Will immersive sports amount to anything? I don’t think anyone knows, but this latest experiment suggests that the people producing Apple’s immersive sports are actually learning and experimenting. I’m encouraged, and I’ll tune in next week for sure.

Immersive sports feels like one of the killer use cases Vision Pro has been waiting for, and I’m rooting for it to actually go somewhere. Cosm is already proving the appetite exists, just with a domed screen instead of a headset strapped to your face. Superfans will pay real money to feel closer to the game.

None of that gets me closer to buying a Vision Pro, though. I don’t have a use case that justifies the price tag yet. But the day the NFL or EPL figures out how to put me in a virtual seat for the Bucs or Liverpool, that’s the day they get my money.

We’ll get back to normally scheduled programming soon.

Behavioral evaluation becomes practical at scale: Thanks to AI, the previously high cost of retrieving and analyzing the pertinent evidence has collapsed. AI systems could analyze patterns of employee interaction to surface evidence of higher-order competencies that traditional metrics miss entirely. They could examine communication networks to identify employees who consistently help others solve problems, analyze decision-making patterns in project discussions to spot strategic thinking, or map influence flows across email threads and meeting transcripts to reveal leadership in action.

Earlier this year, we built a rudimentary fact-gathering bot for managers. Its job was simple: pull together evidence of our team operating at their given skill level, without making any judgments, just finding concrete examples of when someone did a certain thing and linking to that data. For each area of the skills map we use for career progression, it would build out a list of evidence it could find, with a link to the relevant PR, conversation, document, or whatever else. It saved us a ton of time and led to a much more equitable set of reviews.

Assuming you have things wired up in a way that’s AI friendly at your org, this kind of tool can make the review writing process a lot easier and fairer. I even took the evidence documents to my individual team members and asked them to validate how accurate it was, and with a few exceptions, people thought it was pretty amazing.

Lots of caveats here, but as long as you’re accounting for the tooling’s blind spots and doing your best to bust the bias that gen AI tools always seem to have, this can be a powerful addition to your tool belt, one that saves you a ton of time.

There’s also a noticeable sharpness to the icons, along with a flattening of the Liquid Glass effect. I’m not sure yet whether this is simply an early-beta artifact or the intended final look. For example, while I really like the redesigned Finder icon, the sharp black edges around the nose currently feel a little unrefined.

Thank goodness – I felt like I was going blind looking at Tahoe icons on the Mac. The glassy, blurry nature of the style seems to be on the way out.

So well done to the Webkit team for Safari 27. And shame on the Chrome team for Chrome 148.

It’s nice to see the “Snow Leopard” year has extended beyond just the OS platforms and over to WebKit as well. Jeremy Keith gives a brief rundown of some of the features he’s most excited about and catalogs a number of the 500+ bugfixes coming our way as well.

SignalTrace “links devices that regularly travel together, correlating them to license plate.” It is a surveillance product that will sweep up and add all sorts of Bluetooth and other data to license plate readers, linking specific devices—and people—to cars.

We already know about the pervasive spread of Flock Safety and other Automatic License Plate Recognition (ALPR) cameras. But Motorola’s Vigilant Solutions is actively pitching a handy upgrade for any budding authoritarians: adding Bluetooth and RFID sensors to these street cameras.

This means they aren’t just logging your car anymore. They want to sniff out the unique MAC addresses of your phone, your smartwatch, and your AirPods as you pass by, tying your digital footprint directly to your vehicle’s location.

The kicker? The pitch documents explicitly mention tracking RFID pet microchips. They are literally turning neighborhood security cameras into a giant, unified surveillance database that logs everything from your daily tech, to your commute, to your daily dog walk. It is a total privacy nightmare.

I always enjoy Michael Tsai’s roundups of WWDC wishlists as well as the reaction posts afterwards as well. I’ve found a handful of new voices to follow as a result.

Most of us have been that manager at some point. The schedule fills up, the weeks blur together, and the 1:1 becomes one more thing happening to you rather than something you’re actually driving.

