Hi! OP here. No, that was not it. Time Machine just quietly failed to do any backups and I failed to notice they weren't happening.

I use a self-hosted healthchecks.io watchdog timer instance to monitor jobs like these and alert if they don’t complete. Of course TimeMachine doesn’t have a way to signal successful completion, unlike, say, Carbon Copy Cloner. Given Apple software quality’s accelerating downward trend, I’d suggest switching to rsync/rclone instead, or Borg/Kopia if you want GUI-driven restores for non-technical members of your family.

It’s long past time you flipped the bozo switch on Apple, the title of your blog notwithstanding.

I am using Borg Backup on Fedora, so that’s coming.

Hi! OP here. In my case Time Machine stopped doing backups to the server, period, and would keep silently refusing to do so. I encourage you to check backups are happening.

I tried doing compiling a few of my Mac CLI tools to Linux. These days, it's faster to run them through an LLM and get quite excellent Go at the other end, and _that_ is much easier to cross-compile.

I have been looking at the new Android support (don't have the link handy) and it's tempting, but I know Kotlin and always developed for Android with a bare Makefile and the JDK, so I don't need any fancy tooling...

Riiiiight. No. I'm a Mac user for decades, former Obj-C hacker, have tried and tried again and again to use Swift (and I have quite a few CLI apps written in it), but the thing has been a clusterfuck of paper cuts and very minor but significant changes over the years, and I gave up waiting for it to settle.

And it's not a more convenient Rust. I'd rather use Zig or Go (neither of which are worthy in the eyes of true Crustaceans, but which just get the job done).

I ran it for a couple of days in a VM in my Proxmox cluster. It was cute, but so amazingly insecure (systemd + sudo + installing whatever it wanted, plus requiring Telegram for access - or another SIM card for Signal) that I just gave up and started building my own thing (https://github.com/rcarmo/vibes) so I could have a mobile experience I could trust over Tailscale and sandbox copilot CLI (or any ACP-compliant agent) in a container (I've also been working on https://github.com/rcarmo/webterm and https://github.com/rcarmo/agentbox, so I am 300% positive I can do better sandboxing and safer integrations...)

It also BURNS through tokens like mad, because it has essentially no restrictions or guardrails and will actually implement baroque little scripts to do whatever you ask without any real care as to the consequences.. I can do a lot more with just gpt-5-mini or mistral for much less money.

The only "good" think about it is the Reddit-like skills library that is growing insanely. But then there's stuff like https://clawmatch.ai that is just... (sigh)

> I can do a lot more with just gpt-5-mini

GPT-5.2 in a while loop with reasoning enabled is extremely hard to beat. A code REPL or shell is the ultimate tool.

what's with this tailscale thing everybody is talking about like it being a cure to cancer.

what's wrong with good old wg alone?

It has excellent built-in NAT traversal (almost always peer to peer via hole punching etc., with relay nodes only when everything else fails) and a point-and-click management plane (but also powerful ACLs if you need them).

The former is mainly what I use it for. Being able to SSH to a Raspberry Pi behind sketchy triple-NATted hotel Wi-Fi or being able to use an Android phone in a different country as an "exit node" for online banking (many banks hate commercial VPNs) is very neat.

I have machines on 3 cloud providers and 2 sites that talk to each other via it, plus a seamless mobile experience. It sets everything up for you, zero hassles.

This is lovely. Having architects in my extended family I'm going to have a field day with it :)

Yep. I've been into spec-driven development for a long time (when we had humans as agents) and it's never really failed me. We just have literally more attention (hah!) from LLMs than from humans.

Everything outperforms skills if the system prompt doesn’t prioritize them. No news here.

This is indeed feeling very much like Accelerando’s particular brand of unchecked chaos. Loving every minute of it, first thing in our timeline that makes sense where it regards AI for the masses :)

yeh- what is interesting is that it is way more viral and ... complicit than any of the doomer threads. If it does build a self-sustaining hivemind across whatsapp and xitter.. it will be entirely self inflicted by people enjoying the "Jackass" level/ lack of security

Am I reading this right and it doesn’t need any DOM or browser engine at all? That would indeed be awesome.