By Daniel J. Siegel and Tina Payne Bryson

Rating: 4 / 5.

While this is very much written for raising kids, it’s really about all of us. It’s an accessible, illustrative guide to understanding how the brain works and develops and using that knowledge to encourage habits and behavior that benefit us individually and collectively.

It got a little repetitive at times and I wish it had moved a little faster, but neuroscience is always interesting to me and I liked the emphasis on practical application. I particularly liked the metaphors—a river, a hub of “centeredness” with spokes and outer edges we all tend to get stuck on—the authors used to visualize behavioral balance and how it maps to brain activity. I tend to remember simple, visual ideas fairly well.

The book spends most of its time on the individual before looking at how “mindsight” (awareness, empathy) connects us to other people and leads to a more collective harmony.

A friend’s therapist mentioned liking this book and it’s easy to see why. Worthwhile read whether you have kids or not! I would have finished it much more quickly if not for a reading pileup and I’m digging my way out of.

By Karen Hao

Rating: 5 / 5.

This was a fascinating and unsettling read.

I’m already aware that technology can redistribute power and leave people worse off, so it was with great hesitation that I eventually started figuring out what to do with coding agents like Claude and Codex.

I realized before finishing this book that it’s unlikely that there will be a future where I’m relying on anything from OpenAI and feeling okay about it.

Hao’s detailed and far-reaching account starts with the early days of the company and it seemed like the book was going to continue straight up to the day I finished reading it. (Which isn’t a thing because of how book publishing works.)

I didn’t want it to end. It seemed like a clear-eyed, accessible view of the moves OpenAI has been making and the complex dynamics that ultimately bend toward rapid exploitation and capitalistic dominance. You know, journalism rather than marketing.

The critiques that I lay out in this book of OpenAI’s and Silicon Valley’s broader vision are not by any means meant to dismiss AI in its entirety. What I reject is the dangerous notion that broad benefit from AI can only be derived from—indeed, will ever emerge from—a vision for the technology that requires the complete capitulation of our privacy, our agency, and our worth, including the value of our labor and art, toward an ultimately imperial centralization project.

Frankly, this all sucks.

I’m tired of charismatic people with grossly self-serving aims that don’t seem interested in changing. I’m tired of unceasing growth being a common ideal. I’m tired of hyped products and services just obscuring the harm they do and so many people not knowing or caring.

The epilogue offered a heartening glimpse of things going another way, and I was grateful for that.

This feels like an important read for any modern person and I thought Hao did a nice job making sure you don’t have to be a tech person to follow along with the story.

If you’re solidly in a pro-AI camp, particularly a developer using LLMs in your work, I’m curious what sort of thoughts you’d have after reading this. I imagine a lot of people would need to willfully ignore it, and that’s why I’m nervous about our collective future and I hope I’m too pessimistic and ultimately wrong.

What are we doing here?

I’m glad you asked! I’ve been enjoying a new mechanical keyboard I’ve wanted to write about, but a review felt out of place and boring. A Q&A seems fun and strange, so I’m going with that.

Why should anyone care what you think about mechanical keyboards?

They probably shouldn’t, actually.

I used whatever clunky keyboads came with my first Windows computers and computer lab Macs. Then years of super-thin Apple keyboards interrupted by a Logitech MX Keys that was pretty much the same thing, a WASD CODE keyboard with cherry MX clears, a Keychron K1 with red switches, and now this new thing.

I’ve dabbled and always given up and retreated to the default, non-interesting keyboards for one reason or another.

How awkward. Let’s get this over with: what’s this new keyboard?

It’s a NuPhy Air75 V3 with Blush nano switches.

I can see you’re not going to elaborate. Why this one?

A friend stirred my mechanical keyboard curiosity again, which seems to happen every few years. A smaller desk footprint would be ideal for less mouse collision, and I’ve never actually had a nice mechanical keyboard.

(For the uninitiated, red switches are typically quiet and linear, meaning least sonically annoying and absolutely no tactile personality whatsoever.)

I decided that a keyboard represents the most significant tactile experience I have with the computer I spend lots of time on, so improving that feel would very likely increase my productivity and solve all my life’s problems.

Reviewers seemed to generally like this thing and it’s not crazy expensive, so I picked the quiet switches and went for it.

And how is it?

It actually feels great to type on.

Reviewers like to throw around the word “creamy” and it makes me wonder what sort of experiences we’re all having with cream because that word would never cross my mind otherwise. But it is smooth and pleasant and there’s something about a complete lack of secondary movement or hollowness or wobble that’s quite satisfying. Not unlike going from a car with a creaky, bouncy suspension to a sports car that seems to smoothly and capably eat bumps and stay tight to the road. I can return a library book just fine with either one, but there’s a tactile experience with the latter that leaves me faintly sad to stop driving.

Unlike my K1 with wobbly keys and the slightly warped base it came with, this NuPhy keyboard feels nicely dampened and utterly solid. The movement is all in the keys with none in the base. When I really get going the sound of it is a treat, in the same rhythmic ballpark as gentle rain on a roof.

I was also excited about the glorious little knob, to have a quickly-accessible physical control for volume and muting. Unfortunately it was disappointing and nobody seemed to mention that in reviews. The knob is stiff and unsatisfying to turn, and so close to its neighboring keys that I often end up bumping them. I replaced it with a regular key.

I don’t care much about the backlighting, but the little bars on the sides are useful and the battery life is great. Mostly I just like the feel of typing on it, enough that I’m willing to type out my password roughly a thousand times per day in lieu of Touch ID.

I can control how long it waits until going into sleep mode, but even with its defaults I spend less time waiting around when I get back to my desk for some computering.

It takes up less space on my desk, it sounds great, and I like staring idly at it.

I think even these fancy key caps are going to get worn and shiny by my caustic fingertips, but I’m starting to accept that it’s just a thing that happens that no amount of cleaning is going to solve. No material known to mankind is going to withstand the erosive force of my fingers.

What did you want that knob to feel like?

Smooth and heavy with lots of rotational resistance, like an old stereo receiver. Ideally a solid hunk of metal with a knurled barrel.

Was that was a realistic expectation at this price point?

I suppose not, but I’ve been surprised before.

Why do you prefer boring switches?

I’m a quiet person that works from home where I live with another quiet person, and I’ve previously worked around people with jackhammer keys and wanted to strangle them.

People seemed to like these “Blush nano” switches and I thought I was daring to move away from reds. Since it’s just us here I’m willing to admit that I’ve wondered what blues or browns would feel like in this keyboard. Would I want to strangle myself? Would I adapt to the noise and have a blast with key presses that feel more crisp and distinct and less mushy?

I’ve had one of those little switch tester sets for a while, which was a nice way to try them without buying multiple keyboards, but one switch can only tell you so much.

I love that this keyboard’s switches are hot-swappable—that’s a first for me—and I have some quiet tactile ones picked out that are sadly out of stock.

I’d love to let my hair down and try some not-silent switches and see if I still recognize myself a week or two later.

So would you recommend this keyboard?

For anyone thinking about it that’s on the verge, yes.

However!

I also thought about the newer Node series and I probably could have gone with that, saved a little money, and been just as happy.

It’s still low profile. Its base is not aluminum, but I honestly doubt that matters. More than one reviewer grabbed and twisted the keyboard base to illustrate how sturdy (or not) it was, but who ever does that for any reason in real life?

Instead of a disappointing knob, the Node has a touch-sensitive control I imagine feeling equally clunky and unsatisfying. If I’m wrong about that, great! If I’m not, it’s at least visually hidden and easy to ignore. The Node otherwise seems to be a similar keyboard that’s just as handsome, and I’m intrigued that they offer 3D-printable accessories for it.

Did the keyboard solve all your problems?

No. But I’ll for sure find something else to buy that will.

By Timothy Snyder

Rating: 4.5 / 5.

It was chilling to read this knowing it was written almost a decade ago, which further underscored that we’re not experiencing anything new.

Historically, we’ve seen all this before. We know what to anticipate, and we know what we can do with our mental framing and our actions to resist.

You might one day be offered the opportunity to display symbols of loyalty. Make sure that such symbols include your fellow citizens rather than exclude them.

Snyder’s most powerful points were made simply by recounting how authoritarian regimes rose to power and the effects they had on their neighbors. How deftly a populace could be turned on itself, and how vital it was that some were able to recognize what was going on and stand against it.

This was a heavy read, but many practical takeaways left me feeling contemplative and faintly optimistic if not patriotic.

Having old friends is the politics of last resort. And making new ones is the first step toward change.

I’m finally launching a brand new Eleventy site at this domain, combining my shy Kirby blog with the former Astro site. It could be a mess for a little bit particularly if you’re an RSS reader, and I apologize in advance. You’ll probably notice some “new” posts I merged in from the old side blog.

I’m excited to share this with you, though!

I’ve been anxious about having my “weird garden” side blog posts appear alongside those normally on this site, and I’m trying to get over that and simplify. I hope you can deal with me being weird and rambly.

Nick Simson introduced me to Vulf Mono ages ago and I’ve desperately wanted to spend more time with it. I finally bought a license and it has not disappointed.

I’ve been inspired by Nicolas Magand’s posts where he seems to relish identifying and removing whatever’s unnecessary. There’s still a lot crammed under the hood here, but it’s trending simpler and smaller.

The layout is weird and off-kilter and simultaneously retro and modern, which feels right for where I am right now. I’m sure there are rough edges to grind down and things I’ll need to fix, but I like it and I’ll have fun improving it.

It’s not just some merged posts with a fresh coat of paint, though!

Features

New Layout

Vulf Mono and Vulf Sans are doing most of the work here, and they’ve been a joy to work with. It was tough to cut weights and keep things somewhat reasonable.

I went for a monochromatic look that leans hard on the type and plays with hash mark and halftone effects that attempt to keep it feeling like some kind of odd anachronistic zine.

I’ve got no top navigation, which feels daring if not stupid. I reorganized so the footer is more focused and reinforces a gentle site hierarchy.

Simpler

This project’s package.json file declares a total of nine dependencies. Vanilla CSS and no Tailwind, simple bundling.

Locally, npm install takes 3 seconds. A cold build of 380 pages, including image transforms and processing, takes 14 seconds. (Less than 2 seconds once the cache is warm.)

I was using Markdoc in my Astro build so I could neatly add my own markup for images and pull quotes in posts. That added another post file type (.md and .mdoc) along with more complexity around RSS handling and footnotes. I’m back to simple Markdown now, and thanks to a handy little API I can dynamically inject a set of macros everywhere and very simply use them in my Markdown. For example:

---
title: Example Post
date: 2026-02-18
---

Pretend post, here! I want a nice quote with attribution:

{{ macro.quote(
  "She thinks I’m too critical. That’s another fault of hers.",
  "Lucille Bluth"
) }}

Oops now I want a photo with a caption:

{{ macro.picture(
  src="/assets/images/butte-sunset.jpeg",
  caption="Sunset seen from the trail on Pilot Butte.",
  alt="View from Pilot Butte looking west, sun setting in a hazy sky."
) }}

Look at those optional named arguments! Nice when there are a lot of them.

Tiny Back End

My Kirby site had a guestbook and mileage log, both of which were easy to manage with a handsome, mobile-friendly CMS. The Astro site also had a Letterbird contact form.

This is an entirely static site, which means user-submitted forms are trickier.

I didn’t want to over-use Cloudflare Workers and get too wrapped up in Cloudflare-specific features. I’m already using workers for open graph images and proxying the form submission URLs, along with their handy _redirects.

I decided to make a small Laravel app for collecting form details, sending email notifications, and letting me quickly approve guestbook entries and add mileage logs. There’s a simple admin UI that uses Terminal CSS.

Updates that are reflected on public pages trigger a new build via webhook, so while it’s not instant a new guestbook entry or activity log can appear a few minutes later.

Unnecessary Charts

Each listing page (posts, thoughts, and books) now includes gratuitous charts that I enjoy being there.

The Century Club page still has its bar charts and they use the same style.

Popular & Recommended Thoughts

With a growing body of posts that frankly are all over the place, a newcomer may want to know what to bother with. I plucked some of my favorite posts and whipped up a build-time script to find the most-trafficked posts based on Umami analytics.

External data like guestbook entries, popular posts, and mileage logs still lives in committed JSON blobs in the repository so the site can always build even if there’s an external connectivity failure.

Series Navigation

On my side blog, I’ve published posts in a few short series. The titles were really long and they had no specific relationship to each other and this always bugged me. I added a better way of handling those so the titles aren’t so long and the posts can automatically get special navigation for more leaping to posts in that series.

Reading Note Improvements

I previously kept my reading status in Kirby and published reading notes in Astro when I was done with a book. I also extract highlights from my Kobo when I can, so finishing a book has involved a fair amount of ceremony and redundant efforts.

Now these things live in one place, and I have a local utility to seed book details using the Hardcover API.

When I add a book to my reading list or finish one to make notes on, I can use the clerk CLI to start a file for it:

❯ ./clerk "it's only drowning"
Searching for: it's only drowning

Multiple books found. Please select one:

1. It's Only Drowning: A True Story of Learning to Surf and the Search for Common Ground by David Litt
2. It's Only Drowning

Enter selection (1-2):

I pick a book, select the status (Interested, Reading, Finished, Abandoned) and the app writes a new Markdown file in my Eleventy project with pre-filled frontmatter and placeholders I need to fill out.

---
date: 2026-02-24
title: It’s Only Drowning
subtitle: A True Story of Learning to Surf and the Search for Common Ground
author: David Litt
publisher: Simon & Schuster
publishYear: 2025
isbn10: 1668035359
isbn13: 9781668035351
coverImage: "https://assets.hardcover.app/edition/32171050/ec4ad90cea62b5b19eaab56a68183b978077c1e9.jpeg"
hardcoverURL: https://hardcover.app/books/its-only-drowning
state: Finished
rating:
summary:
startDate:
finishDate:
note:
format:
source:
pageCount: 304
links: []
---

Seems nice so far, so we’ll see how it goes!

I’m making little adjustments like separating titles and subtitles, and exposing a few more book details on post pages.

Metadata Improvement

I’ve taken the opportunity to audit tags and descriptions and fill in gaps. I still have old book records without any ratings or notes, but I’ve cleaned up some URLs and added missing tags and now that everything is in one place that’s likely to be a trend that continues.

I don’t know why you should care about this, but I’m telling you.

Nunjucks

I’m used to Twig so the Nunjucks templating language feels familiar. When I build with Twig, I frequently use includes like this:

{% include "partial.twig" with {
  heading: "Hi",
  subheading: "Just demonstrating!",
} only %}

In partial.twig, I’ll use comments to document the expected parameters. Everything’s clear and explicit and that only means that no other context is passed to the partial to muddy the waters.

Nunjucks has no concept of including a partial with explicit variables. You can use macros for that, and I have in many places, but I also tried to embrace the global scope and make sure partials find what they need.

I miss how tidy, explicit, and composable Astro’s components were, but my templates also make sense to me and feel comfortably Twiggy.

I have a lot of fondness for Kirby and Astro and the people that make them, and I stalled for a long time not wanting to move away from either one even though I have too many moving parts.

I’m also still nervous about putting more strange, personal posts here but I’m not getting any younger I’d like to stop hiding thoughts off to the side. This could be a Very Bad Idea™ but I’m trusting that it’s better to bring it all together and waste less energy keeping things separate.

If you have any thoughts about what you’re seeing here, whether you love it or you’re sleepless with rage, I’d love to hear from you!

By Alexandra Petri

Rating: 3.5 / 5.

I was dazzled by Alexandra Petri’s humorous and self-deprecating Atlantic article where she attempted to personally handle the functions of a government.

We knew together, author and reader, that it was not a scientific experiment but a cheeky dance with absurdity, but I was charmed the whole way through and impressed by her commitment to the story. It also made a serious point about an enduring topic of national contention.

I was eager, then, to expore more of Petri’s writing in this collection of essays.

Petri continued to be off-beat and funny, and a lot of it felt too far over the top for my liking—the kind of funny that offers too few glimpses of authenticity. The party guest that has me laughing for hours, but leaves me feeling like I never really met them.

Her wit and way with words was still evident, and I appreciated a lot of how she characterized the journey into adulthood and how fleeting moments and memories can be:

I suppose the strange selectivity of memory is half its charm. Our lives are burning houses, and we come running out with whatever we can carry.

Time eats people, always.

I enjoyed how well she could convey some of the people in her life and would have loved to read more of her writing about writing, since she clearly has a knack for storytelling and a passion for the craft.

Turning people into characters does a kind of violence to them. You lose a dimension or two pinning them down to the page. No, you say. Stop. Don’t move. This won’t work if you move. You are the story I tell about you.

I wish more of this had resonated with me, but it was a fun read.

By David Litt

Rating: 4 / 5.

A blog post from Dave Rupert landed this on my reading list, and I got to read it with a friend.

I was reassured before I even got to the first page that I’d be charmed by the author’s self-deprecation, and the book that followed did not disappoint.

I’ve never surfed and I’m pretty sure I’d be awful at it, but it seems like one of those activities that’s rich with metaphors for living life with balance and taking risks and a lot of what this book got into. It wasn’t packed full of heavy-handed takeaways though, it was mostly just a fun and sometimes gripping story to follow.

Not long ago, I’d often felt that the happy warriors I knew were in denial. How wrong I’d been. To love life in a world full of tragedy doesn’t make you complicit. It makes you complete.

I can’t even tell if I like where it ended. It felt real and also kind of unsatisfying.

The true dividing line is between those who embrace the terrifying hugeness of the world and those who fight to shrink it.

Litt’s sense of humor, depth, and knack for storytelling made this a pleasure to read.

BetterMouse is the best option I’ve found for my MX Master 3, and so far it’s working great and a relief to ditch Logi Options+.

Excessive Context

I was just minding my own business this week, not thinking about my simmering annoyance with Logi Options+, not wondering once again why a person would ever need an AI prompt builder for their mouse, blissfully unaware that the already-annoying software depends on cloud service in order for the peripheral to function.

And then Robb ruined it.

His Mastodon post said that a breakage in the internet pipes meant his mouse didn’t work, which is of course needlessly stupid. I’m glad I noticed, however, because @reillypascal kindly pointed out that there are options beyond Logi Options!

I immediately uninstalled Logi Options and began trying them. This is not the smartest order to work in, but I was eager to hurl that software into the sun and I assumed something else would work.

I use an MX Master 3 and I like it.

