Numeric Citizen Blog

If you’re shopping for a Mac mini in Canada right now, the Apple Store configurator is both powerful and opaque. It shows you one combination at a time, with no easy way to compare shipping wait times across the full lineup. So I decided to do something about it.

Using Claude’s browser automation tools, I navigated the Apple Canada store programmatically, cycling through every possible Mac mini configuration: all 52 of them, spanning two chips (M4 and M4 Pro), two CPU/GPU variants, two memory tiers, five storage options, and two Ethernet speeds. I captured the live price and shipping lead time for each one directly from the store. No guessing, no third-party data. Everything you see was pulled in real time from apple.com/ca.

The result is the reference table below. The headline finding is blunt: if you want a Mac mini without a long wait, your options are narrow. Only the base M4 model ships in 3 to 4 weeks. Upgrade almost anything (memory, storage, Ethernet, or chip), and you’re looking at 10 to 12 weeks. The Mac mini may be Apple’s most configurable desktop, but right now it’s also one of its hardest to get quickly.

Google’s Gemini Mac App Is Native, in a Distinctly Google Way, But Annoyingly Presumptuous, but John Gruber had this to say:

(Sidenote: The Gemini Mac app is a native Mac app, but it is … weird. Gus Mueller poked around at it and found that it’s the product of a Java-to-Objective-C converter that Google made, and much of it was originally written for Android.)

A trillion dollars company can’t make great Mac app, i.e. native Mac app. 🤢 Pass.

Working on my iPad and Microsoft apps during lunch while attending a webinar… if you think Apple software is full of paper cuts, you ain’t see nothing because Microsoft apps like Outlook and Teams are literally full of bugs, mostly user interface related. 😩🫤

So you were expecting a long blog post about my expectations for WWDC26, right? Well, I don’t have a clue. I think it would be better to write what I want, not what I expect. Here’s what I want.

• A slider-type control for toning down Liquid Glass. This UI must disappear somehow, to a degree.
• A useful Apple Intelligence, requiring minimal third-party support in order to be useful. Apple’s relationship with the developers is at an all-time low; it’s not the right strategy to depend on them for basic OS feature support.
• Speaking of relationships: I want Apple to care about developers, for real. Don’t pretend. Be honest. Be humble.
• Finally, I want bug fixes, all sorts of bug fixes.

That’s it.

This morning, after reading and asking Kagi Summarizer for a summary of this article, I wanted to write a response and attempted to craft a counterargument, first using Kagi Summarizer, then using Claude AI. I reviewed numerous versions but remained unsatisfied with the results. After reviewing my options, I ultimately decided to create this version, entirely my own. I still have a feeling that AI helped forge my thoughts.

Since reading David’s article, “the solution might be cancelling my AI subscription”, I cannot stop thinking about how different our experiences are.

Six months ago, right before subscribing to Anthropic’s Claude AI, none of the following custom-built apps existed: a useful, personally fitting bookmark manager; a purpose-built RSS reader that works hand in hand with my bookmark manager; and a simplified task manager that also works in conjunction with the other two. As an experienced user of many RSS readers, bookmark managers, and task managers, knowing my friction points with these apps, I have always dreamed of what would be perfect-for-me versions of them. So I built those, one by one.

Thanks to Claude Code and my vision for what would make the perfect versions of each of these apps, I could build them without being constrained by not knowing Next.js, TypeScript, CSS, etc. I don’t plan to make commercial versions of these nor open-source them. There are a few, and I can tweak them as I see fit and as my needs evolve; this is where I’ll keep focusing. AI empowered me and will continue to do.

You see, the solution for me is to keep iterating and keep my subscription. Two different stories, two different outcomes. Maybe there is something that I didn’t catch in David’s experience.

If this year’s OS releases are Apple’s way of fixing its software, do these rumors still make sense if work has already started on next year’s major updates? How can Apple manage to do both and maintain them simultaneously? I understand that some features take months or even years to develop, but this doesn’t seem to fit with the rumors. Alternatively, the rumors might just be reflecting common expectations.

Apple AI glasses launch pushed back to late 2027, Vision Air to arrive by 2029: report — 9to5Mac

Apple believes it has a massive opportunity here in the eyewear space, and wants to potentially capture billions of people who depend on prescription glasses, casually wear sunglasses, or use glasses as a fashion accessory.

Until Apple released the iPhone, wireless providers dictated everything about the buying experience, but Apple sidelined them. Could Apple repeat a similar tour de force with prescription glasses? Imagine getting your prescription, then heading to the Apple Store to pick up your smart glasses. I do see this as Apple’s next disruption. Really.

Taking a short break today for a 50 KM bike ride around Montreal Island. The expected weather is cloudy and rather cold, at 12C, with a 40% chance of scattered showers. I wished for better weather, but hey… could be much worse. Have a great day. 🚴🏻‍♂️

I’m getting there… by myself. I’m finally getting the reading workflow and tools that I always dreamed of. It might sound complex, but it isn’t. I can start at any reading circle level; no need to go with the smallest one (my blog roll). Seventy percent of the time is now spent in Ink⋅well. More to come.

