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.
From time to time, I open a different browser to see if I’m missing anything. I tried the Dia Browser again, but I still don’t quite like it. I feel unsure about it. Its design doesn’t appeal to me, and it seems slower than the ARC Browser. My default choice remains Apple Safari. Oh, I want to like the Zen browser, but I don’t like it to be based on Firefox.
In “Google Betrayed the web”, Mike Elgan argues that Google has broken the foundational “Grand Bargain” of the internet—trading traffic to content creators in exchange for indexing their content. Google now uses creator content to train AI models and serves answers directly to users, bypassing the original sources. This shift eliminates referral traffic to publishers and independent sites while allowing Google to profit from their work twice over. Elgan calls for a post-Google internet as a response to this betrayal. His response centers on Google’s alternatives: Kagi, Perplexity, Fastmail, and Antropic. All alternatives that I personally use, except Perplexity.
If all goes according to plan, I’ll be upgrading my aging iPhone 15 Pro Max to the iPhone 18 Pro Max this coming September. For each upgrade, when I upgraded to the iPhone 13 Pro, and then to the iPhone 15 Pro Max, I wrote a personal take. Both are quite different, and I’d like to put another one this year. The big difference: the place AI could take to help me put this together. I have all summer to think about this. And I know what some people might be thinking about those who use AI and write about using AI to do their things.
Our healthcare system does miracles, but it also relies on practices that are sometimes questionable.
Yesterday, when I went to pick up my wife from the hospital after her jaw surgery, she needed to have one last X-ray. The device is the same as at a dentist’s office, with a rotating head that can turn 360 degrees around the patient’s head. This device is connected to a PC running imaging software. Once the X-ray was taken, the doctor examined it and decided to share it with a remote specialist for a second opinion. The method used was quite surprising.
The doctor started Chrome, opened WhatsApp Web, and authenticated with his WhatsApp account on his personal phone. Then? He took a screenshot of the X-ray using Windows’ utility and sent it in a WhatsApp chat with his colleague. It’s an unusual procedure, but it stems from a flaw in the medical imaging software: the lack of a secure messaging feature. Is this approach approved? I doubt it, but I can’t blame the doctor. His action allowed all of us to save time. And what about the quality of the image on the other end? Using the Windows screenshow utility is certainly not the best way to share a detailed view of my wife’s jawbone.
Oh, and another surprise: the PC had a label with a password written on it to enable a recovery function in case of problems. I thought this practice was outdated. Apparently not.