Newer models have come out, and I’ve been tempted especially as I’ve worn several smooth spots into it with use, but now we’re old friends. I freshened it up with some grip tape and kept on. The wheels still turn and the buttons still click, and even though the grip tape is quite worn now we’ve had a lot of good laughs together.

I use it almost stock, with a higher sensitivity than macOS allows by itself and a press of the thumb button to spread all my windows into tiles with Mission Control. When I lose my place with too many windows open, a thumb boop lets me find whatever I’m looking for. I rely on that regularly, and I’ve never met a nicer scroll wheel in my life. The horizontal one is just there, but the vertical scroll wheel between the left and right buttons is glorious. Solid metal with a nice weight and just enough texture to feel satisfying and be functional. A smooth but tactile scroll, where a strong enough flick can translate into a free spin. After this scroll wheel, every other one I’ve tried feels like joyless, primitive nub made of rubberized sadness.

I really can’t tell you where I am on the tracking and scroll speed/acceleration spectrum. All I know is that I can perceive even the slightest change to these settings, and said change makes it feel like I’m drunk and it’s my first time using a mouse.

A friend coaching me through Escape From Tarkov was once bewildered and enraged at my mouse sensitivity being way too high, though that was a different mouse and a different computer. But now we’re really off track, here.

Field Notes

I tried each option Robb listed, hoping to use my MX Master 3 and get

1. Normal left+right click behavior.
2. Stock vertical scroll behavior, including the little clutch button.
3. Stock horizontal scroll behavior.
4. Thumb button Mission Control.

I use the forward/back buttons maybe twice a year when I remember they’re there, so I don’t care much about them.

LinearMouse was where I started, and while it felt pleasantly simple I couldn’t get anything to happen with the thumb button. So no Mission Control. (Same was true of beta v0.10.3-beta.2.)

Mac Mouse Fix was next, and it didn’t register the thumb button either.

SteerMouse was my winner for the remainder of the day. I didn’t expect it to be, because it was more expensive and more dated-looking, but it registered the thumb button and everything worked like I wanted. Then I had Reeder open and was scrolling its sidebar and something felt wildly off. The scroll rate felt totally different and there was a strange lag to it. Clearly Reeder does some kind of smoothing or adjustment to scroll input in that panel, but it was never weird with Logi Options. I didn’t take notes testing the day before, so I tried all the apps. Again.

BetterMouse was chronologically earlier in my testing loop, but I wrote it off because even though it registered the thumb button I couldn’t map it to Mission Control. It turns out I was just being dull: Button 5 simply needs to be mapped to App Launch → Mission Control. (I hadn’t noticed that it was in the submenu, and yes I tie my own shoes.) BetterMouse has a weirder interface with more options than I need, but it’s working wonderfully and it doesn’t behave strangely with Reeder’s sidebar. I trust sidestepping this weird issue means I’ll avoid other ones I would have bumped into with SteerMouse. (BetterMouse also has app-specific profiles, so if I do have app-specific oddities I’ll have a way of fixing them.) The vertical scroll behavior is ever so slightly smoother now in a way that I actually like, though I’m not sharp enough to explain why.

If you’ve been stuck with Logi Options and you’re about to liberate your beloved mouse, I hope it works for you too!

It seems like every company suddenly puts out a “wrapped” summary, and it gives me a sinking feeling that even our local library borrowed the language. But it also feels very 2025, so I’m following along.

2025 was a year of noteworthy changes for me that felt purposeful and optimistic.

This is coming off of some harder years where I felt more isolated, more unsure of myself and my career, more stuck.

I got a new car that literally lets me go places I couldn’t before, and started talking with a therapist at about the same time I joined improv classes. The crossover between these things has led to literal and metaphorical traction this year, with surprises I didn’t see coming.

A fast friend and one of my favorite people has burst onto the scene, an unlikely relationship that’s blossomed into many new things that include improv, writing, drawing, music, pickleball, shows, friend hangouts, and lots of new little adventures.

We’ve been organizing something we call Friend Practice, which is a weekly, informal improv meetup where we practice and play with the stuff we’ve been learning. I wrote a Laravel app we use to privately coordinate our meetup locations and RSVPs, and it’s simultaneously not a huge deal and one of the most important things I’m doing right now. It’s a safe place for people to connect and be creative, and a heartening number of participants have expressed what a great thing it’s been for them. I’ve never been responsible for something quite like that, a place where people gather. It’s satisfying and something I take seriously. I want it to keep being a safe and supportive space, and I know that’s something that needs care and vigilance. One of the perks of middle age is that sometimes I feel prepared for stuff that comes along, and I can welcome this special thing I never anticipated having in my life.

My partner abandoned an increasingly-depressing job search to pivot into an entirely different career. That’s taken a lot of energy, and I’m glad that we’re each able to be supportive and honest and somewhat adventurous with the change.

Work for me has never been so strange. I’ve been learning a lot, experimenting with a lot, and spending time with Claude and Codex instead of ignoring them entirely. I’ve gone an unprecedented length of time without much work, then had surges of new and different sorts of projects. It worries me that my greatest period of financial instability is now, not anywhere else in the almost-two-decades I’ve been building stuff professionally on the internet. The good news, the part that I’m deeply happy with, is that I’ve mostly managed not to freak out. (Just-starting-out me would be a much bigger mess.) I didn’t know what would happen, but I had a productive, connected, increasingly-energetic year in spite of it. I did not—and I’m talking directly to you now inner worrier—die alone under a bridge.

My improv comedy efforts led to being on a regular team that performed in shows. People around town sometimes recognize my face because of that, usually with a smile because we made them laugh. That matters a lot to me, because if you and I can make each other laugh then I think we can figure anything out no matter who we are to each other. Humor is something I need for myself, and something I can share with others. We can feel light, even if it’s just for a moment, no matter what we’re carrying.

I also made the difficult decision to leave this improv team, for reasons that are varied and complex and not worth getting into in a blog post. It was hard.

I started talking with my dad for the first time in fifteen years. I won’t delve into that here either, but it’s been good.

Speaking of things I have a hard time saying publicly, I made the choice to sell a rifle I didn’t feel great about owning and turned the proceeds into a small camera I’m much happier shooting with.

I also sold my beloved custom gaming PC, finally making the choice to get rid of a fun thing and spend my time elsewhere. I’ve waffled about this for years! It was damaged in shipping, and my first shipping insurance claim has been a whole journey. I’m thankful for a patient, understanding, and helpful buyer though.

I’ve done a better job with my body this year. I reached my goal of walking 100 miles on Pilot Butte, and I’m still at it in little spurts. I walk more with friends because it’s free and it’s a nice way to spend time with anyone at all. I’ve gotten back into lifting dumbbells and pushing myself because the discipline and more sturdy upper body feel good. I finished physical therapy for a hand injury, got a vasectomy, and got the remaining ear pierced. No tattoos yet, but I’m ready when the right ideas arrive.

I started talking with a psychiatrist I really like, and she heard me perfectly and set me up with a mild anti-anxiety medication I can use as little or as much as I want as a tool for managing energy and smoothing out the highs and lows so they’re not quite so draining.

I reconnected in various ways with some old friends, and I’m working on my debut 5-minute standup material that an alarming number of people have insisted they want to see me perform.

I’ll tell you a secret while we’re here: my partner and I stole a fish. We rescued it from a dire situation in an office where it was in bad shape. We spent a surprising amount of time and money giving it a better home and caring for it—my partner much more than me. We’re not sure if it was just old or if it never recovered from the physical shock of its former neglect, but we said goodbye to this fish just before the year was out. It’s buried in the yard next to the frozen-to-death robins she found last winter, with aquarium gravel marking the spot.

There was a lot of movement and change for me this year. I spent less energy on worry and more energy on reaching out, playing with ideas, and trying new things. Not everything went well, but in 2025 the formerly-spinning wheels of my life found some traction and got moving. I’m not a particularly optimistic creature, but I’m genuinely excited to see where I get to go next.

Can I be more open and weird and caring and less afraid? Can I write about things here I’m afraid of being dumb or too much? Can I consolidate my various internet selves into one site or project and not hide things in different places? Can I stay excited about technology and my career trajectory even though I have no idea where it’ll go or if it’ll even work? Can I be quieter sometimes, and louder others?

We’ll see.

Thanks for being here and reading.

I’ve been going on about Coolify because I’ve enjoyed tinkering with it, and after the initial learning curve I’ve expanded the types of projects I can comfortably self-host.

Coolify has built-in backup mechanisms for databases, but I’m on my own for filesystem backups.

“The whole point is ephemeral containers Matt, so if you need to back up files you’re doing something wrong,” you say.

You’re mostly right and you could’ve been more gentle, but hear me out.

Most modern web apps can use cloud storage like Amazon S3 for user-uploaded files, and I often take advantage of that. I do, however, still want to back up local files stored by the main Coolify instance—environment variables, SSH keys, etc. And I do have more than one app running that stores user-uploaded files locally. It could happen to you someday, too.

“Okay fine,” I imagine you saying hesitantly.

Years ago, I wrote shell scripts to manage remote backups using restic. I wanted a pinch of abstraction to set up, run, and mount backups without having to dig through my notes on infrequently-used commands. restic was ideal because it could efficiently make incremental, encrypted backups and store them inexpensively on Backblaze B2. It worked nicely and the setup was easy to recreate in a few minutes on a fresh Laravel Forge server.

I moved from Forge to Ploi and now to Coolify, where it’s easier to spin up Duplicati from Coolify’s service list. I’d set up a Duplicati instance for each server I had local files to back up, mount those host directories into its container, and put it on a schedule backing them up remotely. I’d check occasionally to make sure things were working, because I have made mistakes and learned from some of them.

This was all working fine and I had no reason to change anything.

Then I noticed something called Zerobyte in Ethan Sholly’s post so I had to try it. Sure things are fine over here, but oooooh what about that thing over there?

It didn’t take long to get an instance of Zerobyte running and mount a host directory into it for backing up.

Here’s what my Docker Compose file looks like for the main Coolify VPS:

services:
  zerobyte:
    image: 'ghcr.io/nicotsx/zerobyte:v0.18'
    restart: unless-stopped
    cap_add:
      - SYS_ADMIN
    devices:
      - '/dev/fuse:/dev/fuse'
    environment:
      - TZ=America/Los_Angeles
    volumes:
      - '/etc/localtime:/etc/localtime:ro'
      - 'zerobyte_data:/var/lib/zerobyte'
      - '/data/coolify:/coolify'
    ports:
      - '4096:4096'
volumes:
  zerobyte_data: null

Everything there is stock, and it’s the /data/coolify:/coolify line that mounts the host machine’s /data/coolify directory as /coolify inside the Zerobyte container.

For each server that runs a Zerobyte instance¹, the process is the same:

1. Spin up Zerobyte with host directories mounted into it.
2. Create a user account and store the downloaded restic key in 1Password.
3. Establish those mounted directories as Volumes to be backed up.
4. Set up a remote, S3-compatible Repository where backups can be stored.
5. Create a Backup so that each Volume is backed up to a Repository.
6. Optionally set up Notifications (I use email and Discord) and use them to keep an eye on each Backup routine.

I bolded the labels you’ll see in Zerobyte’s UI. Those are the vital pieces.

The files stored in S3 are all encrypted, so you can’t see their contents or even their names.

I originally set up DNS entries for each Zerobyte instance before I realized I could skip them and access directly via Tailscale with a URL like http://tailscale-machine-name:4096. I’ll only ever access the web UI via a device on my tailnet, so there’s no need to have a public subdomain.

The UI is more straightfoward than Duplicati’s and downright gorgeous, and so far my backups and notifications have been working flawlessly. Despite the project’s current v0.18.0 release, I’ve already appreciated nice touches like the realtime backup display and the ability to test each notification which is naturally useful setting it up.

The only drawback I’ve found so far is that a Volume can’t be set up in a subdirectory. Previously I had Duplicati instances share an S3 bucket by having each one limit itself to a subfolder. Now I just establish a bucket for each Zerobyte instance, which isn’t a big deal.

I’m not even taking advantage of remote mounts or rclone, and I’ll probably improve this setup at some point. So far, however, Zerobyte has been really pleasant to work with!

By Helen Phillips

Rating: 4 / 5.

I picked this out with a friend at the library, browsing the shelves and deciding on something we could read together.

Hum puts us vaguely further ahead in time, where we’re still in the process of wrecking the planet, texting on cell phones, and Amazon Prime (though not by that name or color scheme) is off the rails and/or operating at its peak. Hum is set in what seems like contemporary New York and contemporary Earth in general except for technological advancements that feel like natural, almost-obvious continuations of where we’re already headed.

Just like the things we’re living with now, this is great and terrible.

We follow a mother that agrees to have her face physically altered for an experiment, and in doing so earn money that buys her family time while they figure out how to stay financially afloat. They keep needing things and wanting things, and struggling with and against each other as the story progresses.

The book is split into three parts, and I started to worry toward the end of part one that it was going to be a slow read. Then I found myself fully drawn in and whooshed through the rest of the story.

The plot didn’t rise and fall with such a rush that I ended with the electricity that makes me want to start again. But Phillips wrote characters with such emotional clarity and relatable conflict that I felt invested and perpetually concerned about whatever would happen next.

Many sentences in this book could have been meditations on modern life.

The portrayal of commerce and surveillance rushing into private life managed to be funny and painful.

More than once I felt like the story and the world it drew me into were too close to home, too soon, to common and dreary to be fiction I could escape with and enjoy. But I’m glad I read this, because Phillips dialed up the intensity of some common dynamics to reiterate that our humanity, our togetherness, and our presence matter despite all the raging nonsense.

Cold on the heels of the 2023 and 2024 editions, the emoji list of apps I used regularly in the past year:

📨 Mail Client: Mail.app, MailMate + SpamSieve + GPG Suite
📮 Mail Server: MXroute, CraneMail
📝 Notes: Obsidian, Tot
✅ To-Do: Obsidian
📷 Photo Shooting: iPhone, Lumix GH5, Fuji X100VI
🎨 Photo Editing: Photomator, Pixelmator Pro, Acorn
📆 Calendar: Apple Calendar
📁 Cloud File Storage: iCloud, Syncthing, MinIO
📖 RSS: Reeder
🙍🏻‍♂️ Contacts: Apple Contacts
🌐 Browser: Safari
💬 Chat: Apple Messages, Discord, Slack
🔖 Bookmarks: Raindrop
📑 Read It Later: Obsidian
📜 Word Processing: Apple Pages
📈 Spreadsheets: Apple Numbers
📊 Presentations: Apple Keynote
🛒 Shopping Lists: Apple Reminders
🍴 Meal Planning: Mela
💰 Budgeting and Personal Finance: Soulver + Actual
📰 News: Mastodon + a jillion blogs + 404 Media
🎵 Music: Apple Music
🎤 Podcasts: Apple Podcasts
🔐 Password Management: 1Password
🧑‍💻 Code Editor: VS Code, PhpStorm, Sublime Text
✈️ VPN: NordVPN

Bonus Items

Important enough to include here, ordered roughly by how often I rely on each.

🚀 Launcher: Alfred
🐚 Terminal: iTerm 2
☂️ Backup: Arq + Hostbrr Storagebox, Zerobyte
🚫 Ad Blocking: Vinegar, Wipr 2, StopTheMadness
🔎 Search Engine: Kagi
📓 Journaling: Day One
🗂️ Version Control: Fork
🖼️ Screenshots: CleanShot X
🐘 Mastodon Client: Ivory
👨‍💻 Local Development: DDEV
🗄️ Code Repositories: GitHub
💽 Database Manager: TablePlus
📖 Reading: Kobo, Apple Books
✍️ Writing: Obsidian, VS Code, iA Writer, Beat, MarkEdit
🧾 Invoicing and Time Tracking: Harvest
👨‍🎨 Design: Figma
🕹️ Games: BattleBit Remastered, Hell Let Loose, Star Citizen, Battlefield 6
📊 Web Analytics: Umami
🗓️ Schedule Booking: Cal.com
🤖 Server Provisioning: Coolify, Ansible
🩺 App + Server Monitoring: HetrixTools, Sentry, GlitchTip, Checkmate
📦 Package Tracking: Parcel
🧠 Brainstorming: Obsidian + reMarkable + an actual whiteboard
☎️ Video Calls: Cal.com, Zoom
📋 Snippet Sharing: HedgeDoc
🎛️ 3D Modeling and Slicing: Autodesk Fusion, PrusaSlicer
🗺️ Maps + Driving Directions: Apple Maps + CarPlay
🔥 Wildfire + Air Quality Tracking: Watch Duty
🎬 Filmography Reference: Callsheet
📽️ Video Editing: DaVinci Resolve
🐎 Motion Graphics: Apple Motion
🎹 DAW: GarageBand
👀 Editor Analytics: WakaTime

Bonus Commentary

My trend toward smaller apps and services and self-hosting continues. I didn’t change a whole lot, which is probably good!

I changed my Git client from Tower to Fork after a long trial, introduced Acorn for fast image editing from a developer I’m happy to buy a license from, and leapt into another lifetime email account (😅) I’ve been happy with.

I tried and failed to find something on par with Kaleidoscope for resolving merge conflicts. It feels too expensive for how infrequently I use it, but I can’t beat that editable three-column view with wonderfully-granular smart merging. It really does help me find my way through comparisons and merge conflicts confidently. Which is saying a lot, because I usually have to take a deep breath before reviewing any merge conflict.

I also got excited about Everlog as an alternative to Day One. I ran into trouble importing my Day One journals and the developer was friendly and diligent about fixing the admittedly mysterious issue I was having. I couldn’t even scroll my list of imported entries without the view locking up, however, and I didn’t want to keep being a pain with support requests. I’ll probably try again, because I’d rather get away from Automattic and support an indie developer.

I caved and splurged on an AAA title and enjoyed some long sessions with Battlefield 6, then pivoted sharply and put my gaming PC up for sale. I have too many things I’d like to do and gaming just eats time.

I did some of my first screenwriting with Beat, and inspired by Nicolas Magand’s praise I found that I have a pleasant time writing with MarkEdit.

My Coolify spree continues, with Actual and Checkmate being new self-hosted apps that have stuck around. Zerobyte was a late addition thanks to Ethan Sholly’s article, and it quickly replaced the Duplicati instances that were clunkier to manage. Despite UI issues and a frustrating name change, useSend has been flawless for managing transactional email on top of SES.

My big physical changes have been a Fuji X100VI camera I’m having fun getting to know, and a whiteboard that now adorns my office wall where I can always see it. The whiteboard is my habit tracker, idea scribbler, and a fantastic place for tracking personal and social goals that don’t need to be on my daily Obsidian todo list.

