Inspired by Kottke’s recent Emptying the Tabs, here’s my own dump from my own digital hoarding:

Understanding the Apple ][. Because, y’know… I might need to reference this later.

Douglas Adams on the English–American cultural divide over “heroes”:

You cannot make jokes about failure in the States. It’s like cancer, it just isn’t funny at any level. In England, though, for some reason it’s the thing we love most. So Arthur may not seem like much of a hero to Americans – he doesn’t have any stock options, he doesn’t have anything to exchange high fives about round the water-cooler. But to the English, he is a hero. Terrible things happen to him, he complains about it a bit quite articulately, so we can really feel it along with him - then calms down and has a cup of tea. My kind of guy!

Ah, good ol’ Arthur Dent. He was my kind of hero when I was growing up. Also, I’d love to give Blackbooks a good re-watch sometime.

Japan’s Dodo Land. A place to be annoyed, for those of us who love to be annoyed.

How Money Works. It leaves your wallet? What else do you need to know?

Inner Drumming. I’ve yet to have a chance to learn drumming, but this sounds like my kind of learning.

Why the open social web matters now. I mean, the Internet seems to be in decay, so maybe this posting on my blog is all for naught?

Adaptive Mixtures of Local Experts. Some light reading for when you want to dive in to how a Mixture of Experts (MoE) LLM model works, straight from the source. Been wrapping my head around MoE’s vs dense models.

Why did AMD just buy this REALLY WEIRD chip company? YouTube video discussing the acquisition of Taalas by AMD. Smells like positioning for local AI, and has hints of what folks were doing for making ASICs for crypto mining.

Why your local LLM feels dumber than it is.

Remote for OpenCode. Looks promising, and fancy way to handshake via iCloud, but doesn’t seem like it works yet outside of the local network. Maybe Tailscale is the right way forward?

Aging Brains Blend Memories Together Instead of Just Forgetting Them, Study Finds. Our context is just filled up, alright. Need to compact.

There’s more than this, but that’s it for now. I can least feel like I can close these tabs and come back to this list if I need to come back to them. Thank you for your attention to this matter.

Now it’s faster and easier to add a new plan on Ready Time. With Ready Time 3.6, you can now select from 46 pre-defined routines to start your plan, including:

• Morning routine. Getting up can be tricky for some of us, and having a regular routine definitely helps out.
• Doctor appointment. You scheduled it, so you gotta make sure you get out the door and to the doctor’s office in time.
• Concert night. Don’t wanna miss the headliner!

And now, for folks who are discovering Ready Time from the web, if you visit a routine page from Safari on iOS, you can start an App Clip right from the page to quickly start planning for a deadline without even needing to install the full app! Just open a routine page like Getting out of bed and open the App Clip right from Safari.

The App Clip lets you get notifications for each step in your plan, just like the full app. And when you’re ready to create more plans or customize your plans more, you can quickly download the full Ready Time app.

We hope these updates will make using and trying out Ready Time faster and easier.

Let me know if you have any feedback. I love to hear from you.

“It’s parenting, but somewhere else”

– overheard

Packing List

Some things I brought along for the con:

• Flipper Zero. Never used it.
• Baofeng UV-5R. I thought I’d charged it before the con but I didn’t and so couldn’t use it, gah!
• Zojirushi Thermos. Came in clutch for having hot(-ish) coffee throughout the day, especially when I was in Ham In A Day.
• RFID protected wallet. Learned from the previous year not to let my hotel key sit naked in my bag (I wasted way too much time waiting for the hotel to help me back into my room).

Things Learned

• I got my Technician Ham License! This was one of my childhood dreams until I was stifled by the Morse Code part of the test. But thanks to the Ham In A Day class, I’m now KO6OUJ! Thank you, Ham Radio Village!
• Fox Hunting. Even though I wasn’t able to use my Baofeng that I brought (boo!), the generous Cat Dragon from the Ham Radio Village let mikwat and I borrow a spare radio to try hunting foxes. We had no idea what we were doing, but with some perseverance and trial and error, we learned some techniques:
  • body shielding
  • detuning
  • tightening the squelch Even though we ultimately didn’t find a fox, we learned a lot along the way, met various random folks in the journey, and just had loads of fun!
• Meshtastic. I’d heard chatter about this online, but didn’t have a real sense of how many people were using it. I met so many folks who had radios at the con, went to talks about the technology, and then went and got myself a radio. It was a great place to get started, getting onto mesh, and listening in on the primary channel. Been playing more with Meshtastic and learning more since I’ve been back home.
• Bruce Schneier. What a treat to attend a talk by the legend himself! I read his blog, so it was amazing to be able to see him in person. Inspiring.
• Silent disco headphones. I thought the use of headphones were a clever way to pack more folks into the large hall without the problems with audio leaking between talks. In previous years, it was pretty common for me to have difficulty hearing speakers, so I really liked this. It was rather controversial though, as a lot of attendees, I imagine, are likely germaphobes.