I don’t remember the last time I built a full automation workflow from the ground up in n8n, thanks to Claude AI and MCP support. I started manually before knowing n8n had MCP support integrated, which makes me feel more competent in understanding what’s going on. It reminds me of when I bought an iPad touch so that I could learn to build apps with Objective-C and Xcode; I like to understand what’s going on under the hood, in the digital world, at least.

I feel like I’ve reached a stage where I’ve completed all the projects aimed at improving my digital ecosystem. The native apps I wanted to replace have been replaced, and the new apps perfectly suited to my wishes are now a reality. I’m entering a stage of sporadic fine-tuning. It’s sometimes satisfying not to always have to work on the toolbox and instead simply be in a mode of using these tools.

I just completed a new workflow: automatically saving new bookmarks stored on Micro.blog to my custom-built bookmark manager web app running on Vercel. Since Micro.blog doesn’t support webhook calls, I had to resort to a scheduled n8n workflow that pulls any newly saved bookmarks via Micro.blog APIs and saves each one using a new API route on my bookmark manager web app on Vercel. It’s much more efficient than asking Claude AI to do this using a skill (which was working perfectly, by the way) to save into a Craft Collection block entry.

What if, as soon as we shared content on the Internet, you couldn’t remove it as soon as someone was referring to it or embedded it in some other content? I’m thinking about the open web here… would this hypothetical web be called the open web anyway? Would we be more intentional when sharing content having this rule baked in? This chain of thoughts was triggered when I came across a website with a blog post with some embedded content from YouTube. Some videos were no longer available and made the blog post more or less diminished.

iOS 27 might convince a lot of people to upgrade to a new iPhone — 9to5Mac

Before now, Apple Intelligence arguably hasn’t provided enough reason to motivate upgrades for most users, but in iOS 27, that could very well change.

Unless Apple hits a home run with this year’s Apple Intelligence upgrade, I tend to think that we overestimate the attractiveness of AI for ‘ordinary’ iPhone users.

I saw Matt Birchler’s work in his latest Quick Read application here and thought it was nice UI work. I wanted to see if I could do something similar for the color configuration for tags in my task manager. It was much harder than I thought, but I’m happy with the results.

Melius’ Ben Reitzes raises price target on Apple to $385 — Asymco

An agentic version of Siri could boost 2030 revenue by as much as $65 billion, Wamsi and his team estimated, adding up to $2 billion in incremental earnings over the next four years.

Siri+. I hope Apple don’t read his blog.

Report: Apple Plans to Make On-Device AI a Key WWDC Focus:

The arrangement represents a noticeable departure from Apple’s original Apple Intelligence announcement, in which the company said all cloud-bound queries would be handled exclusively by its own Private Cloud Compute infrastructure running on Apple silicon. Apple is likely to retain the Private Cloud Compute branding despite the change, people familiar with the partnership told The Information.

There is much left unanswered in this article. How much of Private Cloud Compute is in use today and for what purpose? How much of this capacity will be running the new Apple Intelligence? Will Apple expand PCC into Google’s datacenters? If so, what is Apple’s own infrastructure going to serve? And how much of an improvement can we expect to run local models on our devices compared to the original Apple Intelligence model?

It took me the drive, Codex and Claude Code, four days, and voilà! Things 3 is now a thing of the past; I’m using a perfect-for-me custom-built task manager web application running on Vercel/Neon. Things 4 might go subscription-only; I don’t care.

While moving my tasks out of Things 3, I noticed that many possible projects or tasks were really just daydreams. Migration is an ideal opportunity to reevaluate everything with a fresh perspective. Then, the new digital home will host a new set of daydreaming projects and tasks that I’ll revisit with a smile and kill right off the bat.

You know Apple created chaos when Apple Publishes Document to Help Users Tell Creator Studio Apps Apart. 🤔🤦🏻‍♂️

After largely completing the transition from Things.app to a version built with vibe coding on Next.js and hosted on Vercel, I decided to switch from Airtable to a Postgres backend. The free-tier on Airtable only allows 1000 public API calls, which wasn’t sufficient. To avoid hitting this cap constantly, I, with the help of Codex, migrated to a Postgres database called Neon, available through Vercel’s marketplace. I already use Neon for my bookmarking web app, so the change was quite smooth. I didn’t need to move any data since I am still in the late development phase. Now returning to other fine-tuning tasks.

Well, well, well, it seems to be happening much quicker than originally thought. I’m about 70% done with this already! I still like Claude Code more, but Codex is more efficient at testing the UI in its browser than Claude Code. And the in-browser cursor, which is 100% independant than the Mac cursor, is super cool to see in action.

WordPress 7 shipped with new AI features, where are they hiding?:

The latest version of WordPress (WP), seven, which shipped a few days ago, comes with, according to the accompanying release notes/marketing copy, a number of AI features.

I’m yet to see even one of these, despite installing version seven last week now. The only noticeable difference I can discern — to date — is a change in some of the hyperlink colours on the dashboard.

Not that I’m looking forward to it, but Ghost still doesn’t have any AI-related features.

So far so good with OpenAI Codex. I prefer Claude Code’s look and feel, but Codex seems more like something Microsoft would build: plain, without soul. My data model is completed and implemented in AirTable. The basic web UI is running, but is lacking many basic operations like create, update, and delete for tasks, projects, etc. It took me less than two hours to deplete my token allotment. It’s a slower rate than with Claude Code… but still, it’s quickly gone.