Blog

Nowhere Near My Art

I’ve been using AI in limited ways more frequently—mostly at work, where the expectation that we leverage Copilot (in all its guises) wherever possible is very, very high. At home, I use it sparingly, mostly as an aid to learn specific technical topics a little quicker (Apple development topics, in particular). I don’t use it to write any sort of prose, generate art or music, or speak on my behalf in any way, shape, or form. Which leads me to what’s quickly becoming the main tenet of my personal approach to AI:

Nowhere near my art.

I think there is utility in LLMs, especially for technical topics like coding. I’ve found it useful for helping me to analyze and understand a new code base, and to research technical topics I understand at a high level but want to learn more about (e.g. SwiftUI development, APIs, etc.). But I don’t view coding at home or work as my art.

Work is work. I take pride in what I do while I’m clocked in and do everything I possibly can to insure what I produce maintains high standards (frequently fighting with AI and other internal tooling and processes while I’m at it). But it’s not the art of my life.

At home, that art consists of things like writing, making music, drawing, collage, and thinking through the interesting ideas behind all of it. I often code things to support that art (like this website, my email newsletter, random tools), but that code is never doing the thinking behind or the generating of that art. And I don’t ever want it to. There will never be a point where I want to prompt a song, blog post, or illustration. At that point, it’s not my art, it’s the average of whatever the LLM was trained on.

Taking the average, well-worn path when it comes to coding (at least for the projects I work on) is usually a good idea. I’m not doing anything truly novel on the code front, so sticking to boring technologies and well tested methods is the way to go. LLMs can be very useful here.

But I don’t want the things I consider my art to be average and boring. They might end up that way given my current skill/talent levels, but those can (hopefully) be improved through hard work and time. As they should be. The work and journey that goes into that is where the fun is. Why the hell would I want to replace that?

So, the tenet remains. AI can be helpful as a supporting tool (which is where I think it’ll ultimately land once the bubble bursts and the hype subsides). But I don’t want it generating something that’s supposed to be in my voice. The day I prompt my way to a song or blog post is the day I lose my voice. Nowhere near my art.

I still don’t know how to square up any of this with the massive ethical problems AI and the billionaires who control it are creating and doubt I ever will. For now, I take the only approach that’s available to most of us living in a capitalist society: reluctant participation. I use what I can to improve life for me and my family and friends, do what I can to change things for the better where possible, and hold on to my humanity and individuality at all costs.

Irregular Communication

I’ve been sending an email newsletter in some capacity for something like 15 years. The current version is called Regular Communication and consists of 38 published issues. There are probably about 200 or so in various abandoned email service provider accounts that I never cared to archive. Let’s call it 250 for safe measure. Over the course of 15 years, that averages out to about 16 emails per year… not exactly a regular practice.

I’ve oscillated between two working methods:

1. Trying to send on a regular cadence, at times an email a week.
2. Only sending when I feel like it, with sometimes weeks or months between emails.

Obviously, the second method has been my natural tendency. The first method was usually a result of wanting to grow the newsletter or use it as a marketing tool for a side project like the email books and courses I used to publish. It leaned heavy into the oft heard advice of “publish frequently” that content creators have been subjected to since before the term content creator even existed.

But that’s not really what I use the newsletter for anymore. I turned off the lights in the email shop a while back and don’t have plans to monetize my newsletter anytime soon. Growing it organically—if that even exists anymore—would be cool, but I have a strong aversion to social media and ads so you won’t see me posting about it on LinkedIn or paying for promotion.

I use it almost entirely to force a sense of curiosity on my end and share things I find interesting with others.

So, it dawned on me that it might be time for a slight name change. And by slight, I mean a difference of two measly letters. As of today, I’m calling my email newsletter Irregular Communication. Yep, just tacking an I and an R onto the front to better reflect my intentions. You probably already figured it out based on the title of this post, haha.

Some of you may recall that it actually used to be called The Intermittent Newsletter. Similar concept, but a bit more of a mouthful. Either way, the same concept applies:

It’s an email newsletter about nothing and everything, with a side of music.

