A weekly list of interesting things I found on the internet, posted on Sundays. Sometimes themed, often not.

---

1️⃣ I’m mesmerized by this dancing magnetic liquid. [🔗 @sciencefunn // instagram.com]

2️⃣ Here’s an animation in macOS that I didn’t know existed, but that I love. [🔗 BasicAppleGuy // mastodon.social]

3️⃣ Lex Friedman has some good tips on politely saying “no”, and protecting your time. [🔗 Lex Friedman // lexfriedman.substack.com]

4️⃣ This iOverlander app helps you to find camping and overnight spots filtered by free and paid, amenities, and and whatnot. [🔗 apps.apple.com]

5️⃣ Shoutouts.lol by Vincent Ritter is awesome. It’s an easy way to build a database of sorts of things that you want to highlight. And then with a simple embedded script, it presents one at random. It’s what I use to show off the ‘Stuff I Love’ at the bottom of my site. And only $10 a year! [🔗 Vincent Ritter // shoutouts.lol]

6️⃣ Pepsi’s got a new/old logo. I like it! I think. [🔗 pepsico.com]

7️⃣ Where I’m headed right now for a spur of the moment climbing day. 🧗‍♂️ [📍 mountainproject.com]

---

Take a Chance

---

Thanks for reading 7 Things. If you enjoyed these links or have something neat to share, please let me know.

7 Things

I can’t get over how well done the feature suggestion and bug reporting flow is in Readwise Reader. Take a look:

Too many systems require endless dropdown menus, they need step-by-step instructions to reproduce the problem, and end up feeling like way more trouble than their worth. I’m looking directly at you, Apple’s Feedback Assistant.

Reader’s open-ended text box allows me to write out my thoughts plainly, as if I’m just sending a quick message over to the app’s developers. It feels informal, and I never get frustrated trying to send along my thoughts. Which is good for the folks on the receiving end because by the time I’m done filling out most feedback forms, I’m more pissed off than I was with the initial problem and I’m sure that comes across in my feedback.

Reader earns bonus points for the prompt, well, feedback on any report sent over in the way of an email confirmation. I’ve also gotten direct messages in their Discord regarding my bug reports and feature requests.

In the end, it’s the results from feature suggestions and bug reports that really matters, but Reader’s no-fuss system tells me that they really do want to receive a bunch of feedback so that they can improve their product. Most app developers could learn a lot by taking a leaf out of the Reader book.

Apps

Joseph White, writing for Reuters:

General Motors plans to phase out widely-used Apple CarPlay and Android Auto technologies that allow drivers to bypass a vehicle’s infotainment systems, shifting instead to built-in infotainment systems developed with Google for future electric vehicles.

This seems like a very short-sighted move on GM’s part. I wonder how badly their arm was twisted by Google as their development partner.

I’d be astounded if GM doesn’t reverse course within, let’s say, two years. Let’s not forget Apple’s claim when they introduced the upcoming next-generation of CarPlay (timestamp of Apple’s YouTube video | The Verge write-up) last year:

I’m no transportation market analyst, but I have to assume that a big segment of folks interested in electric vehicles use iPhones.

Linked

Manton Reece on an exciting new feature addition to their podcasting platform:

We’ve launched a new feature for Micro.blog Premium customers: automatic podcast episode transcripts, powered by OpenAI’s Whisper model. I’m excited about this because it’s one of the more practical, time-saving solutions coming out of the rise of AI. The automatic transcripts are so accurate they can be used as-is, or edited by hand as you have time.

[…]

The new transcripts feature is available to anyone hosting their podcast on Micro.blog. When you upload a new MP3, Micro.blog will process it to generate a transcript. You can then edit the transcript or link it from your podcast page.

What a great, practical use of Whisper. I love that it’s automatic, and can be linked for listeners who want to follow along. Anything that helps make the internet a more accessible place for us all gets a thumbs up from me.