So my killer app this year is a whiteboard I got from Home Depot. Nice contrast, infinite runtime without recharging, no subscription, no AI features, no software updates, no firmware updates, no ads, no data mining, no parent company acquisition and business model shift. This says a lot about my relationship to technology in 2025. Yes, I’m fun at parties.

A little more than a year ago I consolidated most of my self-hosting to Coolify.

Despite my technical wanderlust, I’m happy to report that Coolify has stuck!

Setting up PHP apps is more cumbersome than it was with Ploi, and my Docker Compose deployments come with a few seconds of downtime rather than none at all. But I love that I can host whatever I want, each project with isolated components and easy backups, and throw as many servers into the equation as I feel like.

I often try out new apps, host half-finished projects I end up retiring, and change things in general, but I thought I’d let you know how it’s going. I don’t know why you care, but it’s really nice of you.

Servers

I’ve got myself down to four individual servers, which for me is a feat. I love trying out services from smaller providers and finding reliable hosting that’s more powerful than big ones at a fraction of the cost.

• 2-core, 4GB Los Angeles VPS running Coolify
• 4-core, 6GB San Francisco VPS running Mastodon and Checkmate
• 4-core, 8GB Los Angeles VPS running lots of stuff
• 4-core, 16GB San Jose VPS running lots more stuff

Each is on Ubuntu 24.04 LTS with an EPYC CPU, NVMe storage, and more bandwidth than I’ll ever use.

I keep them on a Tailscale network for SSH access, and manually update their system packages from the command line every now and then.

Many have come and gone, but these four servers are my stable, well-performing, well-connected champions¹. I renew them a year at a time, and combined they cost $16.10 per month.

There’s enough headroom that I don’t have to worry too much about resources or speed. They don’t pool resources unfortunately, but migrating a needy app from one server to another isn’t all that painful.

Apps

These are the apps (“Projects”) I have running right now. Most of them have been live for months if not the whole year:

• Actual for budgeting
• Axiom Syslog Proxy for forwarding logs to Axiom
• Kirby development site
• Bluesky PDS even though I’m rarely on Bluesky
• Checkmate for monitoring site uptime and PageSpeed scores
• Checkmate Capture because I needed to see what Checkmate’s hardware monitoring looks like
• small hosted utility written on the Flight framework
• barebones Craft CMS instance I use for testing my Docker setup
• Dekindler Demo for an old Kindle note parser I wrote
• DocuSeal for document signing
• Duplicati for facilitating remote backups of Docker data
• Feed Canary, my Laravel app for monitoring RSS feed health
• DumbDrop for people to easily send me files
• private Laravel app for managing meetups with friends
• GlitchTip to take place of Sentry for personal projects
• HedgeDoc for sharing words with people
• Kirby for my other blog
• Mastodon for t00t.cloud
• MinIO for various S3-compatible storage
• Monica personal CRM
• Outline as an experimental Notion replacement
• a small collection of flattened client sites
• Typesense for fun and powering this site’s secret-ish search page
• Umami for website analytics
• Unsend useSend for SES-backed transactional email

MinIO and useSend have stood out to me.

I might be the last person to realize how useful MinIO can be since so many projects support S3 storage, but it’s nice to have cloud storage buckets on fast servers of my choosing.

useSend has made it possible to consolidate most transactional email into one place, easily manage and keep an eye on things, pay very little for the few emails I send from these projects, and still enjoy reliable delivery.

Working Impression

Keeping Coolify up to date has been painless. The release cadence has been all over the place, and while there have been a few disruptive bugs communication and fixes have usually been swift. Overall my experience relying on Coolify has been largely stable and without drama—most problems were my own failure to read release notes, or actively derping my way through things like migrating projects between servers.

I migrated the Coolify instance, by the way, to a more reliable and less expensive VPS and it went surprisingly well.

I like that predefined services are added all the time. It feels like Coolify has enough momentum that if a self-hostable app isn’t already available someone will likely have written a post about it by the time I’m looking.

Paul Anthony Webb’s notes about Stalwart led me to Aldert Vaandering’s post about running it on Coolify, and I was close to getting my own email server running before I realized I was about to get my own email server running. I have not one but two perfectly good lifetime email hosting accounts I’m happy with and I’ve been down this road before. I do not need to run my own email server. I already learned the hard way. But I could have had Stalwart running. It’s the kind of thing I keep doing with Coolify.

If you don’t want to be messing with Docker or Docker Compose I can’t recommend Coolify. While you can easily spin up predefined services, hosting your own projects means you’ll need to spend time figuring out how to deploy whatever you made.

If you want to be learning these things, however, and you’re the type to spin up projects just to be able to poke at them and learn, Coolify may be a fun and satisfying tool to explore.

I’ve improved my Docker setups over time. I experimented with Laravel Cloud and that inspired me to get Octane + FrankenPHP working with self-hosted projects.

Broadly, Coolify has fallen into place as a hosting Swiss Army knife, and we’re having a nice time together. No major drama or container that can’t be rebuilt and redeployed. I’ll post an update if things manage to go sideways or I find some newer, shinier thing to play with, but I see myself sticking with Coolify and being happy to watch it evolve.

In a time where big annoying companies seem to have a stranglehold on the internet², it’s nice to have a corner where I’m cheerfully running stuff I like to use and play with.

I’m not a “photographer” as much as a person with cameras that’s fastidious and prone to overthinking. This is where I’m at managing photos in the actual year 2025.

Backstory

In the days of Aperture, I had no complaints.

I imported files from a camera, like the Canon Digital Rebel that replaced my 35mm Rebel 2000, and they went into Aperture’s library. I pruned out the many mistakes, made my edits, and sometimes pushed a few to Flickr.

Then Aperture went away and nothing ever quite replaced it.

Lightroom worked well, but Adobe seemed increasingly hell-bent on moving to subscriptions. (I’m done being irritated with subscriptions, but wary relying on them with words and memories I’ll still want someday when I’m ancient with an unknown income.)

Synology Photos never stopped feeling clunky.

Luminar was okay, but full of upsells and increasingly more excited about its AI features than I was.

ON1 was pretty good.

Photomator was nearly perfect, only missing video previews. I bought a lifetime license a few weeks before it was acquired by Apple, and I assume that if Photomator manages to stick around my precious folder support will vanish.

I haven’t yet mentioned my self-imposed complications:

1. Each application has had its own way of managing source files and metadata, and I’m tired of losing information and hierarchy so I insist on managing my own folders and not trusting big library files anymore.
2. The collection has grown to a point that it’s now on external storage. A fast USB4 NVMe enclosure has taken some pain out of that.
3. I’ve merged my iPhone and camera media into one library, where memories increasingly include videos in addition to photos.

I’m pretty sure a lot of you use Apple Photos because it’s easy. There are lots of nice apps that can work with your Photos library.

Many others use Google Photos or Dropbox because you can get plenty of storage and easy sync and automatic backup.

But really I don’t know what most of you are doing.

I have a spouse, no kids, and one dog. Not a ton of subjects, or a life packed with far-flung travels or “interestingness”—but I like bothering with a physical camera sometimes because the images are better and it forces me to look at everything more carefully instead of reactively capturing moments.

Ever since a dog entered our lives my photo volume has hockeysticked with photos of the creature—often doing adorable things like breathing, sleeping, or looking at me with an empty expression. A lot of you are raising human children, so I can only assume your media pile is vastly greater than mine given the cute and momentous things they do.

How anyone manages photos remains a great mystery, is what I’m saying.

Ingredients

Cameras:

• iPhone 14 Pro photos and video, kept in the usual camera roll
• Panasonic GH5 I use for video projects
• new Fuji X100VI that replaced my old Digital Rebel T2i in a form factor closer to the DMC-LX3 I miss

Storage:

• 2TB NVMe stick in a fast USB4 enclosure that’s always connected to my Mac Studio and backed up with Arq

Software:

• Image Capture (macOS utility)
• ExifRenamer
• ditto
• Photomator

Recipe

1. Go do a thing and take photos and/or video.
2. Use Image Capture to transfer new media from the iPhone (cable), X100VI (cable), or GH5 (SD card) to the ~/Downloads directory.
3. Drop those files onto ExifRenamer, which is configured to write a {year}/{month}/{day} structure in a temporary _sorted directory.
4. Use ditto to merge _sorted into the existing library:
   ditto /Volumes/Example/_sorted /Volumes/Example/Library
5. Confirm joy, then manually delete the _sorted copy.

Then I can edit in Photomator, which is great for viewing and working with photos quickly. I can flag and delete cruft, rate the better specimens, crop and fine tune, and sometimes export for sharing or posting to Mastodon or whatever it is I do.

I pretty much forget about videos until I poke around in Finder and wonder if Photomator will ever add video preview support. (I don’t need to be able to edit videos there, I just want to see that they exist—ON1 had this worked out.)

Meh

ExifRenamer is great, and this is a good enough workflow to merge media from various devices into one library where I have control over its structure, how it’s backed up, and whatever it costs to maintain.

It also feels absurd. I keep flirting with the idea of dropping everything into Photos and letting it manage the library like Aperture used to, and worrying about obligatory iCloud subscription bumps if and when they’re a problem.

Can I trust Photos and its library, though? I don’t want to get hurt again geotagging and writing captions and tagging people only to have those details vanish in a few years when I decide to change software.

How do you manage your photos, dear reader? What do you shoot, and what with? How do you edit? How do you back up? How long have you been using that system, and are you happy with it?

I decided to wipe and sell my beloved Windows gaming PC, Thunder Nugget.

It started as a pandemic project I took on because I was inside a lot and I hadn’t built my own computer since high school.

It’s a whole lot easier to build a custom PC now! You can pop by PCPartPicker to make a plan and find good prices for parts, collect all the boxes that arrive, and assemble everything. You still have to wonder if your anti-static hygiene is sufficient and stress out until you power it on for the first time, so it’s nice that some things don’t change.

Graphics cards are like phone camera bumps though: they’re obnoxiously large and I think we’ve been pretending they’re not long enough that nobody cares.

It was a mini-ITX build, meaning not very big. Air cooled (as opposed to water, which is a real thing if you’re not into this scene), fairly quiet, and no RGB. I wanted a gaming machine, not a clown box.

I played various games. Joined some groups for a while. Let it sit unused for long stretches of time, because I’m not a gamer but a person who sometimes plays games with reckless abandon—then moves on and doesn’t play them at all or keep up with gaming news. I’m not a “good gamer.”

I clocked a slight change in my reaction time with age. I tried different mice.

In the winter, I’d open a window for the free cooling. Thunder Nugget was a space heater running a big AAA title.

Securely wiping Thunder Nugget was hard. Not emotionally, but operationally. (Okay and a tiny bit emotionally.)

I have no idea how Windows works now, apparently, and my instincts were all wrong.

Did I create a bootable Windows 11 thumb drive before wiping the primary disk? You bet I didn’t!

Did I have another Windows machine to use for that? Nope!

Good thing for that little Ubuntu Live stick I cooked up! I’m not sure what I would’ve done without it.

ChatGPT helped me navigate some BIOS adventures, NVMe-wiping commands, and the rather baffling number of problems that came up, but it also sent me down some holes and seemed eager to try too many different approaches to things. I had some harsh words for it and only feel a little bad.

But it’s up for sale now. Liberating it from my overengineered cable management took a while, and the wipe + Windows reinstall took the better part of a day.

I like playing games because I can disappear into adrenaline romps—yes those games—when I want to. For hours at a time, so focused there’s no room to think about anything else.

I’ve struggled wondering if I should get rid of the thing just to free up time and brain space for whatever else I might do. I’ve met friends playing, used the energy from or approaching gaming sessions to get other stuff done, and it’s possible that having fun might just be a good enough reason to do something.

I don’t have any feelings about whether you play games or don’t, by the way. This is just me.

I finally decided that I’m ready to see what happens when I get this thing out of my life. I’ve already felt the pang of withdrawal, but I did some writing and started a show I’ve wanted to catch up on. I haven’t even sold the thing yet, but already I feel lighter and ready for something new.

I’ve been more aggressively cleaning up and donating other things that take up mental and physical space in my life they don’t need to, and that feels good too. Sometimes it’s time for change!

By Stephen Rosenfield

Rating: 4 / 5.

A friend liked reading this and I did, too!

I’m working up the courage to go try an open mic, and there was a lot of motivation and practical advice in this book. Early on there were descriptions of comedy I found deeply satisfying:

By taking the things we struggle with and worry about, and by ridiculing these things, comedy transforms them from overwhelming to laughable. (…) For a glorious moment the comedian lifts our worries off our shoulders and unites us in laughter. The underlying message of comedy is this: You have problems; I have problems. But we’re OK. You are not alone. We’re in this damn thing together.

The spirit of the comedian is vulnerable but indestructible. The comedian looks unblinkingly at his or her shortcomings and utilizes them not to create sympathy or pity, but to create laughter.

The book started extolling the virtues of the medium, for example the fact that standup and improv require nothing but a stage and imagination and how that’s different from most other forms of entertainment. Later, it got into more structured and practical device for developing material, practicing, performing and hosting at clubs, etc.

I most loved this concept of “joyous communication” that was new to me here:

Joyous communication is the single most important technique in performing stand-up comedy. Joyous communication does not mean you’re communicating how joyous you feel. Happiness is a wonderful thing to experience in life, but it is not funny. It’s hard to get a laugh on how great things are going for you. What joyous communication means is that you take joy in communicating to the audience your emotions, be they anger, confusion, outrage, excitement, frustration, love, or whatever else you’re feeling.

This isn’t limited to standup. I’ve been around people that I never tire of because they’re so engaging and connected—the tone and topic don’t even matter. They seem to simply enjoy storytelling and conversation in a way that’s easy to feel and stay with. It would make sense that this is a vital part of relating to an audience and performing good standup. (Or possibly anything.)

The book seemed to repeat itself more toward the end and it re-used a few examples enough that I wished it pulled from a broader pool of them, but I liked the casual tone and apparent authoritativeness of what Rosenfield covered.

I’d recommend reading this if you’re even remotely interested in standup!

I mostly approach what we’re calling AI with skepticism because of who controls it and the impact it has on power dynamics and the planet.

I also realize it’s transforming how developers work, and it was Jeff’s reading list that got me thinking more about it and recently working with AI coding agents like Sketch, Claude, and Cursor CLI.

I’ve already been tinkering with Ollama for a few months and exclusively limiting my AI chat experience to local models. I couldn’t manage to get any of them working well for coding. Or at least not nearly as well as popular commercial ones.

I haven’t asked an agent to build me an app from scratch yet. I’m not sure that I would, because even reviewing minor refactoring and improvements has taken a fair amount of time.

It is novel and impressive to watch an agent evaluate a code base, make a clear todo list, suggest improvements in a short amount of time, run tests, and fix things all on its own. Directing it just like a person is natural. I find that I am polite and use proper punctuation which may be insane.

For me, the time to review with care is still significant. I’m not sure that it’s a massive time saver compared to thinking and coding my way through a problem with only my old brain.

However.

What I really, really like is getting suggestions I wouldn’t have come up with. It could be an architectural improvement or utilizing a framework feature I wasn’t aware of. One such example was using a scoped query in a Laravel model. Not a huge deal, but it looked weird so I consulted the docs and learned that it’s a thing! (And this isn’t a given; Ollama confidently told me I could use some Craft CMS features that don’t exist!) It took Cursor CLI (ChatGPT 5) less time to implement query scoping than it did for me to learn about it and realize the syntax differed from the docs because I unwittingly took myself to the wrong documentation version. Cursor got it right and I made the error.

I agree with something David Crawshaw pointed out, which is that working with an agent provides a boost to momentum that can be powerful. This happens with pair programming too, in my experience: tackling a problem together has its own sort of energy that can be the difference between letting something languish and actually getting it done. My experiments had me making several improvements to different projects—I felt a sense of accomplishment and probably wouldn’t have ticked off the same items on a normal organic day.

The idea I keep grappling with most is this:

Does an intern cost $20/month? Because that’s what Cursor.ai costs.

My first experience working with coding agents has been like working with really smart interns. They do tedious work for me, offer some great ideas, and still make mistakes I have to be careful to either call out or fix myself. This is easily worth more than $20/month.

But what about the intern starting out in the field gaining experience?

What happens to lower-wage developers that had been earning a decent enough living to support themselves?

What happens when we all adopt these tools and moving faster with agents becomes the new norm? (And do we end up with better software, or more complex messes to fix with more complex tools?)

Working with AI coding agents has so far been easy to wade into, and there’s an overwhelming amount of activity in this rapidly-changing space. Knowing how to use these tools well seems like a skill of its own.

Let’s also acknowledge the hype. I know many of you are very excited, but there are a lot of self-congratulatory and breathless videos and we could all calm down just a bit. My hype allergy really flares up around this stuff.

If you’re a seasoned tech bro building software for business, the immediate cost is trivial and the benefit to you is potentially massive. If you’re a hobbyist or a newcomer to the field or not doing work for a massive company, is this good for you too?

Maybe power consumption and model accessibility are things that improve over time. Maybe these tools can truly help everyone and lead to better work and greater understanding and reduce some of the tedium that nobody needs.

I’ll continue to be wary as I see a benefit and wonder who controls the technology and what my use of the thing means for other people.

A quote in Minsuk Kang’s daily sentences struck me:

I’ve found only one metaphor that encapsulates the nature of what these AI power players are: empires.

The older I get, the more wariness I have for massive companies and the impact they have on human beings. I like whatever puts people ahead of business or scale. I can’t tell yet whether that puts me at odds with The New Way™, whether I just have to find my own balance with it, or whether I should get the hell out of here and build things with my hands.

I’ve been using Unsend all over the place now that I’m self-hosting it.

In my Laravel apps I’d started out using a generic SMTP connection before realizing a simple custom transport could connect with the Unsend API directly.

The documentation was straightforward, and the bulk of the effort was mapping Laravel’s message details to a suitable API payload and testing it.