As always, you can sign up for it right here and you’ll get the next issue whenever I feel like there’s enough worth sharing.

Trusted vs. Untrusted Agency Amplifiers

Like everyone else in tech (and increasingly outside of it), I’ve spent a lot of the last few years thinking about AI and how it can be used in my work. More and more, the framework I’ve been using to evaluate AI tools—or any tools for that matter—can be boiled down into the terms: trusted agency amplifiers (TAA) vs. untrusted agency amplifiers (UAA).

I think of tools as agency amplifiers. They give you agency as it’s traditionally defined: the ability to make decisions and act independently. The right tool allows you to do things without roping someone else into the equation or makes team efforts more efficient with better outcomes.

A pencil lets you think on paper. A pen makes the record of those thoughts more durable. A printed book can spread those ideas around the world.

All of them amplify your agency, but to different degrees and with different levels of trust. A pencil and pen are very trustworthy, as they create direct copies of your ideas. It’s your hand putting those ideas down on paper. A printed book is relatively trustworthy—still your ideas using words you wrote, but the editing, typesetting, and printing processes can introduce errors. Writing on the internet often compounds those types of problems and introduces a whole host of others. You get the idea.

I value trusted agency amplifiers over untrusted ones. Obvious, yes, but increasingly important in today’s world.

Software has been a huge boon for personal agency: calculators, spreadsheets, email, desktop publishing, the World Wide Web, e-commerce, early social media, etc. It’s empowered people to build and connect in ways that weren’t even imaginable a hundred years ago. Software has eaten the world. And with AI now eating all of software, there is a massive proliferation of agency amplifiers out there.

But what we’ve seemed to lose is trust in a lot of those tools. There’s more software—both from individuals and smaller companies as well as tech behemoths—but so much of it is broken. Plenty of it was broken when it was just humans building it, but it feels so much worse now that AI is tasked with writing code and building the systems people are supposed to rely on for damned near every aspect of their lives. And the avenues for remediation are narrowing.

I want to use tools that both amplify my intentions and are trustworthy. My definition of trustworthiness here includes two criteria:

• Fidelity: The tool scales my intentions faithfully e.g. it doesn’t hallucinate words or code I would never write.
• Reliability: The tool works as advertised, with minimal fussing and clear paths to remediation if problems are encountered.

Beyond that, my framework consists of asking myself two questions:

1. How much agency will this tool provide?
2. Do I trust that it will amplify my agency in the right way?

The thing to note is that the second question isn’t, “How much will this tool amplify things?” It’s about doing it correctly. It’s tempting to want to maximize as much as possible (late capitalism’s default mode), but that frequently leads to problems that are too numerous to list here. I want to use tools that amplify but do so in a way that preserves the quality and care I like to think I prioritize in anything I do. I could enlist an AI bot to write blog posts for me, post across social media, optimize for engagement and growing an audience… but all evidence points to those tools producing error prone slop that I would never want to pin to my reputation. And the audience it’d attract would be people to whom slop is acceptable—not the audience I want to court.

If a tool is lacking in either fidelity or reliability, then it needs to provide a hell of a lot of agency to adopt it—and that adoption would be in a very measured way. When it comes to AI tools like Claude, Claude Code, and Copilot, I think there is a lot of agency they can provide, but both the fidelity and reliability are lacking, so I temper their role in my workflows. Instead of prompting a coding agent to build something in its entirety, I prefer to ask questions about architecture and implementation, then review those suggestions, verify any claims as much as possible, and write the code myself before testing it as I normally would pre-AI.

My fear these days is that a lot of people are only looking to maximize their agency without caring about whether or not the tools they use are doing anything correctly. And that lack of care is when we get into real trouble.

In case you’re wondering what tools I actually use and how they fit into my framework, here are a few examples. Have any of your own? Email me and share them.

