Now that trains are running again, it got me wondering whether my temporary commute, where I drove to a nearby station, is worth keeping. I don’t think it is. It got me into work earlier, which is nice, but with all the road traffic, getting home was slower and more of a hassle.

Seeing sponsored features in Vivaldi is… well, it is what it is: they need to make their money somehow. And atleast they’re easy to remove. But I’m left wondering if Netscape’s attempted approach of charging people for a browser was the better way to go in the long run.

Still wishing for MR review software that allows one to make private notes. Some of the reviews I’m looking at today are quite large, and it would be good for me to annotate things I’ve seen. I could use notes or something, but UI for marking up the code is already there.

I’m finding, as I go through my Godot journey, that a good indication that a particular approach to a problem is the one preferred by the designers, is the amount of “infrastructure” that goes into making that approach easy to execute.

Case in point, I ran into an issue in my Godot game yesterday, where a nested scene, one included in another nested scene, was unable to get a unique node near the root of the tree. After a bit of reading I learnt about groups, which look to be a way of collecting nodes, regardless of where they are in the tree, in a group that is easy to reference later.

This indeed solved my problem, but it was’t my first approach. There were ways of traversing the tree to find the node in question that I explored first. And I probably could’ve gotten that to work with a bit less-than-great of code. But given that there are 10 methods that deal with groups on the SceneTree class, along with UI in the editor to create and manage groups, it’s clear that groups was what the developers intended as the right approach.

Frustrating to be facing a task at work where my answers to when it will be done look like:

I don’t know.
I don’t know.
I don’t know.
I don’t know.
I don’t know.
Ok, it’s done.

“Vibe coding” I can take or leave, but this poster’s point on iOS distribution is spot on:

I recently built a small iOS app for myself. I can install it on my phone directly from Xcode but it expires after seven days because I’m using a free Apple Developer account. I’m not trying to avoid paying Apple, but there’s enough friction involved in switching to a paid account that I simply haven’t been bothered.

I had a passing interest at looking at the iPhone SDK when it came out. What steered me towards learning the Android SDK was learning that any app I built for the iPhone had to distributed via the App Store¹. That wouldn’t work for apps that I built for myself, or maybe a few friends, which is pretty much what I was interested in doing. Such apps are a completely different kettle of fish than building something meant for the whole world to use.

I don’t know what it’s like now, but despite how locked down Android is now, compared to those early years, it’s still seems to me to be much more opened to these sorts of apps than iOS.

🔗 Adactio: Journal—Design processing

There’s no doubt about it, using a generative large language model helped a non-designer to get past the blank page. But it was less useful in subsequent iterations that rely on decision-making:

I’ve said it before and I’ll say it again: design is deciding. The best designers are the best deciders.

Another writer coming to the conclusion that the effective use of these AI tools rely on taste.

This is just a passing though but it’s interesting to think of these tools acting less like a multiplier and more like an “adder”. Unlike something that amplifies the skills someone already has, it simply elevates it, like the tide lifting all boats. Maybe it’s a little of both.

Via: Jim Nielsen’s Notes

Not sure what’s going on but both Safari on the iPad, and Vivaldi on my phone have been feeling very sluggish these last couple of days. Could very well be a particular site I visit, but why would I still be experiencing slowdown after closing all tabs? Very strange.

I believe the gen AI could have a place in software development, but I wouldn’t say that I’m ready to go all in on “vibe coding”. I believe the best utility of gen AI comes from knowing what good code looks like. Someone with, if I may use the word here, a sense of taste. And I think the only way to develop that is to be hands on in the craft of writing code. You have to “touch” it, to feel it; almost like how a carpenter feels the grain of a new cabinet. And I don’t think you can get that if you see the code generated before your eyes.

Oof! Something I wrote has gone a little viral. It made it onto Hacker News a few days ago, and has had a bit of traction. But this weekend it’s just exploded. 12,376 views over the last two days and counting. Took me about a year and a half to get to 20k total views, and two days to get to 30k.

Ugh! I’ll use WhatsApp if I must because of network effects (i.e. my friends are there) but let me state for the record that I do not like it, nor the company that owns it. Wonder if I could persuade my friends to move to something else. The network runs deep so it might be a tough sell.

I kinda wish I had a nice “bespoke” keyboard. The Microsoft Sculpt I’m using is fine, and it keeps my RSI at bay. But it’s uninteresting. Seeing — and more importantly, hearing — the keyboard others are using at work, it would be nice to have something new.

Although, I guess I could get something for work to replace the Apple Magic keyboard I tollerate.

Pitch for the first act of a romance movie: a barista expresses their love for a customer via latte art. Maybe the customer’s sitting there, wishing for love, maybe with a friend. Then they get their drinks and the friend notices that there’s an awful lot of hearts there. The rest writes itself. ☕️

Ugh, using IMDB for anything is just awful. Would love to give Callsheet a try, if only I didn’t have an Android phone.