I started by creating app/Mail/Transports/UnsendTransport.php:

<?php

namespace App\Mail\Transports;

use Illuminate\Support\Facades\Http;
use Illuminate\Support\Uri;
use Symfony\Component\Mailer\SentMessage;
use Symfony\Component\Mailer\Transport\AbstractTransport;
use Symfony\Component\Mime\Address;
use Symfony\Component\Mime\MessageConverter;
use Symfony\Component\Mime\Part\DataPart;

class UnsendTransport extends AbstractTransport
{
    /**
     * {@inheritDoc}
     */
    protected function doSend(SentMessage $message): void
    {
        $email = MessageConverter::toEmail($message->getOriginalMessage());
        $apiKey = env('UNSEND_API_KEY');
        $baseUrl = Uri::to('/')
            ->withHost(env('UNSEND_DOMAIN'))
            ->withScheme('https')
            ->value();

        if ($from = collect($email->getFrom())->first()) {
            $fromValue = $from->toString();
        }

        $postBody = [
            'to' => collect($email->getTo())->map(function (Address $email) {
                return $email->toString();
            })->toArray(),
            'from' => $fromValue ?? null,
            'subject' => $email->getSubject(),
            'replyTo' => collect($email->getReplyTo())->map(function (Address $email) {
                return $email->getAddress();
            })->toArray(),
            'cc' => collect($email->getCc())->map(function (Address $email) {
                return $email->toString();
            })->toArray(),
            'bcc' => collect($email->getBcc())->map(function (Address $email) {
                return $email->toString();
            })->toArray(),
            'text' => $email->getTextBody(),
            'html' => $email->getHtmlBody(),
            'attachments' => collect($email->getAttachments())->map(function (DataPart $part) {
                return [
                    'filename' => $part->getFilename(),
                    'content' => base64_encode($part->getBody()),
                ];
            })->toArray(),
        ];

        Http::withHeaders([
            'Content-Type' => 'application/json',
            'Authorization' => 'Bearer '.$apiKey,
        ])
            ->baseUrl($baseUrl)
            ->post('/api/v1/emails', $postBody);
    }

    /**
     * Get the string representation of the transport.
     */
    public function __toString(): string
    {
        return 'unsend';
    }
}

This needs to be registered in app/providers/AppServiceProvider.php:

/**
 * Bootstrap any application services.
 */
public function boot(): void
{
    // ...

    Mail::extend('unsend', static function () {
        return new \App\Mail\Transports\UnsendTransport;
    });
}

You can then define a mailer in config/mail.php:

// ...
'mailers' => [
    'unsend' => [
        'transport' => 'unsend',
    ],
],

Finally, designate the mailer and add your API credentials:

MAIL_MAILER=unsend
UNSEND_API_KEY=us_•••••••••••••••••••••••••••••••••••••••••••
UNSEND_DOMAIN=unsend.example.com

You can see if it works by opening a Tinker shell with php artisan tinker, then using the Mail facade to send a quick test:

Mail::raw('Hi!', function($m) { $m->to('me@example.com')->subject('Test Email'); });

By C.S. Lewis

Rating: 3.5 / 5.

I faintly recall reading this somewhere in late high school or early college and I don’t remember if I finished it. I doubt I did.

A not-religious friend enjoyed the book a while back and was curious how I’d react to it, which made me wonder too. The religious era of my life is long gone, but I finished the book this time around!

I read even slower than usual; for some reason the sentence structure was anything but flowing, so I’d often have to read and re-read to track a thought to its conclusion. It was odd to step around some Christian fundamentals rather than nod along with them, but also worth it because so much about people seemed so accurately portrayed in these letters.

There is nothing like suspense and anxiety for barricading a human’s mind against the Enemy. He wants men to be concerned with what they do; our business is to keep them thinking about what will happen to them.

Plenty of descriptions like this made me smile:

Fun is closely related to Joy—a sort of emotional froth arising from the play instinct.

Hot take on social media (from 1942):

You should always try to make the patient abandon the people or food or books he really likes in favour of the ‘best’ people, the ‘right’ food, the ‘important’ books. I have known a human defended from strong temptations to social ambition by a still stronger taste for tripe and onions.

The entire book is one working devil writing to another and a clever way to elucidate Christianity, but even with framing I’m not sold on there were some lovely characterizations I can readily get on board with:

The Enemy wants him, in the end, to be so free from any bias in his own favour that he can rejoice in his own talents as frankly and gratefully as in his neighbour’s talents—or in a sunrise, an elephant, or a waterfall. He wants each man, in the long run, to be able to recognise all creatures (even himself) as glorious and excellent things.

I wish it’d been a more comfortable read, but it was interesting and ultimately worthwhile.

By Kevin Kelly

Rating: 4 / 5.

Brie Wolfson wrote an essay called Flounder Mode that recently introduced me to Kevin Kelly. I wasn’t familiar with either person but now I’m swooning at both.

The library had a copy of this available and I read it all yesterday. I’m a slow-ish reader so this would be a feat if the book wasn’t very short sentences. It reminded me of How to Live and I felt guilty blowing through it despite all the highlighting.

Draw to discover what you see.
Write to discover what you think.

There was a time in my life where I’d start a day by reading a line from a religious devotional and then sitting with it for a little while. That seems like a more appropriate way to absorb something like this.

To be interesting just tell your own story with uncommon honesty.

Advice ranges from small and practical to huge and philosophical, and picking out a few of my favorite highlights was really challenging here!

If you repeated what you did today 365 more times will you be where you want to be next year?

Very much recommended if you’d like a quick read with some potent advice.

Masto.host has been great in the year and a half I’ve had my Mastodon instance running there. Hugo guided me through a smooth migration process and I stopped thinking about Mastodon hosting and updates. If you’d like to run your own instance and not have to futz with it, Masto.host is solid!

My database and media usage bumped me into an uncomfortable pricing tier, so I wondered how much pain it would be to migrate to an unused 2-core, 2GB Ryzen 9 VPS with plenty of storage.

You already know it worked or I wouldn’t have written this.

In hindsight, two fast CPU cores are okay but 2GB of memory is tight. If you’re used to self-hosting the pieces are relatively straightforward: PostgreSQL, Redis, and containers for a web interface, streaming server, and Sidekiq queue. Mastodon is not lightweight though: it’s a busy app that uses quite a bit of memory—probably more sensible for multiple people sharing a server.

If I had memory to spare I might have added Elasticsearch to the equation, but I know that’s even more of a memory hog so I didn’t bother.

What I Did

1. Created a Coolify project.
2. Provisioned standalone Redis and PostgreSQL services.
3. Added a “Docker Compose Empty” resource with the web, streaming, and queue containers.
4. Got a database dump and environment variables from a Masto.host backup.
5. Configured everything like my live instance and started it up.
6. Added a PostgreSQL role and imported a database dump with Coolify’s GUI.
7. Confirmed that everything basically worked, then stopped my live instance to get a fresh copy of the database and the massive media folder.
8. Failed to appreciate how necessary the 17GB of caches/ were.
9. Ran a bunch of tootctl commands to clean up broken emojis, account images, and media.
10. Added a lightweight Cloudflare Workers CDN for media files.

Coolify Project Parts

There isn’t a predefined Coolify service for Mastodon yet, but armed with Mastodon’s own Docker Compose example and my proclivity for needless adventure, I knocked out the bulk of this in about two hours.

I started making a new Coolify project and provisioning standalone PostgreSQL and Redis services, then a custom Docker Compose setup for the web, streaming, and Sidekiq containers. I hacked down the example to get to this:

services:
  web:
    image: 'ghcr.io/mastodon/mastodon:v4.3.8'
    restart: always
    env_file: .env
    command: 'bundle exec puma -C config/puma.rb'
    healthcheck:
      test:
        - CMD-SHELL
        - "curl -s --noproxy localhost localhost:3000/health | grep -q 'OK' || exit 1"
    ports:
      - '3000:3000'
    volumes:
      - './public/system:/mastodon/public/system'
  streaming:
    image: 'ghcr.io/mastodon/mastodon-streaming:v4.3.8'
    restart: always
    env_file: .env
    command: 'node ./streaming/index.js'
    healthcheck:
      test:
        - CMD-SHELL
        - "curl -s --noproxy localhost localhost:4000/api/v1/streaming/health | grep -q 'OK' || exit 1"
    ports:
      - '4000:4000'
  sidekiq:
    image: 'ghcr.io/mastodon/mastodon:v4.3.8'
    restart: always
    env_file: .env
    command: 'bundle exec sidekiq'
    volumes:
      - './public/system:/mastodon/public/system'
    healthcheck:
      test:
        - CMD-SHELL
        - "ps aux | grep '[s]idekiq 7' || false"

Notice the env_file: .env lines above.

The environment variables you edit in Coolify’s GUI are stored in a .env file right alongside Coolify’s version of the project’s docker-compose.yaml file. Since the .env file is sitting right there, you can add that one env_file line to each service so it has access to the complete set of environment variables.

And there are a lot of them:

LOCAL_DOMAIN=t00t.cloud
REDIS_HOST=•••
REDIS_PORT=6379
REDIS_USER=default
REDIS_PASSWORD=•••
DB_HOST=•••
DB_USER=postgres
DB_NAME=postgres
DB_PASS=•••
DB_PORT=5432
SECRET_KEY_BASE=•••
OTP_SECRET=•••
ACTIVE_RECORD_ENCRYPTION_DETERMINISTIC_KEY=•••
ACTIVE_RECORD_ENCRYPTION_KEY_DERIVATION_SALT=•••
ACTIVE_RECORD_ENCRYPTION_PRIMARY_KEY=•••
VAPID_PRIVATE_KEY=•••
VAPID_PUBLIC_KEY=•••
SMTP_SERVER=•••
SMTP_PORT=587
SMTP_LOGIN=unsend
SMTP_PASSWORD=•••
SMTP_FROM_ADDRESS=mastodon@t00t.cloud
S3_ENABLED=true
S3_REGION=•••
S3_ENDPOINT=•••
S3_HOSTNAME=•••
S3_BUCKET=•••
AWS_ACCESS_KEY_ID=•••
AWS_SECRET_ACCESS_KEY=•••
EXTRA_MEDIA_HOSTS=•••
IP_RETENTION_PERIOD=31556952
SESSION_RETENTION_PERIOD=31556952
FETCH_REPLIES_ENABLED=false
FETCH_REPLIES_COOLDOWN_MINUTES=15
FETCH_REPLIES_INITIAL_WAIT_MINUTES=5
FETCH_REPLIES_MAX_GLOBAL=1000
FETCH_REPLIES_MAX_SINGLE=500
FETCH_REPLIES_MAX_PAGES=500
STREAMING_API_BASE_URL=wss://streaming.t00t.cloud

(We’ll come back to populating the secrets in a moment.)

I grabbed the Redis and PostgreSQL connection details from the standalone services I’d already started, and as usual I was careful to make sure the web app’s Connect To Predefined Network setting was checked.

I used self-hosted Unsend for email via SMTP and MinIO for S3 storage.

The working app stack looked like this:

Zooming out one level to resources:

It wasn’t obvious how I could test Mastodon’s SMTP settings outside the initial setup wizard, but a Reddit post nudged me in the right direction.

First, dive into a Ruby console:

RAILS_ENV=production bin/rails c

Then create a mailer instance and send a test message:

> mailer = ActionMailer::Base.new
> mailer.mail(to: 'me@example.com', subject: 'Test', body: "You’ve got mail!").deliver

It worked on my first try, which made me feel pretty smart.

Secrets and PostgreSQL Data

Next I needed to bring in stuff from the current instance.

Masto.host makes it easy to download backups, and they come in two flavors:

1. Snapshots you can download at any time that’ll be slightly stale and include a minimal set of media files.
2. An up-to-date database dump and entire set of media (including caches) you have to stop your server in order to generate.

Each is an archive with contents following the same structure:

media/
pg_dump.custom
README.txt
secrets.txt

I started out using a snapshot to test the waters, and the environment variables I needed were in secrets.txt.

I uploaded the backup archive’s database dump to the Coolify VPS at /tmp/pg_dump.custom. (Coolify’s GUI includes an option to point to that file and import it.)

My first attempt failed because I needed to create the t00tcloud role that Masto.host relied on. (Yours will be specific to your server name.)

CREATE ROLE t00tcloud;
GRANT ALL ON SCHEMA public TO t00tcloud;

Provide the --clean option for pg_restore and everything should work.

At this point I was able to edit /etc/hosts and send my t00t.cloud domain to this new server and interact with the running web instance in a browser. Because I used one of the static snapshots, it had missing images and activity was slightly out of date.

Migrating For Real

I returned to Masto.host, stopped the server, and initiated a real-time backup.

It took around 10 minutes to generate. I downloaded its ~18GB archive, uploaded and imported the database dump, and used Transmit to sync up everything but the huge caches/ directory. This last part was a mistake, and you’d be wise to find some way of neatly merging that huge media/ directory with the one in your S3 volume.

It turns out Mastodon keeps references to what it thinks is in the cache directory, so you can’t just delete that stuff. If it thinks a file is there, it’s not going to double check—it’ll just serve an image URL and let it be broken in the web UI. (I’m sorry I let you down like that, Mastodon.)

I spent a while using tootctl commands to clean and repair missing assets.

These commands can take a long time to execute, and because Coolify’s handy in-browser terminal flakes out after a while I used SSH via a regular terminal instead. (From the server, run docker ps, look for the ID of the running web container, and run docker exec -it [id] /bin/bash to get in there.)

• tootctl accounts refresh can fix profile images and banners.
• tootctl media refresh can fix file references in posts/toots.
• tootctl emoji purge --remote-only can clear out the jillions of custom emojis accumulated from other servers. (This ran for hours and there was no output; check your “Custom emojis” page at /admin/custom_emojis?remote=1 to confirm it’s busy.)

Most of these commands have a --concurrency setting you can increase to significantly speed things up and/or crash your server. I did both.

Cloudflare Worker Reverse Proxy

This was a fun little side quest that’s not relevant if you use local images, aren’t using Cloudflare, or you’ve already got a CDN figured out.

I store the Mastodon server’s media in a MinIO bucket (similar to S3), and at first I was serving images directly from its URLs. But I wanted Cloudflare to be able to cache them without having it proxy the entire domain.

I don’t use Cloudflare Workers (outside of Cloudflare Pages) much, but all I needed was this workers.js file to forward requests and cache the result:

addEventListener("fetch", event => {
  event.respondWith(handleRequest(event.request));
});

async function handleRequest(request) {
  let url = new URL(request.url);

  url.hostname = "storage.m7n.foo"; // backend domain
  url.protocol = "https"; // ensure secure connection

  // create a new request preserving method and headers
  let newRequest = new Request(url, request);

  // fetch and return the response from back end
  return fetch(newRequest);
}

I attached a custom subdomain to it, which you’ll see now in URLs like this.

Conclusion

It worked. The instance behaves pretty well. Like all these other Coolify setups, establishing database backups is wonderfully simple.

If I have a real banger of a toot at some point this little instance will probably melt and struggle to stay online, so if I get a chance I’ll probably move it to one with more memory.

I ran my own instance prior to the Masto.host setup, and moving to Coolify was significantly easier now that I’m comfortable with it. I love that I’m not all that familiar with Ruby but hosting the project is comfortable with a Docker Compose setup.

I already got to update the Mastodon version, which was as simple as editing docker-compose.yaml to update the image versions and redeploying.

Hopefully something here is helpful! I’d love to know if you try this or if you noticed anything I should’ve been smarter about here.

July 20, 2025 Update

It took me a few weeks to realize I left something out that I should have remembered from my first Mastodon self-hosting experience: cleanup tasks are important to keep storage in check.

I started getting 500 errors posting images, which turned out to be my storage bucket hitting the 50GB limit I whimsically established for it. I raised that limit, added some scheduled cleanup tasks, and now my ~60GB of storage is down to 8GB. I expect it will cruise in the 8–12GB range now.

I followed pruning advice from Ricard Torres and used Coolify’s UI to add individual scheduled tasks:

Each runs in the web container, X minutes into each third hour of a given day:

0 */3 * * * tootctl accounts prune
5 */3 * * * tootctl statuses remove --days 4
10 */3 * * * tootctl media remove --days 4
15 */3 * * * tootctl media remove --remove-headers --include-follows --days 0
20 */3 * * * tootctl preview_cards remove --days 4
25 */3 * * * tootctl media remove-orphans

A few failed running for the first time—mostly processes being killed—because there was so much to churn through, but now they run smoothly.

I’ll adjust this if it’s too aggressive about pruning, but for now the cache reduction is wild and I don’t notice any significant change in front-end experience.

Now that it’s nice outside, I’ve recommitted myself to a walking goal for this year.

A friend drew my attention to an inconspicuous sign at the base of a nearby butte. Beneath it is a clear bin with little cards, and you can take one and record your walks up and around the butte and then mail that in when you’ve reached 100 miles to join the Century Club. Which means you get to have your name printed in the relevant mileage tier on the sign.

Of course there are people who walk and run this butte with such dedication that they’re in staggeringly-high tiers of their own. I don’t have that sort of aspiration, but I do want to get my name on the board.

I have a hard time getting excited about exercise in a room, but I love walking this thing and getting fresh air and seeing people and their dogs. Deer, quail, eagles red-tailed hawks, chipmunks, rock chucks, gold-mantled ground squirrels, many more birds I can’t identify easily yet, and sometimes low-flying planes circling the butte to get a look at it.

So anyway this is a thing I’m doing and I made a page to track my progress. Walk with me if you want, literally or with your imagination via internet.

Sit down, friend. I have something to tell you.

There are times in our lives when we must make hard choices.

Surely we can delude ourselves into fashioning a veil for greed, lust, or vanity. We can engage in crude acts of self-deception. We can varnish a turd.

And I have to admit to you, my dear, that perhaps I have.

Perhaps it’s fear that has driven me to this, because certainly it is not joy.

The desire to connect is real though. Surely that’s something.

You already know what I’m going to tell you.

You’re imagining the thought pieces and upsells and mining actual human relationships for overt commercial gain.

I don’t like it either.

But I cannot despair. I must rage against the dying of the light.

Now please raise your head, dry your tears, and join my professional network.

By Rainbow Rowell

Rating: 4 / 5.

I exchanged book recommendations with a friend, and she chose this one because it’s a story she’s read and enjoyed many times.

It’s just over five hundred pages of what are obviously teen wizards complete with wands and magic. Also vampires. But these wizards are called “magicians,” which is tricky for me because I can’t encounter the word without thinking of Gob Bluth.

These kids have a lot to figure out about the plot, about themselves, and about the grownups they trust and look up to.

It’s told from the first-person perspective of different characters, some of whom aren’t integrated into the story until much later. I loved this, along with how fun it could be switching between characters’ thoughts in the middle of big, intense action scenes.

The characters grew on me, and Rowell added a number of twists that were refreshing and satisfying and made the story uniquely its own thing.