Some things I consider trusted agency amplifiers:

• Pencil, pen, paper, and well-built analog tools
• Logic Pro X
• Obsidian
• Email
• Good friends, families, and professional networks
• Ghostty
• NetNewsWire
• Git
• The non-corporate internet
• HTML and CSS
• Apple hardware

And some untrusted agency amplifiers:

• AI in all of its forms
• Modern social media

Finally, a few in-betweeners:

• MacOS and iOS
• Microsoft products
• Establishment news outlets
• A lot of web and technology frameworks

I will note that I think this is, like so much in life, a spectrum. That in-betweeners group is much larger than what’s listed above. There are degrees to which things are trustworthy. But it’s still a good framework for me and think it might be helpful for others, too.

Curation Forces Curiosity

One of the main reasons I like sending out an email newsletter is that it forces me to be curious about things.

I mean, I’m naturally curious about things, but I have found that—as the world gets more complicated and moves at an unsustainable pace—that curiosity starts to get sanded down. I feel information overload very acutely and the first thing to suffer is my sense of curiosity. If there’s too much to keep up with, why bother? It’s a sentiment I hate.

I’ve found that curating and sending a newsletter with stuff I find interesting is a good way to combat that. By committing to sending it, I’m also committing to staying curious about the world and finding interesting things to feature. It forces me to open up my RSS reader (happy 23rd birthday, by the way), actually read things written by other humans, consider their viewpoints, reflect on my own to add some commentary, listen to new music, and more on a regular basis. It’d be a lot easier shut it down and spend my time playing video games, but we all know that’s not exactly healthy…

That forced curiosity makes me a little bit happier. I’m not kidding myself into thinking anyone else is relying on me sending out yet another email newsletter, but just the act of curating, commenting, building, and sending something out into the world is a potent medicine.

The Albums That Made Me

I was talking to a buddy of mine about music and he shared his top 10 albums. I didn’t share mine, because I know I couldn’t whittle them down to 10.

I used to spend an inordinate amount of time compiling such lists with friends for fun and to impress girls back in high school. I mean, High Fidelity is an all-time favorite book and movie, so it makes sense. For a long time, I was Rob, Dick, and even Barry—ruthlessly judging music, making lists, stating opinions as if they were facts, and rating others based on their taste in music. I still do all of that sometimes, but like to think I’ve mellowed out a bit as my own tastes have changed and I’ve realized that music taste is largely subjective.

Going back to that conversation, I was tempted to compile my all-time top 10 greatest albums list. I fired up a new note, starting plugging away, and realized it was a futile attempt. What am I rating? What constitutes greatness? Musicianship? Production? Emotional impact? Sales? Cultural impact? Influence on other musicians?

It’s an impossible task.

So, I took a few steps back and decided to compile a different list, one that consisted of the albums which were most impactful to me. I quickly blew through the nice, round limit of 10. But, I’m happy with where I ended up—a nice, concise 26. Instead of siloing it in a text thread, I decided to share it here for anyone curious about the albums that left an eternal impression on me. I’m not giving any reasons as to why each one impacted me so much (that’d take more time than I’m allotted in this life), and not ranking them in any particular order, but just throwing them out there. If you are curious about the why behind a pick, email me, I’d be happy to rant a bit.

Here they are:—the 26 albums that made the most impact on my life (so far):

• Live at Luther College, Dave Matthews and Tim Reynolds
• Multitudes, Feist
• Yankee Hotel Foxtrot, Wilco
• Kid A, Radiohead
• A Ghost is Born, Wilco
• In the Aeroplane Over the Sea, Neutral Milk Hotel
• Texas Flood, Stevie Ray Vaughan and Double Trouble
• Belafonte at Carnegie Hall, Harry Belafonte
• The Joshua Tree, U2
• The Rhythm of the Saints, Paul Simon
• Give Up, The Postal Service
• Come Away With Me, Norah Jones
• White Ladder, David Gray
• Thank God We Left the Garden, Jeffrey Martin
• Rufus Wainwright, Rufus Wainwright
• Abbey Road, The Beatles
• The Midnight Organ Fight, Frightened Rabbit
• American Recordings, Johnny Cash
• Grass Punks, Tom Brosseau
• Grace, Jeff Buckley
• The Navigator, Hurray for the Riff Raff
• Cê, Caetano Veloso
• Buena Vista Social Club, Buena Vista Social Club
• Sweet Tea, Buddy Guy
• Feathers and Fishhooks, Rayland Baxter
• OK Computer, Radiohead

Which albums helped make you who you are today? Email me and let me know, I’m always looking for more music in my life.

