Zingers and tip jars
Margin notes on blogs, great ideas that arrive after hitting publish, and feedback.
Zingers and tip jars
I’m taking a break from longer essays this week. Writing longer pieces is both challenging and rewarding, a nice change of pace. I hope you’re liking them! Suggestions and feedback are appreciated.
1. Tip jar
Another point of order: I’m experimenting with a “tip-jar” on this newsletter. The software I’m using, Buttondown, will add minimal “upgrade your subscription” call-to-actions. Rest assured, they won’t be modal screen take-overs, like some other sites. These are entirely optional, particularly if you’re already subscribed.
My goal is to build the muscle of exchanging goods and services for money. If I’m going to start a business-shaped thing, might as well start with something I’m proud of. Namely, this newsletter.
But, I would rather not reduce access to my writing. I don’t want a paywall. A pay-what-you-want tip-jar sort of thing seems like the right place to start.
If you enjoy my writing enough to send encouragement in the form of monthly cash, that’s great. Even if not, your readership is appreciated!
2. Shorter ideas of late
Text boxes were not meant for collaboration. We find ourselves interacting with human-like intelligences via text boxes lately. This in addition to short (chat) and long (docs, issues, etc.) interactions with our teammates. Granted, some pretty great things have been created this way. Yet, we should ask: what are we leaving on the table by collaborating this way? What might we achieve if we had collaboration tools as deep as (for example) musical notation?
Calendars display the predictability in our world, but it’s the unpredictability that is often more stressful. Surprises, interruptions, questions aren’t scheduled. Arguably, they shouldn’t. The best work frequently happens when the right people work together at the right time on the right thing. You can’t schedule that. The second-best work happens when new information is acquired because someone was surprised that the world doesn’t reflect their mental model. That said, most recurring meetings should happen inside a textbox instead. 🙃
Process normalizes output at the expense of peaks (and valleys). Strike out less, hit slightly more singles and doubles, but hit fewer home-runs too. As goes writing, so goes software development.
When it comes to endless feeds, let the computer check off the old stuff. I spent last week trying to convince my mom, a not-organized person, that she shouldn’t aspire to organize her email and files, and use search instead. Similarly, I can’t read the entire internet, so I should let a computer click unread next to stuff as it ages out. 🤷🏻♂️
Let’s not reduce intelligence, emotion, art, or the gamut of human intelligence down to the terms by which we try to represent them as billions of floating-point numbers. Related: human-like intelligence won’t eat software estimates.
The web (and blogs) are great because they can get weird. 🤩
3. Director’s commentary
Occasionally, I come up with a zinger after this newsletter goes out. Luckily, I have a blog that is perfect for after-the-fact zingers. I consider it to be the director’s commentary from DVD extras, for the newsletter.
Saturday Night Live happens, in part, thanks to two deadline hacks.
Claude doesn’t like it when I call its work “slop”. These days are peculiar for humans and floating-point numbers.
LLMs won’t eat software estimates, unfortunately. Wherein I ask myself, “do I like estimating projects now?”, and come up with “not entirely, but I have ideas for how to make them worth the time.”
Thanks for reading!