Other Random Tidbits

• “Less luck. More skill. Less Claude.”
• I’ve never had a real handle for myself at the con, but this year I came up with one I like: fourants and I “hacked” my badge to have it show my new handle.
• First Friday at the Arts District at 10PM, 110 degrees, tons of kids hanging out on a summer night!
• We never got a chance to try The Loop, every time we tried to book a car it failed. We’d get an error after putting in payment details rather than up front before the booking form: annoying!
• Spent some time in the arcade, where I got my butt beat in Marvel Super Heroes. I had a bit of better luck at protecting the planet in Galaga instead.
• My flight from Harry Reid got delayed…then canceled. Ended up spending an extra night, though I was too pooped by the end to try to drop in on the closing night’s parties.

We’ve been getting books from Dolly, and I have to say, they are such an incredible gift and delight to receive and read. The stories from the books we’ve gotten are now a part of our family’s lore.

Thank you, Dolly Parton, or, as the little one says in the cutest way: Dolly Person.

Sending Hearts To All My Dearies. A tribute album to Smashing Pumpkins.

If you grew up in the 90’s listening to alternative, you’ll notice every track is one you’ve heard before.

Some re-interpretations, like, Tonight, Tonight by The Midnight, feel a little lackluster to me. But maybe a more mellow version can sometimes be called for, given the right occasion, like, maybe a midnight drive in a red Ferrari Testarossa. I do find it funny that there are 80’s synthwave remakes of 90’s songs.

Good friend Gage has been toying with a new idea: Particles:

Particles is shared memory for humans and AI agents. Each particle is one claim, plus what you need to judge it: who said it, where, when, and how confident they were. Facts, opinions, and memories are all claims, recorded the same way as particles. Particles are not edited or deleted. Particles are superseded, retracted, or disputed in the open. How much to trust it is a perspective applied at query time, never baked into the record.

In the current post-truth world, this an interesting idea: the facts we have have a relative truth based on your own viewpoint and perspective. If we give an agent knowledge/ideas/data, it doesn’t really know what truth/facts are without us telling or asserting its validity or it looking it up from other sources. The more personal our agents become for our own uses, it seems like we’ll want to ensure our agents know our own worldviews and the weights of our claims compared to the information it finds from other sources.

Maybe I’m late to the game, but seems like there’s so much thought right now going around on how to manage the context going into models at inference. Context engineering. Personal RAG.

I was working on fixing a bug in Lilmem 2.0 where switching from the diary’s Calendar mode to Memories mode caused a bug where the diary’s content “stuck” in a specific day. When I’d try to swipe between days after switching modes, the same day would re-appear after swiping.

I asked Claude to help me take a look and consider a fix. It came back with its usual vomit of text, and it left me rather confused. Rather than press it to give me even more text, I asked it to give me a visual explanation for what it was trying to do instead.

Here it’s explaining first the layout of the view so that we knew we were on the same page (ha!):

It outlined the various ways selectedPage changes can happen and its effects. And then it brings in a new concept of a “pinned” date key, which wasn’t clear to me in its wall of text:

And then it explains how and where the pinned date key is used:

After seeing Claude’s diagrams and explanation, I was able to get what it was talking about and start collaborating with it from there.

Super helpful getting these visual explanations. I was inspired by how I’ve seen Simon Willison has gotten Claude to make interactive explanations for him to better understand what the agent is talking about.

This view/interaction is a core part of the UX of Little Memory, and getting it right is critical to the experience of the app, so I had to make sure Claude wasn’t just doing something too wild. Having it show me (rather than just tell me) was incredibly useful to make sure I knew what it was trying to accomplish.

ChatGPT Is a Huge Snitch:

You’re led to believe you’re talking to a little guy in the computer; in fact, as with most computer and smartphone interactions these days, you’re also filing text into a searchable database. Add in the rise of AI-powered meeting transcription and social surveillance, and the range of materials generated in the course of a regular computer-job workday will soon be breathtaking.

Even when you know not to, it’s too easy to accidentally slip and send private data to your agents.

Claude Code has, on multiple occasions, read private keys that it wasn’t supposed to, and I’ve had to rotate them because of the exposure. And with features like GPT’s Computer History (horrible name BTW), it’s even easier to just dump more of your private data into the cloud. Frontier models have known to be relentless in the pursuit of their goals, and guardrails to tell them not to read your home dir or private files doesn’t seem like much use in blocking them.

I think the way forward is to either run only local models or setup a separate workstation just to run remote models/agents. Without some sort of isolation, the agents can just easily suck up all the data they see.

