Ways of Looking #11 (July '25)
In this issue: Useful Ways of Looking, context engineering, Trying to Smile, "Just Call My Cell Anytime," wool shirts, and Tinker Creek.
Welcome to Ways of Looking. Roughly every month, I send my friends links and summaries for anything I’ve written recently, plus a few things I’m enjoying.
Check out my more regular writing and reply to this email anytime. Thanks for reading.
What I wrote last month
Introductory field guide to Context Engineering for LLM users — it’s not just “prompts” that matter for good AI results — context is king. LLM interfaces inject a lot of context themselves, and you can augment it. If you do this well, you’ll get better answers!
Useful Ways of Looking — I started writing a series on concepts / mental models / “ways of looking” I find myself referencing frequently. When I’m facing a challenge (or helping a friend work through one), these are the explanatory tools I commonly reach for. The first five:
OODA loops — Loop through Observing, Orienting, Deciding, and Acting — with a special emphasis on Orientation, and on cycling back to it as quickly as possible; it works for fighter pilots and can work for you, too.
Theory of Constraints — To solve any problem: 1) properly define your goal, 2) clearly outline the system that will dictate the achievement of that goal, 3) determine the bottleneck of that system, and 4) address the bottleneck — and only the bottleneck.
Parkinson's Law, generalized — Work expands to the time allotted for its completion, and elsewhere constraints generally predict outcomes as well; the intelligent definition of constraints is often more important than the energy put into execution.
Ways of Looking — 'The most fundamental fact about any experience is that it depends on the way of looking' (Rob Burbea); by leveraging this, we can make any 'problem' we're facing trivially soluble.
Useful vs. Correct — The usefulness of any idea is not necessarily correlated with how correct it is; it is only within a specified context that attributes like usefulness and correctness of the idea can be evaluated.
A pretty gradient generator — exactly what it says on the tin. Configurable gradient generator I used to create the interface0 “pulse.”
interface0: early July update — more features! (And then many more since I published this…)
AI Captain's Log — usual not-quite-weekly updates on my usage of AI.
Jekyll category filtering with no JS — low-TAM blog post. If the title doesn’t make sense, it’s probably not for you…
What I’m enjoying
Article: "Smile" vs. "Poop!" by Guy/Rivalvoices. On why “Trying” to achieve a direct outcome (“smile!” “get healthy!” “be happy!” “make more money!”) often fails — because the thing you are acting out is the “Trying” rather than the “Doing” — and thus why indirect paths are often more effective. The piece is really, really good. May need a couple read-throughs (and is deserving of such!).
Experiment: Just Call My Cell Anytime. For 95%+ of inbound scheduling requests I want to take, I now just give the other person my cell number and tell them to call me anytime. It is great. Calls now last the appropriate amount of time (often 3-10 minutes) instead of a default 30- or 60-minute block. And perhaps 50% of people end up never calling (even better!). I usually pick up, but if I can’t, I rip through the backlog of missed calls while I’m out for a walk. This has been a huge quality-of-life improvement.
A friend sent me this video from Jason Fried which exactly represents my attitude toward the whole thing. Thanks Alex.
Product: I’ve finally abandoned non-natural materials in my workout gear and am loving Ryker and Wayve cotton/wool shorts. For shirts, I’m doing merino wool from Quince.
Yes, most “100%” merino wool is actually “superwashed” with a chemical treatment and then coated with a polymer layer, but I don’t want to pay $100+ for an untreated/plasma- or enzyme-treated shirt I’m going to destroy with sweat in a month. If you have any other ideas, please let me know!
Book: Pilgrim at Tinker Creek by Annie Dillard. Pulitzer-winning narrative nonfiction in the vein of Walden. I didn’t find the philosophical thread particularly compelling, but the writing and depth of observation is stunning. Worth picking it up and reading a chapter or two whenever you want a dose of inspiration on those fronts.
About me: I’m a multi-time founder (Eco, Lightwork Home Health, CoinList, Sidewire). I’m currently spending time on: Eco, Lightwork, building interface0, investing at Amity, supporting FreeWorld, hacking on a few other products, and advising great companies and founders.
Thanks for reading. I’ll see you next time.
— Andy