I enjoyed reading this and the moving ending it worked up to.

Unsend is an open source transactional email service that sits on top of Amazon SES.

If you’re a non-developer here for my groundbreaking thought pieces, Unsend is a thing software can use to send email.

It’s a young alternative to Resend that makes it convenient to manage sending domains and API keys, track deliverability, and even conduct sending campaigns that I don’t normally think of from a strictly transactional service.

Trying Unsend was straightforward with Coolify, but it took me a while to realize that if you want an SMTP service you’ll have to run the included relay for receiving SMTP requests on their respective ports and forwarding them to the Unsend API.

The proxy runs instances of smtp-server for ports 465, 2465, 25, 587, and 2587—which means it’s ready to handle SSL and TLS connections.

Unsend’s Docker Compose example offered a clear starting point, and Aldert Vaandering’s Setting up Stalward on Coolify post helped me figure out what to do with the certificates unsend/smtp-proxy needs to use for SSL and TLS.

I originally started messing with Traefik labels at the server level, but it turned out I didn’t need to. I followed Aldert’s lead making sure that the server’s certificates could be available for the relay to use by appending the following to the Traefik configuration in Server → Proxy → Configuration.

traefik-certs-dumper:
  image: ghcr.io/kereis/traefik-certs-dumper:latest
  container_name: traefik-certs-dumper
  restart: unless-stopped
  depends_on:
    - traefik
  volumes:
    - /etc/localtime:/etc/localtime:ro
    - /data/coolify/proxy:/traefik:ro
    - /data/coolify/certs:/output

The relay itself just needs to bind some ports and point to those Coolify certificates in order to make SSL connections:

smtp-server:
  container_name: unsend-smtp-server
  image: "unsend/smtp-proxy:latest"
  volumes:
    - /data/coolify/certs/my-unsend-url.example/key.pem:/data/certs/key.pem:ro
    - /data/coolify/certs/my-unsend-url.example/cert.pem:/data/certs/cert.pem:ro
  environment:
    SMTP_AUTH_USERNAME: unsend
    UNSEND_BASE_URL: "https://my-unsend-url.example/"
    UNSEND_API_KEY_PATH: "/data/certs/key.pem"
    UNSEND_API_CERT_PATH: "/data/certs/cert.pem"
  ports:
    - "25:25"
    - "587:587"
    - "2587:2587"
    - "465:465"
    - "2465:2465"
  restart: unless-stopped

My entire Unsend stack now looks like this:

services:
  postgres:
    image: "postgres:16"
    environment:
      - "POSTGRES_USER=${SERVICE_USER_POSTGRES}"
      - "POSTGRES_PASSWORD=${SERVICE_PASSWORD_POSTGRES}"
      - "POSTGRES_DB=${SERVICE_DB_POSTGRES:-unsend}"
    healthcheck:
      test:
        - CMD-SHELL
        - "pg_isready -U $${POSTGRES_USER} -d $${POSTGRES_DB}"
      interval: 5s
      timeout: 20s
      retries: 10
    volumes:
      - "unsend-postgres-data:/var/lib/postgresql/data"
  redis:
    image: "redis:7"
    volumes:
      - "unsend-redis-data:/data"
    command:
      - redis-server
      - "--maxmemory-policy"
      - noeviction
    healthcheck:
      test:
        - CMD
        - redis-cli
        - PING
      interval: 5s
      timeout: 10s
      retries: 20
  unsend:
    image: "unsend/unsend:latest"
    expose:
      - 3000
    environment:
      - SERVICE_FQDN_UNSEND_3000
      - "DATABASE_URL=postgresql://${SERVICE_USER_POSTGRES}:${SERVICE_PASSWORD_POSTGRES}@postgres:5432/${SERVICE_DB_POSTGRES:-unsend}"
      - "NEXTAUTH_URL=${SERVICE_FQDN_UNSEND}"
      - "NEXTAUTH_SECRET=${SERVICE_BASE64_64_NEXTAUTHSECRET}"
      - "AWS_ACCESS_KEY=${AWS_ACCESS_KEY:?}"
      - "AWS_SECRET_KEY=${AWS_SECRET_KEY:?}"
      - "AWS_DEFAULT_REGION=${AWS_DEFAULT_REGION:?}"
      - "GITHUB_ID=${GITHUB_ID}"
      - "GITHUB_SECRET=${GITHUB_SECRET}"
      - "REDIS_URL=redis://redis:6379"
      - "NEXT_PUBLIC_IS_CLOUD=${NEXT_PUBLIC_IS_CLOUD:-false}"
      - "NEXT_PUBLIC_SMTP_HOST=${NEXT_PUBLIC_SMTP_HOST}"
      - "SMTP_HOST=${SMTP_HOST}"
      - "API_RATE_LIMIT=${API_RATE_LIMIT:-1}"
      - HOSTNAME=0.0.0.0
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_healthy
    healthcheck:
      test:
        - CMD-SHELL
        - "wget -qO- http://unsend:3000 || exit 1"
      interval: 5s
      retries: 10
      timeout: 2s
  smtp-server:
    container_name: unsend-smtp-server
    image: "unsend/smtp-proxy:latest"
    volumes:
      - /data/coolify/certs/my-unsend-url.example/key.pem:/data/certs/key.pem:ro
      - /data/coolify/certs/my-unsend-url.example/cert.pem:/data/certs/cert.pem:ro
    environment:
      SMTP_AUTH_USERNAME: unsend
      UNSEND_BASE_URL: "https://my-unsend-url.example/"
      UNSEND_API_KEY_PATH: "/data/certs/key.pem"
      UNSEND_API_CERT_PATH: "/data/certs/cert.pem"
    ports:
      - "25:25"
      - "587:587"
      - "2587:2587"
      - "465:465"
      - "2465:2465"
    restart: unless-stopped

I can now configure various apps to use my Unsend URL’s ports 465 or 587, username unsend and an API key as the password, and emails happen.

Troubleshooting with the relay’s logs can be a little frustrating since issue details get flattened into [Object], but Unsend is in beta and once a logging PR gets merged that should make investigating a breeze.

Sending email to Unsend API at: https://my-unsend-url.example/api/v1/emails
Unsend API error response: { success: false, error: { issues: [ [Object] ], name: 'ZodError' } }

I love that Unsend configures SES with each new domain, so I just have to set up and verify DNS records like I would with any other service. And then I have immediate deliverability reports with message previews so I can be sure everything’s working.

Overall it’s been a nice little learning adventure and I like consolidating a bunch of transactional mail setups into one that so far works well and inexpensively with no plan limits.

Update: thanks to Zai in Unsend’s Discord server for pointing out that I should mount only the certificate files I need to keep things cleaner. The smtp-server section’s volumes and environment variables are updated to reflect that.

We are experiencing higher than normal call volumes and a representative will be with you shortly.

Your call is important to us. Please continue to hold.

You can drive your car or do your laundry or make your lunch with one ear on our hold music and automated assurances.

This system was made by people for connecting people, but it feels like gears of apathy.

You are waiting to speak with someone, and someone is working their way to you.

This is imperfect, but a connection will be made if you are there.

Please continue to hold.

You’re right, Alexandra! I had fun reading your post and I have spontaneously decided to share seven random facts of my own.

Fact #1: I thought for a long time that “Netflix and chill” meant hang out and watch movies.

I’m pretty good at missing random things everybody else is onto, which thankfully offers endless comedy. I think I was heading out one night when I told my partner I didn’t know what my plans were, maybe just “Netflix and chill,” and she stopped and put a hand on me and had me tell her what I thought that meant.

It did not mean what I thought it meant.

If you don’t know, I guess ask your parents or something.

Fact #2: I was a chopper gunner and team coordinator for a Battlefield 3 clan.

I occasionally play video games, and not often the cute contemplative ones but the big run-and-shoot-stuff adrenaline romps. Earlier in my life I played Battlefield a lot, a giant game with a lot of people in teams trying to capture and hold flags with tanks and helicopters and war machines. For reasons that remain unclear, I was disgustingly good at being the gunner in an attack helicopter. I accepted an invitation into a platoon, which is basically a self-organized team that divides up roles and practices and plays other teams. I’d gun for a really, really good pilot while coordinating communication between squads. We were great together and we had a lot of fun! I miss being that good at something, however useless, and working so quickly and smoothly as a team.

Fact #3: I frequently remove tags and labels from stuff.

I’m sensitive enough to logos and decals that I either love them or get rid of them. No license plate covers on the car. No ugly stickers if I can manage to get them off. No stitched-on tags if I can use a seam ripper or my beloved iris scissors to liberate them. (A lot of Carhartt logos are easy to remove!) No white printed text if a black fabric marker can vanish it. If you see me sporting some kind of logo, you can be sure I’m either a big fan or I couldn’t manage to get rid of it.

I managed to remove the big gross airbag warning decals from the sun visors in my last car, but that was a lot of effort and there was lingering visible evidence of the struggle. I have to choose my battles even though I am frequently at war.

Fact #4: I’ve lived in six U.S. states, only two sharing a border.

Ohio, New York, Washington, Florida, Texas, and Oregon—and not in an order that makes any sense for efficient travel.

I’ve not been in the military or witness protection, nor am I fleeing anything in particular. We’ve just moved a lot to see what different places (in the United States) are like. I’m tired of long-distance moves at this point, but I’m glad for the adventure.

Fact #5: My pre-resumé work history includes Lowe’s, Abercrombie & Fitch, a newspaper route, hand-drawing diaper and bottle bag specs for manufacture, hand-painting kitschy sayings on kitchen soffits, and assembling lockers and pallet rack in factories and warehouses.

I was lucky enough to work a bunch of jobs early in my life, which did not feel lucky at the time but in hindsight gave me an appreciation for how lots of different people live and work. I really liked being able to use my hands to make money, especially when it meant drawing or painting or getting a critical eye and artistic inclination involved.

I also have an innate respect for a day’s work that you can see the result of when you’re done. I rarely feel like this with something that lives on a screen, even when I’m happy with it.

Fact #6: I’m annoyed breakfast tacos aren’t popular outside Texas.

Breakfast tacos are the perfect Tex-Mex form factor for that time of day. They were all over Austin at various levels of fanciness, and I don’t know why seemingly every other place on the planet manages to produce breakfast burritos but not breakfast tacos.

Fact #7: I watched the last space shuttle leave Earth.

I just stood there agape near Cape Canaveral, but I’m so glad I got to see that bright starburst in its thunderous streak away from the planet. Some nighttime satellite launches confirmed that it’s mesmerizing to watch anything leave the planet.

A few people have asked whether I’ve managed to get Craft CMS running on Coolify, so I figure it might be helpful to share what I’ve got.

The project I’m running is only a small test site, but it’s been stable and happy. There are lots of ways to host a Craft site, and this is what I’d consider a bare minimum setup:

• PHP 8.4, along with the extensions Craft needs (plus ImageMagick and GD).
• A stable, independent queue runner so the AJAX-based runQueueAutomatically option can be disabled like nature intended.
• Standalone MySQL and Redis containers. (Where you could just as easily use PostgreSQL.)
• Redis for cache and sessions.

I mentioned in an earlier post that I liked using base Docker images from Server Side Up. I’ve continued using them for all my Coolify-hosted PHP projects, and it’s taken a minimal amount of tailoring to be up and running.

I’m not a Docker expert, but I like them for a few reasons:

• They’re maintained by people that know PHP and Docker.
• They’re careful with permissions and defaults because they’re designed for production use.
• They’re customizable via well-documented environment variables and other patterns that are just plain handy, particularly if you use Laravel.

The Pieces

My Dockerfile uses their PHP 8.4 image, customizes some PHP settings and the document root, installs mysqldump, PHP extensions, and Node.js, then updates npm and Composer dependencies:

FROM serversideup/php:8.4-fpm-nginx

ENV PHP_OPCACHE_ENABLE=1
ENV PHP_MEMORY_LIMIT=1024M
ENV NGINX_WEBROOT=/var/www/html/web
ENV CRAFT_WEB_ROOT=/var/www/html/web

USER root

# Install mysqldump so control panel backups work
RUN apt-get update
RUN apt-get install -y lsb-release wget gnupg
RUN curl -sLO https://dev.mysql.com/get/mysql-apt-config_0.8.33-1_all.deb
RUN dpkg -i mysql-apt-config_0.8.33-1_all.deb
RUN apt-get update
RUN apt-get install -y default-mysql-client

# Install PHP extensions
RUN install-php-extensions bcmath gd imagick intl

# Install Node.js v20
RUN curl -sL https://deb.nodesource.com/setup_20.x | bash -
RUN apt-get install -y nodejs

# Tidy up
RUN apt-get clean all

COPY --chown=www-data:www-data . /var/www/html

USER www-data

RUN npm install
RUN npm run build

RUN composer install --no-interaction --optimize-autoloader --no-dev

That Dockerfile is used for the app and queue services I define in the Docker Compose configuration I give to Coolify:

services:
  app:
    build:
      context: .
      dockerfile: Dockerfile
    volumes:
      - ./storage:/var/www/html/storage
    healthcheck:
      test: curl --fail http://localhost:8080/healthcheck || exit 1
      interval: 10s
      retries: 5
      start_period: 10s
      timeout: 10s
  queue:
    build:
      context: .
      dockerfile: Dockerfile
    command: ["/usr/bin/nice", "-n 10", "php", "craft", "queue/listen", "--verbose"]
    volumes:
      - ./storage:/var/www/html/storage
    healthcheck:
      test: curl --fail http://localhost:8080/healthcheck || exit 1
      interval: 10s
      retries: 5
      start_period: 10s
      timeout: 10s

This means that the Coolify app has an always-running container for handling web requests, and another one that’s always running the queue. (That command: line leaves it up to Docker to make sure php craft queue/listen is always running; we don’t need to get cron or supervisor involved!)

The sharp reader may be alarmed at this point to not see any mention of MySQL or Redis.

As I pointed out in my much longer Coolify post, I use Coolify’s GUI to establish persistent services—in this case MySQL and Redis. That way they’re always running independently of deployments, and backups are ridiculously easy to manage via Coolify. The working app layout looks like this:

Once those database containers are running, my Docker Compose services can nearly come to life and interact with them.

Three last things are important:

1. The Docker Compose services need to be able to reach the standalone databases—which means visiting the Advanced menu end enabling Connect To Predefined Network.
2. Craft needs to be able to apply project config changes and migrations. I used the General menu’s Post-deployment setting to run php craft up --interactive=0 on the app container.
3. Environment variables! You’ll need to set those in order to configure Craft for the Coolify environment.

My environment variable set looks like this:

CRAFT_ALLOW_ADMIN_CHANGES=false
CRAFT_APP_ID=•••
CRAFT_BACKUP_ON_UPDATE=false
CRAFT_DB_DATABASE=default
CRAFT_DB_DRIVER=mysql
CRAFT_DB_PASSWORD=•••
CRAFT_DB_SERVER=•••
CRAFT_DB_USER=mysql
CRAFT_DEV_MODE=false
CRAFT_DISALLOW_ROBOTS=true
CRAFT_ENVIRONMENT=staging
CRAFT_RUN_QUEUE_AUTOMATICALLY=false
CRAFT_SECURITY_KEY=•••
PRIMARY_SITE_URL=•••
REDIS_HOSTNAME=•••
REDIS_PASSWORD=•••
REDIS_PORT=6379

I’m assuming you already use Redis, but if not you’ll want to follow Craft’s examples using it for cache and/or session storage.

Fine Print

It took me a while to figure out how to get mysqldump working, so for a while I only had Coolify’s backups.

The only oddity I haven’t solved is that re-indexing assets ends with a generic error: “There was a problem indexing assets.” This will also pop some “A server error occurred.” messages. Re-indexing does, however, complete successfully.

The Steps

Here’s how I set things up from start to finish.

Create Standalone Services

1. Commit Dockerfile and docker-compose.yaml to the root of your project.
2. In Coolify’s GUI, navigate to Projects and click + Add. Enter a Name and optional Description and click Continue.
3. Click the newly-created production environment to set it up.
4. Click + Add New Resource.
5. Choose MySQL. You can leave everything as you see it here, but grab the MySQL URL (internal) value because you’ll need to share details from that with Craft.
6. Your production environment will now include the exited MySQL database. Click + New and choose Redis. Grab the Redis URL (internal) value, too.
7. Navigate into each of those services and click Start. Each one should build and spin up and turn green, meaning you‘ve got MySQL and Redis ready for Craft!
8. Import a MySQL backup from the MySQL service’s Import Backups section, providing a .sql dump from another environment.

Create the App Services

1. Once again from the production environment, click + New. Now we want to point Coolify to the Craft CMS project repository. I used Private Repository (with GitHub App) with an already-established GitHub connection.
   • Choose the project repository from the dropdown menu and click Load Repository.
   • Choose the appropriate branch.
   • For Build Pack, choose Docker Compose. Coolify should detect / as the Base Directory and confirm that the /docker-compose.yaml file exists there.
   • Click Continue.
2. Enter your site’s base URL in Domains for App. (Example: https://my-craft-site.example.)
3. Click Save.
4. Navigate to Advanced and check Connect To Predefined Network.
5. Navigate to Environment Variables to populate those.
   • You’ll need to get the MySQL and Redis connection details from those earlier connection strings.
     • MySQL: mysql://{username}:{password}@{hostname}:{port}/{database}
     • Redis: redis://{username}:{password}@{hostname}:{port}/{database}
6. Click Deploy, and everything should be built and deployed successfully.
7. You should be able to visit the site’s URL in a browser, but you’ll get “/var/www/html/storage isn't writable by PHP. Please fix that.” We have to set those permissions as a one-time thing.
   • From the Coolify application service, click Persistent Storage adn grab the Source Path value.
   • From Coolify, click Servers and the one your app is on.
   • Click Terminal.
   • Enter chown -R www-data:www-data {path-you-copied}.
8. After you’ve started the container for the first time, find Post-deployment, enter php craft up --interactive=0 and specify app in the Container Name field. Click Save, then Redeploy to confirm everything works as expected.

You should be able to visit your front end and control panel and have a fully-working Craft site!

Post-Setup Steps

There are a few things you might want to know if this setup is new to you.

First, database backups are wonderfully straightforward to set up from the MySQL service’s Backups section. They can be kept locally, rotated regularly, and optionally uploaded to remote storage. Be aware that Coolify keeps binary backups, which are different from what Craft dumps out of its backup utility. (Both types are simple to restore from Coolify’s GUI.)