I’ve been reading iOS Dev Weekly for years now. Before this past May, the newsletter was written by Dave Verwer who I’d emailed a handful of times (he responded!) and always had great commentary about the state of the iOS development world. Sometimes, he’d be honest and note he didn’t have much to say that week, but he always brought interesting thoughts and ideas to the table.

Since May, the newsletter was handed over to Mobile Sessions who would take the helm and keep it going. I’ve been skimming the latest issues, as it helps to read about new frameworks or blog posts from the community to keep up with what’s happening on iOS. There’s a rotation of writers who have been taking turns on the newsletter for each week. It wasn’t Dave, but I still felt like I was connecting with people in the community.

The latest issue of iOS Dev Weekly, though, was just riddled with AI/LLM-isms that kept irking me:

A new set of App Store marketing assets arrives this autumn, and the work is design work rather than engineering, which means it probably isn’t on anyone’s board yet.

Ugh, this “and the work is…” just keeps showing up.

Most of us open Instruments, look at the call tree, and quietly close it again. The hierarchy is honest but it buries the answer.

And this “quietly” and “X is Y but Z” pattern as well.

And even in the And Finally…, which, IMO is an easy way to just bust something out by hand:

Somewhere a product manager shouted ship it, nobody asked a follow-up question, and CarPlay is now in a pontoon boat while some automakers are still workshopping their objections.

Oi. This “nobody asked X” pattern that AI loves so much and is so dramatic about.

Looking back at previous issues by Juan Marin, and Issue 759 was also by him. Still pretty AI/agentic.

I’m not against agents writing and helping to write (though, I know some folks are adamantly against any use of AI in writing), but IMO this is a real sloppy use of it. I read enough Claude-isms all day from my agents in Claude Code, and it’s already exhausting there. When I’m reading a newsletter like iOS Dev Weekly, I’m expecting more than just the first pass of what gets spit out. I think it’s lazy and really leaves a bad taste in my mind. A little editing of the original draft can easily clean these up and give the writer a bit more of their own voice and make it lot more interesting to read.

BTW: for my own blog here, I don’t use AI for writing/drafting, but I do have it help with proofreading/editing sometimes. For some of my projects, I do have it help out with marketing copy, but even there I steer it with my own voice guidelines and give a bit more personality/context than just its default generic style of prose.

I found a page with a cool example of how to draw brushed metal with CSS. I previously wanted something like this in Swift for simulating flashlights butttons like for the Noctigon KR4. Stashing this away for future projects.

Dave Jimison shares the idea of asking your LLM how you communicate with the prompt:

Review the way I’ve communicated with you across our past conversations and give me an honest, direct assessment of my communication style with AI. Tell me how I express dissatisfaction, how I tend to work through conflict or disappointment, how I give positive feedback, and what strengths, blind spots, or recurring patterns you notice in how I collaborate with AI.

I got a lengthy analysis from Claude Code. Here are some highlights:

How you express dissatisfaction

Blunt, factual, and almost never emotional. Across 1,345 messages there is no profanity, no sarcasm, and the strongest emotion on record is “ugh.” Your dissatisfaction has a consistent anatomy: name the mismatch, then supply the missing fact or the intended mental model.

I’m pretty sure I’ve written fewer than a handful of profane words to AI before. It certainly wasn’t when using Claude Code. It’s usually in response to something that was surprising. It doesn’t make sense to me to use profanity or sarcasm with an LLM. Doing so feels like anthropomorphising the agent.

The overall shape: you run AI like a staff engineer runs a team — high-trust delegation inside an explicitly constrained envelope, mandatory personal review at the reputational boundary, adversarial review as a habit, and a deliberate teach-the-agent loop that turns every correction into a durable rule. The through-line of both your strengths and your blind spots is the same trait: you optimize for keystroke economy and trust the machine to hold state. When the state-holding works, you’re an unusually effective collaborator; when it doesn’t, that’s where every cycle you lose comes from.

Interesting learnings, and nothing surprising for me here. I think I’m pretty deliberate in thinking like an engineer when using AI agents. I understand the harness is keeping memory, and I use it to make sure I’m not repeating myself (too often). But I can see that I can sometimes lean too much on the agent’s memory and state, so it’s important for me to know (or more likely intuit and feel) the limits of the agent.

Unauthorized Wi-Fi network found on Delta flight after DEF CON:

Delta Flight 591 was traveling from Las Vegas to Atlanta on Aug. 10 when an unauthorized Wi-Fi network was detected aboard the Boeing 757. […] Messages attributed to the flight crew indicate there were concerns that someone aboard the plane had created a network called “Delta WiFi Fast.”

Sounds like someone was playing with their new Pineapple after buying it at DEFCON.