Reuben
537 posts

Reuben
@RubZH
Software engineer & designer. 🐱 Download free stuff: https://t.co/2OoXSEUhgt Contáctame en: https://t.co/v8wuYl4BNV
Mexico 参加日 Mayıs 2010
154 フォロー中46 フォロワー


It’s funny because I never use any kind of file browser for code I just quick open file by name.
Edward Sanchez@edwardsanchez
English
Reuben がリツイート

@frederikRiedel @one_sec_app If your user wants to breathe while scrolling TikTok, let them.
English

Just wanted to share some concerns of Apple deprecating `UIRequiresFullScreen`:
Starting in iOS 27, all apps will be forced by the system to be freely resizable.
However, for our use-case with @one_sec_app, we require a breathing exercise to be displayed full-screen / on top of target apps.
Having one sec pop up side by side with social media apps misses the point.
This change will be great for most apps – especially looking at foldable iPhones in the future, where this probably applies as well.
I’m asking Apple to consider a use-case like one sec as well – and to ensure that we can continue to offer a great experience on multi-window environments as well!
FB18763293

English

@jacobtechtavern Is it a cascade or why can’t some of the steps run asynchronously?
English

The launch path of your app can rapidly get filled with new services, setup, and SDKs. This code that gets executed once your app cold-start up, and runs before your first screen can be interacted with. Often, everything is dumped in didFinishLaunchingWithOptions (screenshot #1)
Simple enough, our LaunchOrchestrator runs everything in series (screenshot #2)
As an illustrative sample project, I tried to be semi-exhaustive as I thought up the common steps you might encounter in a standard app launch.
The result is all the same: 6.5 seconds of utter agony (or whatever it feels like when you’re alone with your thoughts).
If you think 6.5 seconds beggars belief, you’re one of the lucky ones. If you’ve ever encountered B2B SAAS, you’ll know purchasing decisions are rarely made with end-users in mind, and cost control is the name of the game.
Often, mandatory internal tooling is cobbled together by overworked, underpaid, offshore teams, where “performance” is relegated to nice-to-have, and descoped before work even begins.
This start point gives us a lot to work with when it comes to optimising launch time.
Subscribe to my blog to accelerate your iOS career with new posts every week! 🚀 blog.jacobstechtavern.com/subscribe



English

@AetherAurelia I was missing Arc too, decided to move to Zen and it’s a very good replacement. Give it a try.
English

@FloWritesCode I’d give it a few months because it takes quite a bit for people to update.
English

@FloWritesCode Add telemetry to your app and decide wether to cut support for iOS 18 based on that.
English

I also finished my iOS & macOS 26 "redesign" of LaunchBuddy.
There are only three things I actually changed to be a better iOS 26 citizen:
- Custom TextFieldStyle
- Moved some navigation items to a bottom bar
- Hiding the TabView once an app is selected
The only question that remains: Should I cut off iOS 18? It would also make my planned updates easier and I think I should be able to pull it off with my developer target users. I still have some time to decide.
My main reason: I already have lots of #.if os(iOS) checks throughout the code and I really don't want to add checks for OS versions as well...
After fixing a crash last week I think that LaunchBuddy is in a good enough state to be unmaintained on older OSs... This is a tough decision.
English
Reuben がリツイート

I guess I just, won't try copilot for something today? How do people even pay for this
aether@AetherAurelia
I wanted to try Copilot in VSCode after trying other tools, but I just can't - literally no matter what I do, I cannot send messages
English

@dennismuellr @RubZH @tryamie Wouldn't this be feasible with SwiftUI? What's your reason for choosing UIKit if you don't mind sharing?
English

the @tryamie split view revived the company
we had been working on it for 3 months, and prototyping it for another.
i could feel that the team started getting a bit nervous.
"why are we even spending so much time on this?"
and we didn't even launch it. one day it just went live on the testflight.
no newsletter. no push notification. no tweet.
and then @sekachov tweeted this video, @avstorm followed up.
signups went ballistic. and the users stuck around, since it wasn't designed just to be pretty.
it was designed for people using their calendar to plan their todos.
this was months before we launched publicly, and we really needed that moment of hype. of being all over twitter.
would i do it again? maybe, probably! custom design like this takes a crazy amount of time, so it's hard to make up directly through MRR.
it's still my proudest interaction work. designing those is probably when i feel most in flow.
English

@AetherAurelia @t3dotchat In my case sometimes it switches languages. It's been behaving very weird specially with long responses.
English