The fix isn’t complicated, but it does require a little intentionality. A few years ago I started treating my 1:1s less like a recurring meeting and more like something worth preparing for. Once I figured out what the pain points for me were and I came up with solutions for them, I have felt like I show up as a much better version of myself in those interactions.

Monday is for prep, not meetings

The first change I made was protecting Monday mornings for review rather than conversation. Before I talk to anyone, I go back through the previous week. These days that’s easier than it used to be. I use AI tools to pull meeting transcripts and generate detailed notes from my 1:1s, complete with action items and themes. What used to take real effort now takes minutes.

But the synthesis work is still on me. Once I have the notes, I look for patterns across the whole team. What topics kept coming up? What’s everyone anxious about? What’s getting energy, and what’s getting quietly avoided? Usually a few themes surface: a project with murky ownership, an announcement that landed weird, a technical decision that’s quietly worrying people.

From those themes, I draft a few questions I want to bring into the week’s conversations. And I think about those questions at different levels. A staff engineer is going to have a fundamentally different perspective on “why is this project behind?” than someone in their first developer role. Both perspectives matter. But they’re not the same conversation, and you shouldn’t expect them to be.

Structure your days by altitude

Here’s the other thing I changed, and it’s the one that made the biggest practical difference: I stopped scattering my 1:1s randomly across the week and started grouping them by seniority level.

Each day, I try to meet with people at roughly the same level. I start the week with my most senior folks and work down from there, ending the week closer to the ground.

I call the problem this solves “altitude sickness.” It’s what happens when you spend 45 minutes thinking about system architecture and technical strategy, then immediately pivot to a completely different kind of conversation about someone’s growth goals or the basics of a new codebase. Neither conversation is less valuable or worthwhile. But jumping between them constantly is a form of context switching that’s more expensive than you think. You’re not just switching topics, you’re switching the entire register of how you think.

Your mileage may vary on the direction. I go senior to junior, but there’s a reasonable argument for the inverse. Start tactical and build toward strategic, or vice versa. What matters less than the specific order is that the order is intentional. Pick one and stick to it.

One small thing that makes all of this easier

If you run 30 minute 1:1s, don’t stack them back to back. A 10am, a 10:45, and an 11:30 beats a 10am, 10:30, and 11am in almost every way. Fifteen minutes between conversations gives you room to jot a quick note, reset your head, and actually show up present for the next person. It also makes it harder for other meetings to colonize your calendar, which is a side benefit worth appreciating.

The point isn’t the system

It’s pretty clear that none of this is revolutionary. Prepare for meetings, group similar things together, give yourself transition time. But taken together, it changes something real about how you show up. Instead of your calendar running your week, you start to actually show up as the best version of yourself. Your team can tell when you know what’s going on with them. They can tell when you remember what they told you two weeks ago. That kind of attention is the actual job, and a little structure makes it a lot more achievable.

Today, Last.fm begins a new chapter as an independent company. Ownership has changed,
but the product you use every day has not.

Your account, your listening history, and your data remain exactly where they are. The team
building Last.fm is the same. The service continues as normal.

What stays the same

Your account and scrobbles
Your data and privacy settings
Your Pro subscription and billing (if you're a member)
The people behind the product
We'll share more about what comes next in the weeks ahead. For now, everything works
exactly as it did yesterday.

Same service. Same team. New chapter. Thanks for being part of it.

I’m a big fan of Last.fm, and less of a fan of their slimy parent company that bought them 15 or so years ago. Glad to see they are breaking free, and hopeful they can stay up and running. The thing I struggle with a bit is that the $4.99/month plan is a bit more than I need, but I do want to support. Maybe they’ll introduce more tiers now that they’re independent.

There are more than a hundred new feature additions, changes, and refinements, including forward-looking foundational work which brings order-of-magnitude performance improvements in some parts of the product.

Headlines are shortcuts support, search improvements, more AI integration, better git support, and tons of quality of life improvements. I tend to bounce between BBEdit and VS Code these days, but I’ll always have a soft spot for BBEdit as it’s been a mainstay since I started using Macs many eons ago.

Full release notes can be found here.