Manton continues with icing on the cake:

There’s no extra charge for any of this. Micro.blog Premium has always been $10/month and it will continue to be priced that way. It includes podcast hosting and also email newsletters, bookmark archiving, web page highlights, and much more. We think it’s a great value.

If I were to start a podcast today, I would absolutely use Micro.blog’s infrastructure. It’s where I’m already doing my microblogging, the setup looks dead simple, and I listen to at least one high-profile podcast that uses it.

Linked

Jason Fried on his HEY World blog:

First it starts as an outlier. Some behavior you don’t love, but tolerate. Then someone else follows suit, but either you miss it or you let it slide. Then people pile on — repeating what they’ve seen because no one stepped in to course correct.

Then it’s too late. It’s become the culture. The new normal.

[…]

Whatever you do becomes how it’s done. And the higher up it’s done, the more influence it loads on the structure below.

It’s so hard to speak up. But letting it slide is, as they say, just silent agreement. It’s something that, I’m embarrassed to admit, I’ve struggled with my whole life and it’s been top of mind lately.

Linked

Are you ready for the next mind-blowing frontier for these “AI Chatbots” that are taking the world by storm? David Pierce shared an example on The Verge of what is possible when GPT meets your computer lifestream:

Dan Siroker, the CEO of Rewind.ai, gave me a demo of the new feature ahead of its launch. He and I had never met before, but we’d emailed a couple of times, and he’d looked me up ahead of our meeting. So he opened up the ChatGPT For Me window, a separate chat window inside the Rewind app, and typed “how do I know David Pierce?” A few seconds later, it spit back an answer: we had a recent interaction after I reached out to introduce myself. (True and true.) We scheduled a 30-minute Zoom on March 22nd, 2023 (true), and discussed a big launch happening at Rewind (true). It also linked to the calendar event for our meeting, my LinkedIn page from his browser history, and more. With 10 seconds and one paragraph, Rewind detailed our entire relationship.

A personalized large language model-powered assistant is a very exciting possibility. But, done wrong, could have disastrous consequences. Setting aside the potential privacy concerns of letting an app record everything we do on our computers — difficult to set asides, I know — also allowing it to slurp up all that data to train a model that’s a big black box on the web would be nerve wracking, to say the least.

But with things like Rewind’s ChatGPT for Me, I can see glimpses of a future where we can ask natural language questions about our personal lives and get genuinely useful responses back. It’d be the infinitely accessible, summarizable, and interactive super memory that so many of us long for. A truly personal assistant.

I’m not going to say that Apple is the only company I’d trust with this kind of access to my data. But it’s a short list, and I am curious and optimistic to see if something like ChatGPT for Me emerges from them down the road.

Linked

A weekly list of interesting things I found on the internet, posted on Sundays. Sometimes themed, often not.

---

1️⃣ Technology is getting bananas these days. Here’s a recreation of Steve Jobs’ voice responding to novel questions, in his style, through ChatGPT. 🤯 [🔗 @BEASTMODE // twitter.com] (Via John Gruber)

2️⃣ This site/Chrome extension/shortcut makes it easy to share a full ChatGPT conversation. Rather than trying to capture a huge screenshot, just share the URL it creates instead. [🔗 sharegpt.com]

3️⃣ I used to have a general disdain for plug-ins (a general mistrust in things they weren’t natively built in) but I left that feeling behind long ago. Plug-ins for web services, launcher apps, and now ChatGPT massively expand and inspire new capabilities for these tools. Seeing as ChatGPT can write code and interact with existing APIs, I think this is going to be a boon beyond our imaginations. [🔗 Mitchell Clark & James Vincent // theverge.com]

4️⃣ This LighterPack site helps you track what you bring when backpacking and how much each item weighs. It’s apparently popular with thru-hikers who tackle extreme distances and where pack weight is a key concern. [🔗 lighterpack.com] (Via Mitchell Clark // theverge.com)