Next, you can pop into the app service’s Logs section to see (or stream) output from the web and queue containers. This can help confirm that the queue is running and confirm that web requests are being handled properly, but Craft’s own logs are still going to be in its storage/logs/ directory unless you configured things differently.

You can run console commands by visiting the app service’s Terminal section and connecting to the app container.

If you want to get to Craft’s persistent storage on the host filesystem (not in the container), you’ll find it at /data/coolify/applications/{app-id}/storage.

The End

I’m not as comfortable or confident with this setup as I would be relying on Servd or a trusty Ploi or Forge VPS, but it’s been running smoothly and auto-deploying for a few months now and it’s been painless to keep up to date.

I’m sure a few steps could be smoothed out here, but I hope this was easy enough to follow. If you’ve got any questions, additions, or corrections please send me an email—I’d love to hear from you!

Update: I learned after publishing this article that Samuel Reichör published a Craft+Coolify tutorial that takes a slightly different approach I may need to steal ideas from!

Sometimes he lingers, letting fear or wisdom survey the fork and choose a direction.

The wanderer cannot walk every road. From his first steps to his last, only a tiny fraction of trails will know the weight of his frame and bear the impermanent marks of his shoes. He will never encounter the dangers and delights of many other paths.

The wanderer may not truly ever know where he’s headed. He may eventually accept that a fixation on the end leaves less of him for the present, where the only real steps are made.

Occasionally a divergent path is alluring. It calls and entices, bright and lovely with a strong pull into the excitement visible further into the distance. The wanderer, having been lost and imperiled, knows these can be the hardest.

He sometimes pauses to admire the beauty of a path not taken, an adventure forever unknown, before continuing with the one in front of him.

There are many paths but only one wanderer, and he must ultimately choose a single path and wander alone.

By Kathryn Schulz

Rating: 5 / 5.

I finally finished this after starting way back in October. A few other books managed to jump the line, but this was such an interesting read it was almost meditative and I never wanted to rush through it.

There’s so much cultural pressure to avoid or contain errors that it was refreshing to explore how natural and fundamental they are to being a thoughtful, conscientious person.

Art and science are explorations of error, and a person recognizing their wrongs is one that’s growing and changing.

Schulz explores this from many fascinating angles: philosophical arguments, accounts of wrongly-convicted assailants, famous historic blunders, commercial error-correction efforts, humor, and curious psychological cases to name a few.

We already saw that “seeing the world as it is not” is pretty much the definition of erring—but it is also the essence of imagination, invention, and hope.

We have a tendency to believe our own stories and be suspicious of those who disagree, and we’re entirely unaware of our wrongs until some abrupt moment when they become apparent.

I feel like a broken record writing about my discomfort with zealotry and my affinity for doubt, so passages about the nature of certainty and doubt were cathartic to read. The end even got into the fundamental importance of error in comedy, which was convenient for where my head’s at right now.

I’m overwhelmed by my mountain of highlights because there were just so many interesting quotes and points and stories to consider.

In the optimistic model of wrongness, error is not a sign that our past selves were failures and falsehoods. Instead, it is one of those forces, like sap and sunlight, that imperceptibly helps another organic entity—us human beings—to grow up.

Highly recommended for anyone actively being a person.

I distinctly remember how it ended with the Millennium Falcon landing on Yavin 4, abruptly transitioning to the summer Olympics high dive event my parents taped over the end of the movie.

I watched that tape enough times that without re-watching I can see that space hamburger turn and land in the tree line. The helmeted guy with a forehead thermometer watches from his precarious booth in the foreground. A mysterious ending full of questions.

It wasn’t until Star Wars was re-released in theaters that I saw all that bonus footage with (SPOILER!) blowing up the Death Star. My high school friends were bewildered and disappointed by my excitement.

You don’t know what you don’t know until you know. You know?

By Mel Robbins

Rating: 3 / 5.

I read this with a friend and appreciated Robbins’ warmth and humility.

After a lot of self helps books and recent re-reading of The Courage to Be Disliked, this felt like a motivational amalgam of concepts I’ve explored elsewhere with more depth. I was a little irked by the forced branding of the so-called theory and tenuous relationship to science that felt more hand-wavy than substantiated. Experts would be cited not to elaborate on research but to seemingly provide credibility for the book’s main thrust. It often seemed as though a leading expert was ushered in to say “yup, floss your teeth!”

I don’t doubt that Robbins’ advice is supported by research and sound psychology, it just seemed like the book was peppered with “this is science!” and little further explanation.

To its credit though, this was an easy read and built around the kind of succinct ideas I can actually remember and therefore make use of. “Let them” is a powerful and wonderfully-easy phrase to carry around. (In addition to “let me.” As in: let them do a thing I have no control over anyway, and let me think or act in a way that’s healthy and within my realm of concern.) I imagine that memorable framing is what makes this a popular title.

If you’re frequently tied up in other peoples’ business and not making strong choices for yourself, this may be a liberating and encouraging read. If you enjoy digging into psychology and you can already cite some favorite books about boundaries and self-control, there may not be anything new for you here.

Somehow we ended up visiting a modern art gallery.

This may not be funny to you because you grew up in an art-forward, open-minded family where this would be a normal activity—in which case we have amusingly-contrasting origin stories.

I only developed an awareness of modern art in college. I grew up drawing and painting and sculpting and photographing, but artistic practice was a rarity in my family. If I encountered modern art at all, I probably dismissed it as a strange anomaly.

So the exact circumstances that left me standing in front of Bruce Nauman’s Five Marching Men with my father are unclear, but the conversation is one of my all-time favorites.

If you’ve never seen Five Marching Men, we need a brief look because it’s instrumental to the story.

We’d rounded a wall to be confronted by a giant animated neon sign that, upon further inspection, was five naked male silhouettes marching. You can tell without reading any signage that they’re marching, because of the straight backs with raised arms and legs moving in unison. The unexpected detail, which also registers as natural given the nakedness, is each penis that flops along with the marching arms and legs.

I laughed.

My dad turned in disgust: “come on, really?”

My father, a person whose natural curiosity almost certainly never reached the bizarre world of modern art, fell right into the big question: what qualifies as art?

“It’s ridiculous,” he said dismissively, about to keep walking.

“I agree, and that’s why I like it!” I responded. “Not ridiculous for being here, but in what it’s depicting. I think it’s supposed to feel ridiculous. There’s something absurd about the proud, coordinated posturing of a march and the reality of penises flopping around in unison.”

I probably used a lot more words to say something less articulate, but I remember him stopping and thinking about it. I remember the shift in orientation that made the rest of the gallery visit a more interesting discussion rather than a rushed parade of indignities. I was surprised and overjoyed that my dad pushed past revolt to entertain a new way of interpreting what he saw.

I misremember that piece’s title as Soldiers Marching, because it further amplified what I took away from it: that even as we organize into machines of warfare, we are each still people. A military display would only ever involve nakedness to humiliate the enemy. The whole point is to be a unified, unbreakable force of dominance that deliberately obscures humanity.

Individualism and softness and indecisiveness and moral dilemmas are drilled out and exploited and even ridiculed by opponents because that’s the game. The most horrific atrocities are perpetrated by human beings convinced not to see human beings.

I hope it’s not a surprise when I tell you that I don’t see two national genders and more masculine energy as good things. Those are distinctly in the wrong direction. Those are more marching.

An improv friend recently mentioned how she’s been cheerfully and willfully interpreting her favorite stories through the lens of improv commentary—not how they were intended, but a delightful and interesting way to approach them nonetheless.

I took this as an opportunity to stroll through quotes I’ve collected and see what they looked like as advice for improvisational comedy.

And because I think of improv as a life metaphor, most of them read like great improv wisdom.

That rabbit hole shot me into a long-forgotten bookmark where endless photographs of famous artists and thinkers are paired with their own quotes. Many leapt out at me for different reasons, but this one stopped time for a few seconds:

The realization that life is absurd cannot be an end, but only a beginning.

I’m realizing how precious and unlikely sudden insights can be for each of us. I can re-watch a movie or re-read a book months or years later and have a wildly different experience with it because of where I am as an ever-changing observer.

I can go on and on about my epiphany, knowing I’ve really got something, and your most honest reaction might be “oh ... ? [takes another slow, bewildered bite of sandwich]” because you live in your head and not mine.

(I might also have the same reaction to myself a week later.)

But this quote about absurdity struck me in that moment. I immediately started reading about Albert Camus, and trained myself to pronounce his name properly in my head in case we talk about it. (I was way off.) I started reading The Myth of Sisyphus, his famous work that was boiled down into that quote.

I’ve always been drawn to comedy that makes me feel all the feelings, because to me something is not deeply funny and light without a grounding in something heavy and solemn. Or perhaps a relatable strangeness.

Absurdity, I am just learning, ties everything together.

Camus argues that one must eventually realize we need to build systems and structures and rules to try and reach some sense of order that simply doesn’t exist. Each of us, upon realizing we live in meaningless chaos, has two choices: kill yourself (because what’s the point?) or live freely in a brief and vibrant act of rebellion. He only accepts the latter as a reasonable answer.

This may not resonate with you, and that’s fine.

For me, this is everything.

Finding the absurdity in a situation or a scene is ultimately what grounds comedy.

Not a wild goofiness that has no bearing on reality, but something honest and relatable that also touches the absurd.

Holding the stress du jour against a great universal disorder lowers the temperature and makes everything a plunge into the unknown to be relished rather than feared.

My most precious mantra is “everyone dies,” because it makes me feel an appropriate smallness and appreciate simultaneously that all of this ends no matter what and my time is limited so I need to make choices rather than worry endlessly. It makes me still for a moment and gets me moving.

Everything is transformed when I embrace absurdity instead of fearing it.

It’s not meaninglessness or nihilism. It’s proceeding from the foundation that safety and certainty are illusions—and however much I need them to operate, it’s okay that they’re just tools. When I don’t confuse them for reality, I’m free of that worry that comes from seeing that some system of order smacks into disorder or nonsense. In fact, I can flip the whole equation and play at the exact point of interchange where chaos and order are bound by a thin, fraying piece of tape.

Each of us has a limited amount of time to explore this life and make what we can out of it.

We can choose to explore with each other and share what we have and want and know and aspire to, and we should.

I don’t know what to do with the things we must fight over, and I’m not the kind of animal that’s going to be good in a fight.

I want to remind you that in every bold march of ego, there is a flopping penis. (It’s a metaphor; I don’t care about what flops about your earthly vessel.) When we manage to forget or purposefully obscure this, something urgently necessary and humanizing is lost and that’s when shit gets bad.

We can make art, and we can choose good.

We can each stop at the strange, abhorrent thing and find ourselves in it and be open to what comes next.

By Rainbow Rowell

Rating: 2.5 / 5.

I read this because it was a friend’s reference for casual magic as a plot device.

That much was cool. A hardwired, old yellow phone in the protagonist’s childhood bedroom magically and inexplicably reaches her husband’s childhood home in the past. This was at the very beginning of their marriage which, in the present day, seems to be in a drifting and uncertain place.

I struggled because I never liked the main character, and the other essential plot device was two adults who could not manage to use their cell phones like grownups. One chose to stay home for work while the rest of the family traveled for Christmas, yet failed to get said work done and manufactured drama while failing to accept this choice.

I didn’t get much depth from the characters, either. But the plot kept moving and I appreciated some of the discussion and framing about relationships and I liked the literal movement of the end that sort of counterbalanced the whole story.

This was a fun emoji list to compile last time, so here we are again!

📨 Mail Client: Mail.app, MailMate + SpamSieve + GPG Suite
📮 Mail Server: MXroute
📝 Notes: Obsidian + Tot
✅ To-Do: Obsidian
📷 Photo Shooting: iPhone, Lumix GH5
🎨 Photo Editing: Photomator, Affinity Photo, Pixelmator Pro
📆 Calendar: Apple Calendar
📁 Cloud File Storage: iCloud + Syncthing
📖 RSS: Reeder
🙍🏻‍♂️ Contacts: Apple Contacts
🌐 Browser: Safari
💬 Chat: Apple Messages, Discord, Slack
🔖 Bookmarks: Raindrop
📑 Read It Later: Obsidian
📜 Word Processing: Apple Pages
📈 Spreadsheets: Apple Numbers
📊 Presentations: Apple Keynote
🛒 Shopping Lists: Apple Reminders
🍴 Meal Planning: Mela
💰 Budgeting and Personal Finance: Soulver
📰 News: The Atlantic + Mastodon + a jillion blogs
🎵 Music: Apple Music
🎤 Podcasts: Apple Podcasts
🔐 Password Management: 1Password
🧑‍💻 Code Editor: VS Code, PhpStorm, Sublime Text
✈️ VPN: NordVPN

Bonus Items

Important enough to include here, ordered roughly by how often I rely on each.

🚀 Launcher: Alfred
🐚 Terminal: iTerm 2
☂️ Backup: Arq + self-hosted MinIO on a storage VPS (replacing Backblaze)
🚫 Ad Blocking: Vinegar, Wipr 2, StopTheMadness
🔎 Search Engine: Kagi
📓 Journaling: Day One
🗂️ Version Control: Tower
🖼️ Screenshots: CleanShot X
🐘 Mastodon Client: Ivory
👨‍💻 Local Development: DDEV
🗄️ Code Repositories: GitHub
🛌 Sleep Tracking: AutoSleep
💽 Database Manager: TablePlus
📖 Reading: Kobo, Apple Books
✍️ Writing: Obsidian, VS Code, Sublime Text, Scrivener, iA Writer
🧾 Invoicing and Time Tracking: Harvest
👨‍🎨 Design: Figma
🕹️ Games: BattleBit Remastered, Hell Let Loose, Star Citizen
📊 Web Analytics: Plausible
🗓️ Schedule Booking: Cal.com
🤖 Server Provisioning: Coolify + Ansible
🩺 App + Server Monitoring: HetrixTools + Sentry + GlitchTip
📦 Package Tracking: Parcel
🧠 Brainstorming: Mindspace (RIP iThoughtsX 😢), Obsidian, reMarkable
☎️ Video Calls: Cal.com, Zoom
📋 Snippet Sharing: HedgeDoc
🎛️ 3D Modeling and Slicing: Autodesk Fusion 360, PrusaSlicer
🗺️ Maps + Driving Directions: Apple Maps + CarPlay
🔥 Wildfire + Air Quality Tracking: Watch Duty
🎬 Filmography Reference: Callsheet
📽️ Video Editing: DaVinci Resolve
🐎 Motion Graphics: Apple Motion
🎹 DAW: GarageBand
👀 Editor Analytics: WakaTime

Bonus Commentary

The few things that have changed reflect my desire to slowly get away from VC-funded apps and platforms, instead choosing smaller ones even if it leaves me with a bit more work to do.

The one with the biggest daily impact was moving from Raycast back to Alfred, which gave me a chance to audit my workflows and update my custom ones. That was fun.

When we cancelled our Amazon Prime subscription I replaced my Kindle with a Kobo. (With the Bookerly font for reading; I can’t give that up.)

More for the adventure than anything, I moved my perfectly-good mailbox.org email to a lifetime MXroute plan. imapsync made this simple and reassuring.

I wrote about wandering into Coolify even though Ploi is still great and I’d still cheerfully recommend it. Docker, Traefik and I are still uncomfortable friends but it’s been good for learning.

Speaking of infrastructure, I added a backup category just so I could point out moving from Backblaze to Arq and MinIO. Arq lets you back up your Mac to an impressive variety of sources, and MinIO lets you self-host S3-compatible volumes. So I used Coolify to get MinIO running on an inexpensive storage VPS on another continent with plenty of room, and now I have incremental, encrypted backups with a much less frightening restore process than Backblaze offered. Cheaper than Backblaze, too. So far a win all around.

I was particularly thrilled to try Photomator again and discover it’s exactly what I’d like for managing my own photo library. I bought a license days just before the team announced being acquired by Apple. I’m assuming that’s bad news for Photomator continuing to support file management outside the Photos library and I’ll have to find some other app again. Or give in and trust Photos and give up on my directory structure. That’s bitten me in the past, but maybe I should channel my inner Tobias Fünke and hope mindlessly that it’ll be fine.

Tot has been the biggest surprise. I was sure I had no need for such a simple notepad, but after trying it just for fun it turns out I do? The Iconfactory knew before I did: a simple place to dump ephemeral blurbs of text in visually-distinct little buckets, that also happens to sync automatically across machines—I apparently need that to improve my mental RAM. An empty Sublime Text file I never save is fraught with danger. Tot is a lovely little friend to have.

I’m still without a suitable Notion replacement for collaborative docs and planning, particularly with friends outside of work (where subscription fees are hard to justify), but maybe in 2025 I’ll find it.

I realize I’m heavily into the Apple ecosystem and this is all relative, but it’s been nice getting smaller again and I hope I can continue the trend however slowly.

By Ichiro Kishimi and Fumitake Koga

Rating: 4.5 / 5.

I’m a proud and a embarrassed to admit that I just re-read this book.

The irony is telling and darkly amusing: I first read The Courage to Be Disliked in February 2022 but couldn’t bring myself to publish a note for it. I lacked the courage to let you know I read a book about finding courage.

Someone I admire recently shared how life-changing this book was for them, which inspired me to read it again.

I happened to change e-readers between readings and it was interesting to see how my highlights differed. There were a lot of similarities, but with this second reading my highlights faintly moved from the trouble of how things are to what I could think or do. In other words, less “that nicely captures my conundrum” and more “this is worth more thought or action.” Seems positive!

The format and translation was just as distracting this time, but I was startled to realize how much I really absorbed two years ago. Some of the concepts of Adlerian pyschology apparently stuck with me and resurfaced in other things I read, watched, and even wrote.

That all problems are interpersonal problems, and we’re more alike than we’re not.

That freedom is taking responsibility for what we can change and letting go of what we can’t.

That happiness is ultimately about self-love and a sense of communal belonging that comes from active commitment.

That now is the only reality we exist in, and our ideas about the past and future often let us squirm away from being present.

Adlerian psychology is a psychology of courage. Your unhappiness cannot be blamed on your past or your environment. And it isn’t that you lack competence. You just lack courage. One might say you are lacking in the courage to be happy.

There’s a lot more to it than courage; it’s an appeal to taking complete responsibility for your life, your decisions, and not meddling with those of others. (Which, despite how that sounds, is not easy or solitary work.)

We humans are not so fragile as to simply be at the mercy of etiological (cause-and-effect) traumas. From the standpoint of teleology, we choose our lives and our lifestyles ourselves. We have the power to do that.

