Apple is reportedly pushing back on AI vibe coding apps like @Replit and @Vibecode App, tools that let users create apps just by typing prompts.

The issue isn t a new rule. Apple is enforcing an old one:

• Apps must be self-contained

• They can t download or run new code that changes functionality after review

When making small one-file edits most AI coding tools feel pretty similar. But as soon as a change touches six or seven files at once, the experience is completely different.

In Editor UI tools watching the side-by-side diff file by file feels safer. Working through terminal agents is definitely faster for broad edits but it's much easier to miss subtle breaking changes or bad tests until hours later.

How do you decide between visual diff tools and CLI agents when doing large code refactors?

OP-ED?
Recently I've been finding myself actually buying and downloading apps more than before. The common thread? They're all silly things that almost do nothing.
I say almost because what they do offer is a bit of joy during my work day. Some of the recent apps I've purchased or downloaded are @Klack, Googly Eyes, @Docko, @Ball,@TabTab, and @NotchNook.
Some of these do have productivity or quality of life improvements (looking at the last two) but others are simply about making the computer fun again.
For example @Klack has genuinely made me more focused when I type and I've been able to zone in on work. It's like each clickity-clack is driving me closer to where I want to go and idk, the feedback just feels GOOD. The audio is also really nice, not sure how I can explain it, but feels very high-def for something that is mimicking a tactical feeling.
All these apps remind me of a time where shareware and P2P ( @Limewire ) was more popular. Where you might be okay buying a CD or floppy and installing something fun on your computer, then telling (sharing) your buddy about it. And with the rise of vibe coding, I think we're going to see vibeware become a thing. Where users will create something fun, quickly, using AI tools like @Cursor, @Replit, or @bolt.new/@Lovable and then put it at a super low cost or have a free-trial (shareware).
Those that don't want to pay, will create their own iteration of it and choose their own distribution method (P2P) but it won't eat at the original.
It's my genuine feeling that the internet is about to become fun again (it's already started) and I'm curious if I'm the only one feeling this way and/or embracing it?
What do you think? Is the era of vibeware a good thing? And if not why should we refute it?
This piece was written with FKJ - Just Piano in the background.

We built 111 one click apps for Arteza and 9 of them cover most of the runs. I've posted that number before. What I didn't expect is how hard the next step is.

Deleting one takes an afternoon. You check what imports it, whether a paying user has it bookmarked, whether pulling it breaks a shared component three live apps also call. Writing it took twenty minutes. Removing it takes four hours.

So they stay. Every one is something a new user scrolls past and something I have to keep working through model upgrades. Generation is cheap and deletion is full price, and nobody warns you about that half. Has anyone found a way to make cutting as fast as shipping?

There s a weird mental shift that happens when you switch from typing prompts to speaking them. When I write I naturally trim everything down to save effort, which usually gives the AI too little context and leads to messy code output.

When I talk I ramble through the logic, and the results are honestly way more accurate. The only awkward part is doing it in a shared office space without feeling like a lunatic.

Has anyone else made voice dictation a core part of their daily build cycle or does it still feel more like a novelty?

Hey everyone, Connor here. I've been spending a lot of time in the vibecoding space lately, and I want to talk about something that's been nagging at me.

The promise of vibecoding is real. You describe what you want, the model builds it, and suddenly an idea that would have taken a weekend is sitting on your screen in twenty minutes. That feeling is genuinely addictive. It's the closest thing to magic I've experienced as someone who loves building things.

It's so easy to fall into the trap of over engineering a quick weekend project just to keep the styling under control. A massive design system feels like total overkill when you're just testing an idea but without some rules the third screen always looks like a completely different app. I've started keeping a basic note with just a few color and spacing constraints to feed the assistant.

What is your go to trick to keep your UI looking clean without slowing down your build speed?

As a solo dev, I can't afford to ship something and then realize it's unprofitable. But estimating token usage before launch is hard.

Do you just... guess? Build a prototype first? Or is there an actual way to forecast what users will actually cost you?

I ve run into this with longer agent runs and I m curious how others handle it.

The original issue might contain the scope, constraints and evidence required. But when another model or session picks up the work later, those conditions often get reduced to a summary.

Hey everyone. I use Claude for everything. I'm sure you too.

My biggest pain with it? Not knowing when I'll hit the 5-hour window limit.

Turns out, you can use Claude, to update Claude's UI and show you your current usage.