5️⃣ “Underscore” David Smith shares his appreciation for the quietly consistent health data that the Apple Watch collects for you. Even when fitness isn’t your focus, having that historical data is super helpful when looking at trend lines, and, in David’s case, helps reinforce that small changes to habits do make a difference over time. [🔗 David Smith // david-smith.org]

6️⃣ It’s so easy to anthropomorphize cute objects and I did so hard with this video of a slinky on a treadmill set to dramatic music. I can’t help but root for the little guy! [▶️ abzde // youtube.com] (Via Jason Kottke)

7️⃣ I loved this collection of deliberately inconvenient everyday objects, including the “uncomfortable watering can” and “engagement mugs”. Usually, we’re thinking about how to optimize, but in this case architect Katerina Kamprani got to stretch her creativity in the other direction. [🔗 Katerina Kamprani // theuncomfortable.com]

---

Take a Chance

---

Thanks for reading 7 Things. If you enjoyed these links or have something neat to share, please let me know.

7 Things

Yours truly, back in June of 2021, writing about Apple being tight-lipped on which features in macOS Monterey would not be coming to Intel-powered Macs.:

I actually think this was a calculated decision. Apple is rumored to still have more Intel Macs to introduce. Understandably, they wouldn’t want to draw additional attention to OS features that those brand-new Macs won’t be able to use.

Thinking back through the timeline all the way to WWDC 2020, I don’t think this rumor actually panned out. As far as I can tell, not a single Intel Mac has been updated or introduced since the advent of the M1 chip. That rumored Mac Pro update that I linked to didn’t pan out, and the Mac Pro you can still buy today runs the same generation of Intel Xeon W processors.

Apple waited until they had a winning hand with their own silicon, then laid their cards and walked away from the poker table.

My take back then was wrong, but they did (and do!) continue to have an Intel machine for sale. I still tend to think Apple’s comparisons in marketing will get more aggressive once they complete their divorce from Intel.

Linked

Adam Mastroianni, writing for their Substack newsletter, Experimental History:

It turns out that we like people the best when they respond to us the fastest––so fast (mere milliseconds!) that they must be formulating their reply long before we finish our turn.

Uh-oh. I must not be very well-liked at all in conversations because I’m typically a slow thinker. I like to wait until someone gets it all out and then give a measured response. Something to consider, I suppose.

Linked

Srini Kadamati, a data visualizer, waxing eloquent on the virtues of the dearly departed Dark Sky app’s design:

But Dark Sky was much more than just an API or a set of “forecast technologies.” The design of the Dark Sky mobile application represented a hallmark of information design because the team clearly obsessed over how people would actually use the app on a daily basis.

The design of Dark Sky was so wonderful that I could understand the shape of the weather at a glance, even from a zoomed out view of the app.

On the innovative temperature pills that made understanding the gist of the forecast possible at a glance:

In the Dark Sky app, the “temperature pills” representing the forecasted temperatures for the upcoming week preserve their existing magnitude more effectively in the visualization. The temperature values are more tightly integrated with the visual representation, making the combined experience more amenable to quick comparison across multiple days.

Srini concludes with a plea for more contextual data experiences that draw on principles that guided the Dark Sky app:

Dark Sky started with publicly available data, augmented it with contextualized predictions, rigorously iterated on data visualization design, and packaged all of this into a contextualized experience to make weather data useful for me in my daily life.

While the availability of data has never been higher, we’re still missing software experiences that contextualize that data to make our lives better. Data alone isn’t enough.

Make sure you check out his article for the accompanying screenshots.

If you’re a recovering Dark Sky user like me, consider reading this post as “pouring one out” for the Dark Sky app. And to ease the pain, I’ll point you toward CARROT Weather, which introduced a “Dark Sky” layout as a way to welcome users like us. It faithfully recreates the day and week forecast layout that we relied on — with the artful, contextual shape of the weather — which Srini rightfully calls a “data visualization masterpiece”.

Linked