Hi, I’m Adam and past-you thought it would be nice to hear about what I’m excited about right in your email or feeds. I appreciate you sticking around, especially the folks who have been around since 2012!

We continue to feel all sorts of ways about machines generating code, writing on the web, composing images, deciding things that humans used to decide. Robots continue to feel nothing. I bet you can see the problem here.

I’m starting to think that asserting our humanity is one way through this particular forest. We can’t spew text as quickly as machines, but we can act with greater intent and make holistic decisions. We are just as likely as machines to hallucinate ideas and facts, but are better able to ground them in intuition and edit out the junk. (Some call this taste.) We aren’t always the most eloquent, but are far more interesting, surprising collaborators than over-agreeable machines.

Granted, it’s still remarkable to have a capable, fast-coding coding agent as a (nearly) always available teammate and foil. The ability to say “yes” to so many more projects and ideas is probably, on average, still underrated.

But wow it is great to get away, even for a weekend, from monitoring the agents, keeping them moving forward, reviewing their drab prose, grilling ideas into specs, watching agents grind those specs into code, giving them little ideas on how to code better or implement better designs, and keeping up with the myriad Joneses who are doing the same. So much keeping up, so little really considering if what we have is any good.

Even better when that time away from the machines is spent doing deep, rewarding thinking. Not even the coding, thinking about coding, and reading about coding I used to do. Lately, it’s reading philosophy or chunky non-fiction by any one of my dozen or so favorite writers. Perhaps writing about those readings. Turning the words of people we’ve not even met into our own ideas, figuring out what I really think. Reflecting on ideas of people from millennia or just a few months ago. Wild stuff!

In pursuit of keeping our coding agents moving forward and generating outcomes, we’re flattening ourselves. We went from the expressiveness and aesthetic excitement of code to writing out specs, feeding specs through the agents, and trying to find the time to carefully review the results. Almost all the time at delegating and deciding and managing these gizmos. And yeah, that flattening can feel pretty mediocre at the end of the day, regardless of how many outcomes or widgets pop out the other end.

If the clankers have you down, I recommend: revisit your favorites, emulate them. You’ll probably make something novel. It’s called standing on the shoulder of giants, it worked for Sir Isaac Newton!

If that doesn’t work, take a weekend away from the tokens. Touch grass or concrete, let yourself unfurl in the best life has on offer in your neighborhood or city.

The new sabbath was an offline weekend or vacation. The new time offline is time away from tokens.

Related: reading is really my jam, lately.

Give them a chance: dirt, screenshots, and going wildly deep on something. Make interesting things, interestingly.

If I’ve learned anything from working with coding agents on a team, it’s to worry less about that one code review where someone points out a howler of an issue with my code. Whether a coding agent generated that code or I typed it in by hand, acting on my teammate’s feedback is an opportunity to show that I care about quality, I listen, and I follow through.

The outputs of using machine intelligence are a bit inconsistent at this point. Some people are getting great results, some folks are getting mediocre results and putting them out there anyway. Plenty of people are getting countless results but not finding them so compelling that they actually put them out there. Not many folks have put something out that could not have existed before or is a game-changer that was conceived or built entirely with AI. But I’m optimistic that we’re currently draining the swamp of not-so-great ideas that were possible before coding agents but too costly to put software development (or artist commissions) into. Eventually, the swamp will drain and we’ll pursue higher quality ideas.

Or it will turn out we’re all out of good ideas and “Good Vibrations” was the peak outcome of the American cultural/technological experiment. Even if the feeling in that song is teenage dumb-love, at least the way technology meets ideas is fantastic. Hopefully, making technology more accessible means we get better ideas+feelings in the future.

Thanks for reading! Subscribe if you’re new around here.