The format of a youth questioning a philosopher was helpful for me because I learn best pushing against new ideas until I can understand them. The youth even describes my posture toward arguing:

I realized a little while ago that maybe I don’t just want to take apart your argument—I want you to take apart mine, too.

Choosing highlights to share was hard because I have so many of them. But I could probably stand to tattoo this to my face so I never stop reading it in the mirror:

One must not get too serious. Please do not confuse being earnest with being too serious.

Highly recommended.

Too much gray, too much damp, too much gloom.

Maybe there’s nothing we can do to change that, but I’m going to take a moment and try.

Because I love the mist and the fog and the wet streets and clouds of steam and the soft blanket of white noise that falls from the sky. I love the puddles and reflections and rivulets and the way headlights cut through neatly-defined slivers of wet air. I love the moss and the ferns and the freshness of cool damp that you can breathe in and feel drifting growing swirling around you. Plants and trees filling themselves with it like we pull breath emerging from water. I love bundling against soft rain, the steaming hot cup, and huddling inside that’s more cozy and warm because outside is not.

This pairs well with rocky coast where giant trees give way to towering cliffs and jagged thunderous surf that continues to great vibrant salty oceans too far to see and too deep to survive.

Rain is sea exploring land.

The sun brings people out to play and make noise, and the rainy days welcome quiet and reflection. They nudge joys to gather inside, into houses and bars and noodle joints and libraries and places I love to be. There’s a softness and warmth and romance that no dry, sunny day can ever match.

Those that inhabit the damp for months stream outside once the warm sun emerges into a clear blue sky. Maybe they’re enduring misery until summer, but I merely exchange one set of joys for another.

Do not be afraid.

Your home can be small and cramped and quiet, or vast and lively.

Your home changes constantly even though it’s where you start from and return to.

Home is not a place. It’s the people with you.

They may occupy the same dwelling, but it is not the dwelling that makes it home no matter how much care you’ve put into the furnishings and decor.

It’s the people gathered together again, for whatever reason there is to gather, in a place that’s safe and good.

It can be many people in different places, or a few people in that one place.

But it’s the collected souls—not the joists or the beds or the mortgage—that make it home.

It is the same entity that was me twenty years ago, and the same entity that is me today.

But just like teenaged me is very different from today’s me, my future self has changed.

Future self wears age well.

Not just an exterior that looks more at home with smile lines and a measured gait, but a goofiness and creativity that are balanced against a long time watching, listening, making mistakes, and finding connections.

Future self is able to be louder, and quieter, and have the presence to know which best suits the moment.

Future self is more grateful for the time that’s left, yet more calm and less rushed than any previous selves. Future self is less worried about passing time, and more intent on finding meaningful ways to spend it.

Future self is still afraid to do things, to look stupid, and to make mistakes. But he’s more likely to try something and not shy away from it. He’ll find a way to get at it sideways instead of turning around. If he decides against it, he continues to what’s next with less hesitation.

Future self cares less what others might think, and finds more energy for those in front of him.

Future self has a few good friends, and does a better job of nourishing those relationships.

Future self uses a hard-won peace to find others looking for it and help them.

Future self probably has a tattoo.

Future self has found a way to live within his means, and a way to earn and think about money that doesn’t waste energy earning and thinking about money.

Future self is the person I’ve been growing toward, wanting to know, and wanting to share with others. He is thoughtful, strange, kind, and makes you want to be not more like him, but more like you.

Future self is good at listening to and telling stories.

Future self stands up, looks up, and speaks up more with bravery and kindness.

Future self is writing the most interesting part of my story. He is waiting for me to trust myself, take chances, and look away from fear toward joy.

He is waiting for me to realize that my humor and my sensitivity and my curiosity are exactly the tools I need to overcome fear and self doubt.

He is waiting for me to see that I already know what to do, and that I need only do more of it.

Future self makes things out of clay, and shares them.

Future self takes more photographs because finding images helps him see.

Future self plays music. He practices so he can play well with other people.

Future self values what he’s able to share.

Future self makes mistakes, says things other people don’t like, and is misunderstood more. But future self does not stop being kind, or careful. Nothing keeps him from finding his people and seeing love blossom.

Future self ends up dead. But before that, he is alive.

This always stresses me out, because I’ll learn new names I need to remember despite brain-melting exercises and scenes.

It didn’t help that more and more people kept filtering in. Another new face! Familiar ones I hadn’t seen in a while!

The small room buzzed with a cheerful din of conversation.

It was the largest class I’ve been in. Lots of names to get wrong. Lots of people to embarrass myself in front of. Enough bodies that someone opened the door to vent heat.

Familiar warmup exercises were tricky with so many people trying to stay focused while zipping around the room avoiding collisions.

I have a perpetual awareness of the many ways I can screw up, so it’s a fight to keep my attention on the very few and very specific things I need to do right.

I could see my frazzle on other faces. When one exercise got twice as hard, we started clashing and fumbling.

Then something amazing happened.

Our instructor told us to stand firmly and comfortably and breathe. From a calm, attentive place, trust that we could absolutely handle the exercise and be good at it.

And then we were.

Not just a few people, but the whole group. We moved more calmly, more fluidly, fierce and relaxed and smiling. We were good at what we were doing together.

It continued into other exercises, then into scenes.

If you had been in the room, you would have noticed the change after that simple direction.

After a break, we ended up talking together about calm and fear. Each person was tuned in.

People representing an eclectic mix of ages and professions shared how comfortable they are with conventionally-terrifying parts of their work and how afraid they are to show up and do improv. It demands complete attention for something you have no control over, where the only hope of success lies in being present and attentive and creative and keeping fear from pumping the brakes.

“Oh that’s silly, you shouldn’t be afraid because you’re very good at this,” I would repeat to myself as different people spoke. “I’ve watched you feel your way to something spontaneous and smart that made us all laugh.”

And yet from each person who actually was very funny, fear.

How curious.

Fear of forgetting a name, tanking a scene, failing to be a good partner, plunging into confusion few could follow and fewer could laugh at.

We’ve each done those things, but they should not be afraid because look at what they managed to pull off!

The realization came to me, as it so often does, very slowly.

If we had microphones sensitive enough to capture the sounds of inner thought, we might have heard the cold, reverberating “thunk” as something shifted in my mind:

I should be asking myself, too. Why are you afraid?

I have had the same fears, but I have also had my better moments.

What if I am exactly the same?

What if it’s silly that I let fear be such an obstacle that I alone bring with me when others are rooting for me to succeed?

What if calm really does come from practice against stress, and I’ve already been improving at something I don’t need to be so afraid of?

I wouldn’t have arrived at this rush of calm and affection without hearing others describe what I feel—and reacting with the quiet bewilderment of “oh but surely you should not be afraid!”

I write here with the hope of finding something that connects us even if we never write or speak to one another. It’s why I like reading what you write, too.

I also write here, on this sparing little side blog, because I’m afraid to be wrong or rambling or detailing the obvious too close to wherever I’m supposed to be professional and confident and succinct. Ready with answers instead of wondering aloud at what they might be.

Maybe that’s a pointless division of selves that robs energy from something stronger and more unique. Maybe if it’s a distinction rooted in more fear than prudence, it should go. Maybe if it’s a smart and practical division for my work, I should find another line of work.

I don’t know yet, and I can’t dismantle every wall at once.

I’ve been uncharacteristically calm since I left that class.

I’ll be on stage for the first time this weekend and I’m finally less anxious about how I could screw up because I’m excited at what fun it could be.

Please do whatever you do with love and bravery, and share it if you can.

Imagine what we could do together.

And frankly who cares.

It died clutching its heart at the growing roar of “fuck civility!”

Good riddance to this scourge of weakness that kept people from splitting into clear sides and fighting.

What did civility ever do for anyone?

Civility wasted precious time and energy on hopeless tedium: asking questions, looking for common ground, bringing and appealing to one’s better nature.

Civility made a whole thing out of sharing the same neighborhoods, schools, roads, libraries, town halls and bars when it’s more simple to draw lines and take swings at anyone crossing them.

Civility was a facade, not a real and vital place to gather.

Civility would hear nonsensical or misguided words from another and say “can you help me understand that?” instead of swift, humorless condemnation.

Civility had it exactly backwards, always searching for commonality instead of attacking differences.

Civility was often misguided humor trying to diffuse tension and avoid righteous violence.

Civility’s friends were always ignorant about how the world works, and how people operate.

Civility wasn’t even loyal! It would admonish its own friends for being hostile to a stranger that surely deserved it.

Civility is no longer able to get in the way of a brighter future with a more clear concept of “them” that is certainly not “us.”

Civility was a cowardly, short-sighted, privileged, delusional refuge for those duped into imagining peaceful coexistence despite very real differences, very grave stakes, and irreparable wrongs and harm.

Civility can no longer have us waste time with kindness when we know someone has been wrong. It can no longer whisper to each of us that we too have been wrong, misguided, and harmful even when it was not our intent.

Civility duped people into wondering what they might be wrong about instead of feeling the power and solidarity of forcefully extolling what they were right about.

Civility didn’t understand that asking about an ugly thing was the same as embracing and condoning it.

Civility droned on about unity, completely out of touch with how divided we’ve always been.

Civility actually seriously thought it was more important to first see a person rather than their ideology.

Civility knew there could be lasting division, disagreement, and hostility and wasted untold energy exploring other possibilities.

Civility thought the hero and the villain could be the same character. Can you believe that?! It didn’t believe in heroes or villains, just noble, deplorable, and unremarkable acts strung over the span of a life.

And now that civility is not around to hear us say it: what stunning ignorance!

What an utter, self-deluded fool.

People do intend harm!

People do relish exerting power to get revenge and hold others down!

People do use humor and nuance and questioning to deflect, obscure, and distract from more sinister aims.

Finally civility is not here to insist that those people are fewer than we think—that most generally want to do good and mean well, that misunderstanding and miscommunication account for unfathomable amounts of conflict. That hurt people hurt people and wrong is not a permanent condition.

When someone pledges allegiance to a bad idea, they are clearly and solidly and irredeemably the enemy and therefore deserve to be vanquished.

Civility tried to keep us from the sneering judgment, division, hostility, and war that have always made us better.

Let’s dump this casket of obsolete shit into the ground.

How bizarre that our monkey ancestors surely went through some weird, rough times to live with wild wolves and then fast forward and we’re collecting their poop in bags and buying things from Etsy with their names on them even though they can’t read.

They force us to live with fur and dog smells and chewed things and barking at inappropriate times and vet bills and deep pain when they leave us.

But on the other hand, they are love.

They trust and lick and wag because we’re a little pack.

We return home and they are so fucking excited to see us that all previous thoughts are interrupted for the joyful reunion.

Some are so ready for a walk they’ll bring the leash.

We worry about our bullshit but they just want to hang out or play or bask in the sun together.

We drive to the park thinking about work or that car sound and they’ve got a head out the window soaking in life so hard they have to go to the other window where it’s more fresh.

They’ve got much less time and they don’t have our fancy brains, but they know how to be.

★★★★★

Mine takes me back to a middle school math test, somewhere around twelve years old.

There were no laughs there, but it’s where the story starts.

Little me, struggling as he ever did with math, felt a tinge of discomfort in his gut and needed his entire attention on Mrs. Lewicki’s test questions. He also didn’t want to loose a blaring fart in a hushed room of pencil scratches and turning pages.

He got a pass and went to the bathroom, where desperate attempts to make any kind of bowel progress yielded nothing at all. Which was curious, because it felt like something massive was going on and the pressure only increased.

Flash forward to the hospital ER, where little me paces helplessly waiting for an exam room. Cries next to his freaked-out mother, who watches her kid writhe against the most pain he’s ever felt, unable to do anything. (We didn’t yet have phones for field googling.)

Little me vomited into a handy plastic tray they offered, which made exactly no-one feel better.

Finally, a bed with a chair and a curtain. Vitals were recorded, and several doctors parted the curtain to ask the same questions and prod at my abdomen and leave with unsatisfyingly-blank faces.

The last doctor to come in slid a chair next to me and took his time prodding. He reached my right side, just west of my bellybutton, and I grimaced.

“So it hurts right here, when I do this?” he asked, and pressed harder. Several times. Probing fingers, eye-watering pain.

I fought my instinct to swat at him, growled “yes” through closed teeth, and shot a look at my mom that said “of course it fucking hurts which clues are making that unclear?!” She understood, and there was a glimmer of laughter in her eyes—maybe some kind of relief swirled into her reaction to the barbaric prodding.

Appendicitis. My useless intestinal flair ballooned and needed to come out.

Fast forward again, this time to surgery prep.

A nurse apologizes about the saline just coming out of the fridge and I don’t register why that would be an issue. She gets me a huge blanket, and moments later I realize there’s no warming up when cold liquid circulates through the inside of your body.

But bravery and cheer came easily with people determined to make the pain stop.

They gave me morphine and had me breathe nitrous oxide and count backwards.

The next thing I remembered was waking in a fog, with blurry vision, as a nurse was slipping those drawstring hospital pants onto my legs. I was cognizant enough to be embarrassed, so I asked what time it was.

But did I?

Nobody answered. I like to imagine that viewed from the outside, a kid made a grunt that he thought was a perfectly coherent question.

Forward once more to a little room in the ICU, where the drugs would slowly wear off and my mom could relax into boredom and relief.

My abdomen was sore, but the screaming, expanding pain was gone.

They said I was lucky because any longer and it might have ruptured, a medical term for exploded disgustingly that’s also medically considered to be a bad time.

In this hospital room, it was Jell-O and TV shows and check-ups and bandage refreshing and mostly waiting.

This came with an interesting feature though: it hurt to laugh.

Laughter is a gift I exchange with other people, and sometimes it’s all I have to get through something that scares or overwhelms me. It was the same for little me in that moment, primed for laughter with all the physical pain and frightening rush of activity.

The conundrum of this situation reached full force every time I had to pee.

Being old enough that I didn’t want my mother’s help, I would have to gently scoot out of the hospital bed, shuffle my way to the room’s tiny bathroom, and drop my flimsy hospital pants.

Peeing was easy enough—the tricky part came next.

The very first time, I dropped my pants all the way to the floor where they bunched around my feet. My arms don’t reach the ground, and I couldn’t bend my mid-section without wild pain from the still-healing surgical incision.

Narrating this to my mother naturally made her laugh at the hopelessness of the situation, which made me laugh, which hurt.

“Stop laughing!” I laugh-grimaced at her.

That didn’t help either.

In fact, that made it worse.

There I was: pants to ankles, frozen in logistical quagmire, relieved to be okay, grateful to be loved, laughing at the absurdity and wincing at the pain laughter brought.

I don’t remember how I made it out of that first situation. It’s likely that my mom averted her gaze and raised the pants from behind until I could grab them.

We ended up with a policy that she had to leave the room when it was time for me to pee again. Her laughter from the hallway still got me, but I learned to focus and keep hold of the waist line and everything got a little easier.

This was almost three decades ago, and I haven’t spoken with my mother in fourteen years. But I can feel much of it like it just happened, which makes those laughs precious.

Imagine a quiet, aging suburban neighborhood in a dry, dusty version of the Pacific Northwest United States.

Modest, mismatched subdivisions built in the 80’s and late 90’s, massive conifers throughout. If you could fly and you took a modest leap into the air, you’d notice on a clear day that you’re surrounded by rounded earthen lumps (buttes) in every direction. Snow-capped mountains stand silently and gloriously in the distance.

We’re not in Middle Earth; there’s a Safeway and auto body shops and mattress stores, too.

Okay but come back down. Flying isn’t the point right now.

It’s fall.

Clear, cold air. Shocks of red and orange and yellow and brown still clinging to branches. That sweet smell of decaying leaves certifying to the nose that this is fall. Golden sun that can warm your face while your back is cold, like the steaming mug of cider you might hold with frigid fingers.

We’re going to leave now, and it doesn’t matter how. We can go on foot or we can take a car—either way we’re going to see deer.

It may be a gang of does and fawns silently munching in someone’s front lawn. One of the bigger ones will lift its head and enormous ears to watch while the others continue nibbling. A curious fawn may teeter on gangly legs to get a look at the strange hind-leg-walking monkey things, or the shiny wheeled fright machine we drive by in.

Or we may see a buck, wandering alone with its gorgeous crown of antlers. They’re each preparing for “ruttin’” season, bulking up and getting ready for fighting and mating. I imagine every one to be like some sort of monkish frat boy, almost annoyingly male.

It could be like the male that surprised me on a walk. He was standing in some bushes just off the sidewalk ahead, probably eating something but already staring at me by the time I noticed. A short, stocky one with a massive neck.

“Are we cool?” I asked aloud, wondering how many people are regularly murdered by deer on suburban sidewalks because of poor communication.

He stamped his feet, which to me said “why don’t you come closer and find out asshole.”

So I turned and went the other way.

On a different outing it’d be a different deer.

I scan the sides of the road all the time when I’m driving, always just a bit nervous that a huge body on thin legs will leap suddenly across the pavement.

One night, I stopped in my lane as a male with a strange gait crossed slowly in front of me.

At first I was worried that he was injured, because he threw his front legs forward with abandon and made me think of John Cleese doing a silly walk. He did this while managing to cross so slowly I wanted to yell “oh my god hurry because this is dangerous and not every human driving a car is going to notice and stop for you!” He seemed to stare at me as he continued his exaggeratedly-slow saunter.

This injured-looking male reached the far edge of the road, then trotted off not-weirdly. Like any other deer.

Can deer be funny on purpose?

Now I hope the answer is yes.

We already know from munch marks and hoof prints and clusters of soft brown marbles that our front yard is regularly visited by deer passing through. Probably overnight because we’re usually home and have yet to see them. We’ve planted things they allegedly don’t like, and that has not stopped them from nibbling on them. Some antler-rubbing mangled a young tree that my wife later wrapped in aluminum foil, which looks deranged but seems to have saved the remaining branches.

I will meet more deer after I write this and I wonder how they’ll surprise me.

I know we’d both probably rather imagine flying, but what sort of deer would you be?

This AMA format has been fun to read and easier to respond to than posts I arrive at mysteriously, and I was reading Lou’s post when my answer burbled up.

if you could instantly change one internal pattern/thing about yourself, what would it be?

I spend a lot of energy not asking questions, not making jokes, not publishing things I write, not hitting return but deleting messages.

“That’s a weird thing to ask,” it says. It doesn’t care that I’m fascinated and want to know someone or hear more.

“That’s not as funny as you think,” it says. It reminds me that I’m a cis white male and that often there’s a gift in simply not sharing our thoughts.

