So I have to write code to generate the report I want when using k6.

I think it's weird that nobody has built a chat app with Hubot since it's open sourcing in 2013. So I built one today: github.com/hubot-friend...

While the world is figuratively on fire, I'm thinking about project management and tools like Jira, Monday.com and their ilk and why we "hate" them. joeyguerra.com/blog/2026/wh...

digg.com is down.

I am super bothered by these huge corner radiuses in the latest MacOS Tahoe version.

Kaizen UI: ui.gokaizen.io UI components we use in apps we build.

I can hold two truths: - “This system is flawed” - “This system is worth defending” That tension is adulthood.

If I can’t improve a situation, I won’t amplify it thoughtlessly. Some things don’t need to be reposted on the internet. You can just pass right by them without giving them any more energy.

History isn’t saved by heroes most of the time. It’s saved by people who refuse to let fear turn them into something unrecognizable.

Snake years aren’t optimistic. They’re sobering. They’re about: - Seeing clearly - Realizing what assumptions were false - Letting illusions die - Understanding stakes

I've been thinking a lot about how many career decisions look "obviously right" on paper… and still end up wrong a few years later. Titles, pay, logos, scope - they feel like advantages. But a lot of them are just big, visible signals, not actual leverage. joeyguerra.com/blog/2025/pe...

Most organizations don’t have a strategy problem. They have a diagnosis problem. Good strategy doesn’t start with solutions. It starts with understanding the real problem. Do you like my LinkedIn content? My strategy is to post bullshit that’s not bullshit.

I love Rick and Morty. So I don’t know why I waited this long to watch Violent Night.

The risk is building a beautifully reasoned system before people are emotionally invested in the problem.

I’m thinking about wearing this everyday through Christmas

Doobie doobie doo.

A leader, in talking too much - taking up all the space in the room, creates more confusion. It's in the prescription. By prescribing all the steps, you take away people's power to collaborate.

It's all fun and games until you have to spin up more EC2 instances to go faster.

- Discovery never stops - Planning should decrease commitment (refine scope ya'll), not increase it - Code can be a thinking tool, not just a delivery artifact That's heresy in Jira-land.

Do your systems allow exploration without punishment?

Use tactics like GitHub Scientist (shadow traffic: prod response returned, experiment logged) to test safely in production (we did this. "Going live" was not a thing because we were already live, every day, until the last day when all the checks were green, then we left).

Make Experimentation Safe Your existing culture resists experimentation - counter with architecture.

Kevin McAllister as a grown up Batman like dude.

And then? And then?

Remember when cousin Eddie went to Clark’s bosses home and dragged his sorry ass out of bed, marched him over to Clark’s to answer for cancelling this years Christmas bonus? Yeah. Like that.

Was Skynet in an AI data center? Asking for a friend.

I love that little flash a website makes when you click on a link and it goes to the next page.

This is the most amazing thing! 😆

The irony of the GitHub outage is that git is decentralized.

Stay positive ya'll!