Things of interest (July 2026)
Here’s a smattering of things I found interesting lately. Tools, articles, bears, beets, Battlestar Galactica.
SVG has slowly, steadily become one of my absolute favorite web technologies. This set of free SVG-based web graphics by Matt Visiwig will instantly improve your project with a copy/paste.
Email.md - responsive emails, written in Markdown
I don’t write an email newsletter anymore. But if I were ever to start one again, I would 100% use this.
CanIRun.ai — Can your machine run AI models?
Cool mini-site that does its best to determine your machine’s capabilities (via browser APIs) and tells you which models will run:
[great | well | decent | tightly | barely | nope]
It actually guessed my processor wrong, but you can also just fill out the details to ensure it’s accurate.
Jurassic Park computers in excruciating detail
Fabien Sanglarda re-watched Jurrasic Park and researched every computer/software he spotted. A true nerd out. Perhaps this makes me basic, but my favorite is the CM-5 that prompted this choice quote:
“It’s a Unix system. I know this”
Matt Rickard predicts that “winning software teams will throw away orders of magnitude more code than they merge.”
He’s building a new git server implementation and wire protocol to help invent this future.
Proving a human wrote something
Garen Torikian came up with a system for proving that you wrote something. Semoi “records the rhythm of writing — every keystroke, pause, and revision — and seals it with a cryptographic signature the instant you finish. The text is hashed at mint time and never stored. What you publish is a certificate anyone can verify.”
Interesting idea (and kudos to the execution), but too easy to game to work at serious scale. For now, I’ll continue proving I wrote something by injecting original thoughts and pleasant SUrPRISEs into it. 😎
The great divergence in software engineering
Paul Stack, who has teamed up with Adam Jacob on Swamp:
What I’m seeing is three groups. Teams that have embraced AI and are compounding every week. A growing number of teams and leaders actively moving the other way, loudly distrusting AI, insisting their domain is too complex for agents, arguing the costs don’t justify the results, convinced this is a fad that doesn’t apply to them. And then the biggest group: the ones stuck in the middle, running pilots that never graduate, forming committees to evaluate tools their competitors shipped with six months ago.
This article feels sloppy, but I found the water pressure analogy interesting, nonetheless.
The area of Dex’s focus is “lights off software factories”, in which you don’t ever look at any of the code. I think that’s a fair focus, since that’s the ostensible goal of leading tech orgs. There’s lots of gems along the way, too, like this point that has big implications for OpenCode:
Claude Code won because it was better, and that it was better because Anthropic RL’d the model inside the harness – the first time a lab trained a model against the exact tools they were going to ship it with. And it got really, really good at calling those tools in an agentic loop…
If you build a harness but you don’t own the weights and can’t RL the model inside it, you’ll always be at a disadvantage to a team that owns both.
Trifle - track counters, revenue, and hierarchical data across Ruby, Elixir, and Go
This source available time-series metrics store looks really well done. It includes the stats libraries, a CLI for terminal and agent use, and a desktop app for all the visualization stuff. Free if you self-host it.
The best prioritization is no prioritization
The Stay SaaSy folks on why prioritization (for startups) is a trap.
It wastes time, wastes effort, and delivers worse results than just executing faster. Instead of spending your time prioritizing:
- Spend more time focusing on building faster (or literally just spend the time building)
- Shift everyone into durable teams that don’t have to do cross-business prioritization at all, and manage your “prioritization” by the resourcing decisions of growing, downsizing, or splitting teams
There it is. A smattering of things I found interesting lately. What’d you find?