“Actually you should find work,” it says. It reminds me that wonder and play don’t pay for groceries.

“That’s too much enthusiasm and you’re not trying hard enough to seem normal,” it says. It reminds me of the yikes reactions and doesn’t mention smiles and warmth and love.

“You could look like an idiot, or worse reveal that you’re an idiot,” it warns. It doesn’t need to explain further because the weight of this often stops me from squirming.

“Nobody wants to hear that,” it says. I can’t prove it wrong.

“People have already said that better,” it says. I should find a way to link to it and not dump more mediocre thought pollution into the ether.

“Your timing is off and the moment for that comment has passed,” it says. It doesn’t care what I would have said or who I wanted to connect with.

“Your unaddressed bullshit really does have the capacity to be harmful,” it reminds me. This one cuts deep and I don’t have any response.

Self-control is important, but at some point fearful self-censorship wrecks what would have been connection and art and play and ... life.

Oh, to silence that voice!

Drugs and alcohol do it, and make a lot of things seem easier, which is why I’ve started limiting drugs and alcohol because I’ve seen where that can go and I know that learning to live with the critic is hard.

The real answer probably isn’t murdering the critic, but hugging it out. Thanking it for its useful cautioning and feedback, and continuing with matters of heart.

So I suppose what I would change is how readily and unquestioningly I let the critic stop me.

I’m just piling on after Annie offered the question and several others answered:

• Keenan
• Estebantxo
• Kerri Ann
• Lou

I often write drafts I chicken out of posting, but in November I’ll finish and publish four of them—probably one per week. Welcome to post one!

I don’t want to be wrong, but I’d rather know than be a real-life Arrested Development character.

I can bump into a new blunder and recognize it if I’m paying enough attention, but it’s less embarrassing to search them out in a controlled environment.

Sometimes a book or blog post might illuminate the shape of a bad idea, but more often it’s a conversation with a friend that goes like this:

Me: “[thing] is dumb because [dramatized rationale].”

Friend: “Actually you’re dumb because [thing] turns out to be [other way of looking at it].”

Me: “Oh really? So [excessively-sarcastic question that makes a good point and is funny but still serious]?”

Friend: “Yes, that makes perfect sense if you completely ignore [critical detail], but [new idea I will be thinking about days later and/or forever].”

There’s always material, and the parade of ensuing ideas and stories and questions can be mutually entertaining and thought-provoking. And it’s never that formulaic; more flabby ideas might tumble out from unexpected places, or the conversation could leap to a bigger and more interesting clash of perspectives.

Shining light onto murky ideas lets me expand my model of the world and move on to other things I’m probably failing to grasp.

I’m reading an interesting book about being wrong that’s naturally returned me to a great question: what have I been wrong about?

I don’t mean the pronunciation of a word or mistaking a stranger for someone I know, but the big ugly ones. The wrong ideas that, once recognized, led to changes in my life and often for the people around me. Or maybe smaller ones that change how I see something every day.

This question happens to work for the WeblogPoMo AMA challenge, because anyone can ask this and I’d sure read the answers they were willing to share.

What have you been wrong about?

The problem is that I am not eager to tell you, unknown reader, on the internet. I wouldn’t expect you to get excited about telling me either.

I started writing this because it matters to me. I’m genuinely anxious to share, so here are some things I’ve been wrong about.

Grownups know what they’re doing, and when I’m a grownup I will too.

We’ve each had to navigate the rapid changes of childhood, hopefully with steady guidance from parents and teachers and adults who knew what they were doing.

And then maybe you reached the same moment too—perhaps the precise moment of becoming a grownup—first realizing “grownup” is a scam.

Grownups are fellow travelers making things up as they go, they’ve just been at it longer.

Being at it longer doesn’t mean they have solid answers, either! Wisdom is a product of time and insight, which is not guaranteed with age.

For me this was a startling revelation that instantly disintegrated barriers separating me from people I already knew.

Parents and teachers and relatives morphed into companions with stories and insecurities and successes to exchange and learn from.

It had been a kind scam! They offered guidance and safe places to learn while they were still figuring their shit out so I could prepare for a world that is not always kind or beautiful or safe to grapple with.

With the ruse behind us, we can see each other in new ways without that pretense. But any love survives the transition, and it’s abundantly clear even when adulting frequently is not.

I need a church.

I spent many earlier years invested in religious study, outreach, volunteering, and thought—all of which I assumed I needed. (And maybe I did; it was mostly nourishing and challenging and good at the time!)

Eventually I realized that the church wasn’t the answer because some things didn’t square for me and deeper, incompatible beliefs were waiting beyond.

I loved having a social structure I knew how to thrive in, and a sturdy set of ideas to clarify and practice together with kind, earnest people. Rituals offered me profound solace in times of great yikes.

But I don’t need a church. I need connection to other people. I need a sense of community that may be constrained and overly-influenced by a church. I need to struggle with the real stakes of having one life and assuming it ends with a permanent return to nothing. I need to figure out how to live with uncertainty and doubt as companions.

Being gay is wrong or unnatural.

I was sure for a time that being gay was some kind of error. I accepted a soft version of the idea that intended kindness despite registering an aberration.

Then I had gay friends in my life and I just saw love and joy and heartache and bravery and garden variety bullshit and wondered how I managed to fixate so narrowly on a difference nobody ever needed me to weigh in on. (Also who was that helping exactly?)

Any idea I have about a person that leaves me less open to them is probably not a helpful one unless I’m in immediate physical danger.

If I’m preoccupied with our differences that’s a me problem, not an out there problem.

Interaction is a zero-sum game where the more righteous, more ambitious, and more intelligent are able to get their needs met.

I went on for a long time without realizing I thought like this, and learned (slowly) how damaging it is to myself and others.

This is still a popular assumption, whether adherents realize it or not. Everything is a game they intend to win, so there must be losers and unimportant background characters. I’ve been surprised at how far this can apparently go.

But everything doesn’t have to be a contest or a fight. It’s possible to share, ask for help, and listen just to hear someone.

It’s possible to be in the presence of another person without having to give something or get something or fix something, and those moments can be important on their own.

I’d be embarrassed that I had to learn this if I wasn’t so thoroughly grateful to come around with so much life ahead to experience differently.

Asking for help is an act of weakness.

A continuation of the previous one: that things like apologizing and asking for help diminish the strength of your position—the equivalent of lying down and preparing to be walked on.

Now I see that it takes honesty, bravery, and self-awareness to ask for help and own mistakes and venture down an unknown path with the hope of reaching a better place. It turns out that’s what strength looks like.

Emojis are dumb.

Childish, ambiguous, unnecessary little decorations that ruin perfectly capable sentences and phrases!

My reaction was mostly 🙄 until I had some interactions with people whose language I didn’t know, and those stupid little icons were the only thing that helped me discern tone and (miraculously!) meaning.

Aren’t words already symbols with plenty of ambiguity? Now I can 🤷‍♂️ and 😬 and be fascinated by how 👍 is used and interpreted differently across generations.

I may be too old to say it this way, but emojis are 🔥🔥🔥.

Doing the right thing should feel good.

I expected I’d always be able to recognize the “right thing” because it would naturally feel right or good at some point.

Sometimes that happens!

Sometimes.

I’ve been proven wrong by wobbly, deeply-uncomfortable, oh-my-god-can-I-disappear-into-a-hole conversations I needed to initiate as a matter of conscience. The worst of them didn’t even go well or end on an uplifting note. The “right thing” may forever be murky and unresolved. Sometimes the healthiest course of action still felt like utter, inescapable shit.

Tattoos are poorly-considered body decoration.

Questionable decision-making as visible skin art. That’s what I saw for a long time, even as my partner of many years got one.

I’m not sure when it flipped, but now I see my complete existential panic about committing to imagery in such an intimate and permanent way.

Every single tattoo, even the most regretted drunken souvenir of whim, was a commitment with a story and a certain amount of suffering to bear. I love asking people about their tattoos and inevitably learning about something that’s important to them. (I’m still rarely sure when it’s okay to ask though!)

When I get my first tattoo, it will be a small personal triumph.

I will never seriously prefer a vegan pizza.

I miss you, Big Nonna’s.

It doesn’t make any sense because I’m not vegan and I am a big fan of gluten and pepperoni and cheese. The omnivorous pizza joint with all the wood-fired brick ovens was fantastic, and still in the quiet of my heart I often preferred your unfathomable pizzas and marveled at the phenomenon.

I’m including this for the strong, lasting surprise.

Maybe anything is possible.

What matters most is how I think about things.

I can’t be thoughtful without spending time thinking.

My only way of understanding and experiencing everything is through this fancy monkey brain that can get lost in what it wants to see and find clever ways to avoid what it’s afraid of.

Thoughtful is important! I can always stand to be more prudent, more careful, and more kind with any of the creatures in my life.

But my capacity for overthinking can be stunning.

If we’ve spent time together, you’ve watched me do it with something. This is still fresh.

Thinking is important, but what matters most is what I do with the time I have.

What are you doing to prepare for this coming winter?

I live in a place with cold winters and real snow again, but unlike the northeast U.S. only some of it gets plowed and there’s no salt—just crushed lava rock since that’s an abundant resource. Our street stays pretty snowy, and roads everywhere can stay surprisingly icy.

With my humble Volkswagen sedan, my winter strategy was don’t go anywhere for a few months. (We’re just two introverts and a blanket-and-pillow animal, so that works fine.)

But this year is different!

This year I parted with my old sedan to welcome a long-awaited first pickup truck into my life. We’ve talked about getting a pickup for years, to help with landscaping and house projects and of course be a more useful apocalypse vehicle. But EV or gas? And is there any truck on this planet that I actually like? And why part with a car I like that’s paid off?

The storm of midlife crisis nudged me toward action, and I ended up with a Jeep Gladiator and I love it!

We had a Jeep Wrangler growing up that was a lot of fun to drive, and this thing is not quite a Wrangler and not quite a normal pickup truck. It’s a wonderful, anachronistic, and capable identity crisis on wheels. We’re a perfect match, because that’s also a flattering description of me on rollerskates. We’ve already brought home trees and shrubs and pavers and towed ripped-up sod to the recycling center. We took the top off and drove around way before it was warm enough for that to be comfortable.

So my winter strategy can now include going places!

I’ve been riding offroad trails for the first time in my life, apparently without enough equipment or experience—and one thing nobody wants on winter roads or summer trails is to get stuck.

I finally made a little recovery kit with a snatch strap, shackle, gloves, and traction boards that can live in the truck bed for the winter. (I already had a shovel, fire extinguisher, first aid kit, jump-starter, tire pump, and some other emergency items stuffed in there.) That should be enough to get myself un-stuck, or even help others. I’ve wanted to be a benevolent pickup person but it’s harder to pull off with a sedan.

The other thing I’m preparing for is another presidential election.

No matter what happens, I am not thinking the outcome will be full of national unity, understanding, and patriotic goodwill.

Instead of my usual strategy of experiencing dread and trying to become numb inside, I went to an event organized by the local library on post-election civic resilience.

At 41, I was the youngest person in a room of forty or so people.

It was heartening to hear from people leading local organizations already fostering community and encouraging people to mingle and argue and listen to each other. There are lots of local issues beyond politics, like water use and wilderness conservation and the usual strains of a rapidly-growing community.

It was nice to feel among like-minded people meeting and sharing and arguing about yard signs and civic responsibility and welcoming difficult conversations. I left learning more about Braver Angels and books to read and other groups to pay attention to and participate in.

I could still literally and/or metaphorically get stuck in a cold, dark environment ... but there’s a chance I could use the tools I’ve got to dig out, get moving, and help others do the same.

I read manuals front to back, I find shop guides to see how you might take apart or replace bits and pieces, and I give myself modest projects that I can tackle without mortal risk. (I like my car, but we shouldn’t confuse me with a car guy.)

I also like cleaning cars.

I find it satisfying to keep it in good shape and keep it looking and smelling and feeling as pristine as I can. It’s my strange sort of contribution to the life of the thing after the astonishing design and engineering that brought it into existence. Also it’s fun and kind of relaxing.

I haven’t met many people that share my joy of car cleaning, but seemingly everyone loves a clean car and a few have been thrilled to let me have a go at theirs.

I’m not going to keep telling you about cleaning cars though, because I had an epiphany that’s useful even if you don’t have a car or the near-OCD that makes cleaning it fun!

Here it is:

You can use car detailing stuff on things that are not cars.

Allow me to share my favorite examples.

An interior should not be a greasy, shiny, Armor All mess. It should be clean to the touch, have an even luster, and a faint whiff of coating that protects it from drying out and bleaching in the sun. My favorite product for this, which I arrived at after much trial and error, is Griot’s Garage Interior Detailer.

I’m not even linking to it because I’m not here to sell it to you, I’m here to tell you it works great on a mouse, or a keyboard, or a big Cordura mouse mat that’s otherwise tricky to clean. It’s made for tactile car interior surfaces, and electronic gadgets are made of a lot of the same materials.

A cheap set of non-scratching brushes can be great for cleaning out crevices and impossible-to-reach places inside a car or its engine bay, so of course they work great between keyboard keys and mouse buttons and screen edges. (Pro tip: designate a filthy engine/exterior brush and don’t use it for anything else.)

Lots of enthusiastic people on YouTube can show you how to use these things, if you want.

My spouse, who gets me and is a good gift-giver, introduced me to Hyper Dressing. First of all it’s called Hyper Dressing, so it’s got that going for it. Second, you can buy Hyper Dressing concentrate in large quantities—and that’s important because if you like cleaning things and you live in a high-UV environment like Bend you’ll probably appreciate it.

Hyper Dressing is oilier and shinier, so I don’t use it for any interior bits. It’s perfect, however, for plastics, exterior metal, and rubber trim. It can take dried-out, sun-beaten plastic and make it shine like new. It’ll not only make rubber seals look nice, they’ll stay supple and age better, and in some cases (like door jambs) seal more cleanly without being sticky.

But I’m not here to sell you Hyper Dressing either, I’m here to tell you that you can use it on a vinyl window frame, or a metal light fixture, or a plastic deck box, or a powder-coated umbrella stand, or that thing on your porch that got wrecked by the sun and it’d look great if you could find the right thing to restore it. Hyper Dressing is that thing!

I’m dabbling in ceramic coatings now so I’ll probably wind up with more ideas for my not-car toolbox.

I hope this is useful for the one other person that’s excited to think about it. Also, hey you! 👋

While I’ve been playing with new Coolify servers, I’ve started the new habit of limiting SSH access to my Tailscale network.

This drops waves of illegitimate SSH connection attempts and makes me feel smarter.

Arriving at this wasn’t obvious; I first tried Tailscale a few years ago and honestly didn’t know what I should be doing with it.

Tailscale creates a private network you can add devices to, almost like each one is joining your local network from wherever it happens to live. This applies to pretty much whatever device you can think of: your (Mac, Windows, Linux) desktop machine, your phone, your NAS, any servers you have running in the cloud. Take a few seconds to install Tailscale’s client, and the device pops onto your “tailnet.”

This comes with some perks! If the device is only addressable via IPv6—something I’m still figuring out how to deal with—Tailscale makes it easier to access.

Each device has a name you can use to address it, which is nice for SSH. I’ll show you what I mean.

You might connect to a web server via its IPv4 address like this:

ssh ubuntu@55.55.55.55

You’d have to remember, of course, that 55.55.55.55 is your server’s IP address. Or create an alias for it in ~/.ssh/config. Or create a DNS record so my-server.my-domain.tld resolves to it:

ssh ubuntu@my-server.my-domain.tld

Every Tailscale device gets a name, so if I added that server to my tailnet and called it myserver, I could use that name without remembering an IP address or assigning a public hostname to it:

ssh ubuntu@myserver

Pretty cool!

That doesn’t work outside my Tailscale network—it’s only for the special club of authenticated clients using this private tunnel.

Eventually I realized this is a rather important feature.¹

I’d been setting up new web servers to send their logs to Axiom, where I created charts and alerts for SSH authentication attempts and successes.

If you’ve ever paid attention to SSH connection attempts, you know that as soon as you’ve got a server online there’s an endless stream of garbage. It’s unsettling.

Seeing all these sketchy failures again, I thought it’d be nice to only accept SSH connections from my Tailscale network. I already use Tailscale machine names because it’s convenient—why not wall off the public port and use the tunnel instead?

I have yet to see a Tailscale device lose its connection, and if my usual desktop machine burst into flames I’d still have other devices I could use to access to the VPS.

It turns out limiting connections was quick and straightforward! Steps:

1. Enable the UFW firewall. (Typically disabled by default.)
2. Tell UFW to allow traffic on web ports (80, 443), since it’s a web server.
3. Tell UFW to allow incoming connections specifically from Tailscale.
4. Tell UFW to drop all other incoming connection attempts.

Before:

After:

Step by Step Instructions

This assumes you’ve got a VPS running Ubuntu and a Tailscale account, and you’ve got a root session running on the server.

1. Install the Tailscale client

• Run curl -fsSL https://tailscale.com/install.sh | sh to install the package.
• Run tailscale up to connect.
• Click the resulting link, add the device, and optionally disable its expiry. (Machine settings → Disable key expiry)

2. Enable UFW and customize its rules

Don’t freak out! The rules don’t apply until you run ufw reload.

ufw enable
ufw default allow outgoing
ufw default deny incoming
ufw allow http
ufw allow https
ufw allow in on tailscale0

The last line allows all inbound connections from the Tailscale network, which is tailscale0.

3. Reload UFW and restart SSH

Okay now freak out! Take a moment and be extra sure you didn’t forget about any other inbound connections you might need to allow, or non-Tailscale devices that may need to get to this server. If those are urgently necessary and you didn’t add rules for them, you’re about to lock those things out.

Once you’re ready, restart both services:

ufw reload
service ssh restart

4. Confirm joy

You should be able to continue connecting via SSH using your Tailscale device name or addresses, but not from outside. Your UFW logs will also show you all kinds of blocked traffic that can no longer even attempt to establish SSH connections.

And now you can sleep a little easier.

November 13th Update!

Bertrand kindly wrote about this article and asked how I solved port 8000 still being open on my server running Coolify.

It turns out I didn’t! Visiting the bare IP address at port 8000 connected me with Coolify, which is bad because only 80 and 443 should have been responding. This is an issue specifically with Docker and UFW that at least one project aims to fix.

I followed these wonderfully-clear instructions from step 3 onward to bind Docker to Tailscale’s IP address, which solved the problem.