Making Art is Part of The Fight

Take 15 minutes and watch this talk from Amie McNee. I’ll wait.

Fuck, I want to quote the whole thing. There are so many good points.

The part that really got me thinking—especially in light of shit like what’s going on in the US right now with federal agents murdering people on a regular basis—was this part in the middle:

So many of us are experiencing a sort of existential depression. This idea that, you know, we consume the news and we’re like, “I can’t do anything about this. I have no agency, I have no power. I have no purpose.”

And we’re seeing an epidemic of meaninglessness. Like we just don’t feel like there’s any use to us. Dr. Alok Kanojia talks about this, particularly with young men. And because we have this huge, like, “What’s the point of going forward,” we see a lot of people relying on lateral experiences. So because there’s no point to our future, it’s all hopeless, we’re instead indulging in things like phone addiction, gambling addiction, porn addiction, game addiction.

I truly believe a solve here, when we are consuming so much hard news about the world, and it’s so reasonable for us to feel so terribly depressed about the future, is to come back to our art. To come back to our creativity.

And please remember, like, you can be creative in so many different ways. No one is exempt from this conversation. When we create, we have agency in a world that says that we’re too small to enact change.

Two things are stuck in my mind.

One: That those of us who are depressed and feel helpless watching the rise of white supremacist fascism in the US can benefit from making art in any way possible. Even if it seems trivial, it’s worth it. Don’t stop protesting, don’t stop fighting, but spend time away from screens, social media, and the machines extracting our attention, and focus it back on making something that brings you joy, comfort, and community. It will rejuvenate you and give you more energy to keep the fight going.

And, as Amie reminds us, just taking up space sharing the things you create is an inherently political act. So if you’re feeling down about making what seems like silly art, remind yourself that you’re still being political and making the world a better place with your it.

Two: What’s the overlap between ICE agents and people who make art? Probably fucking miniscule.

I can’t help but think that so many of those assholes once had creative hobbies, dreams, and imagination. And, at some point, that sense of creativity was crushed. A parent or teacher telling them it was a waste and unbecoming of them to focus on such frivolous pursuits. A society saying they had zero value to provide outside of economic contributions. Them without any agency over their own lives.

And what happens when people lack agency? They lash out. They do things that hurt themselves and others. They seek it where they can get it. And the world suffers for it.

Right now, the most evil people in the world are promising them agency, a place to channel that anger and exercise power. But it’s ironic: They’re part of an agency, but they don’t have any real agency over themselves. They’re told what to do, where to go, and who to hurt. But they don’t have the agency to think for themselves. They lack the agency to encounter humanity in all its forms, realize they’re a part of it, and change the world for the better by helping all of humanity, not just those who share their pigmentation through dumb luck.

What would’ve happened if those people were encouraged to be creative throughout their lives instead? Pursue things that brought them joy? What if there was social infrastructure in place that allowed them to do that?

I’m not naive enough to think that it would solve all of these problems. There will always be evil, hate-filled monsters in the world. Some people are so broken that hurting others is the only way they can feel joy. But I know that—if we lived in a world that stopped sidelining creativity and extracting all of our attention for money—there’d be a hell of a lot fewer monsters out there.

Humans are meant to create. It’s one of our defining characteristics. We tell stories. We build tools. We sing songs, cook food, and draw pictures. Creative outlets are an integral part of what makes us human.

Without those outlets, we lose our humanity and start seeing other people as less than human, too. With them, we get to share in our collective humanity. We grow individually, in our relationships, and in our communities.

All that to say: Creativity is vital. Exercise it every chance you get. And encourage those around you to do the same, even—especially—when it feels trivial. Keep fighting, but remember that making art is part of the fight, too.