I doubt they'd have to retire the phone - digital radio power levels are normally pretty easy to change in the radio firmware. Which also means it's pretty easy to change, intentionally or unintentionally, in a later OS version.

Perhaps Apple chose to cheat to improve reception after mandatory testing was complete and the phone was available to buy, figuring they'd never get caught out. Perhaps Apple didn't retest with later OS versions and it was unintentional. We will probably never know.

British English voices those letters in most accents. I think the two silent letters is just a North American thing.

Similar to herb.

Pulumi code ends up looking like a DSL anyway with all the stuff you end up using from the top-level pulumi package to do anything vaguely complicated.

Only now, compared with Terraform, you need to worry about resource ordering and program flow, because when you have a dependency between resources, the resource object you depend on has to be instantiated (within the program flow, I mean - Pulumi handles calculating the ordering of actual cloud resource creation) before the dependent resource. This gets old really quickly if you're iterating on a module that creates more than a few interdependent resources. So much cut, paste, reorder. FWIW CDK has the same issue, and for the same reason - because it's using a general-purpose programming language to model a domain which it doesn't fit all that well.

I like Pulumi and it's got a lot going for it, especially if you have complex infrastructure requirements. You get a bunch of little quality of life enhancements that I wish Terraform would adopt, like cloud state management by default, and a built-in mechanism for managing secrets in a sane way. Python/TypeScript etc. modules are much more flexible than Terraform modules, and really help with building large chunks of reusable infrastructure. The extra programmability can be useful, though you need to be extra-careful of side-effects. You get more power, but you also get some extra work.

But for most people deploying a bit - or even quite a lot - of cloud infrastructure, Terraform is honestly just easier. It's usually some fairly simple declarative config with some values passed from one resource to another, and a small amount of variation that might require some limited programmability. Which is exactly what Terraform targets with HCL. It's clear to me that Pulumi sees this too, since they introduced the YAML syntax later on. But IMO HCL > YAML for declarative config.

If you're on-call 24/7/365 without a break, and it's not because you have equity in the company, then find a new job.

If you don't, then your health (physical and mental) will eventually force you to leave anyway. I did it at a startup where I was employee #1 (no equity for me), just me and the founders, and I nearly had a nervous breakdown from it, and ended up quitting from stress. Afterwards I decided I would do no more than 1 week in 3, and life got better after that.

What? Tech companies the world over have people on 24/7 on-call rotas, and it's usually voluntary.

Depending on the company, you might typically do 1 week in 4 on-call, get a nice little retainer bonus for having to have not much of a social life for 1 week in 4, and then get an additional payment for each call you take, plus time worked at x1.5 or x2 the usual rate, plus time off in lieu during the normal workday if the call out takes a long time. If you do on-call for tech and the conditions are worse than this, then your company's on-call policies suck.

I used to do it regularly. Over the years, it paid for the deposit on my first house, plus some nice trips abroad. I enjoyed it - I get a buzz out of being in the middle of a crisis and fixing it. But eventually my family got bored of it, and I got more senior jobs where it wasn't considered a good use of my energies.

Your internet connection, the websites and apps you use, your utilities - they don't fix themselves when they break at 0300.

If TSMC's approach to on-call is bad, then yeah, screw that. I don't see anything in the article that says that one way or the other. But doing an on-call rota at all is a perfectly normal thing to do in tech.

Do we have to bring this up again? It's just boring.

systemd is here and it isn't going anywhere soon. It's an improvement over SysV, but the core init system is arguably less well-designed than some of the other options that were on the table 10 years ago when its adoption started. The systemd userspace ecosystem has significantly stifled development of alternatives that provide equivalent functionality, which has led to less experimentation and innovation in those areas. In many cases those systemd add-on services provide less functionality than what they have replaced, but are adopted simply because they are part of the systemd ecosystem. The core unit file format is verbose and somewhat awkward, and the *ctl utilities are messy and sometimes unfriendly.

Like most Red Hat-originated software written in the last 15 years, it valiantly attempts to solve real problems with Linux, and mostly achieves that, but there are enough corner cases and short-sighted design decisions that it ends up being mediocre and somewhat annoying.

Personally I hope that someone comes along and takes the lessons learned and rewrites it, much like Pulseaudio has been replaced by Pipewire. Perhaps if someone decides it needs rewriting in Rust?

I'm a big fan of Kubernetes, and for larger projects the flexibility and power it brings is unrivalled. But for smaller projects, assuming equal levels of competence, delivery teams using managed Kubernetes are almost universally later and have more issues than teams that use simpler solutions. Container-as-a-service solutions like GCP CloudRun or AWS FarGate help somewhat, but are not cheap for a given amount of compute time.

Terraform (or IaC in general) absolutely has a place, because even if you use Kubernetes, most projects have more infrastructure to manage than just the cluster - at the very least, lemmy.world has a CloudFlare proxy to manage - and clicking buttons in a management portal is not a repeatable way of deploying that, or deploying the Kubernetes clusters themselves.

Ansible also has a place, particularly if you're deploying onto bare metal. I wouldn't use it for new deployments unless I had bare metal to configure and maintain, but lemmy.world is deployed onto a bare metal server as I understand it. Plus, the most effective tooling is generally the one your team understands.

I can only imagine they're shutting it down to replace it with something with different branding, based on an LLM. Microsoft has gone all-in on LLMs and I'm sure they'd love some of that virtual assistant action if they were able to differentiate themselves.

Nice to see Google doing the responsible thing here, because Apple certainly didn't when AirTags were launched.

I still think having cheap, socially acceptable, easily-accessible, highly effective tracking devices with months or more of battery life is something out of dystopian fiction though. It's not good for society in the long run.

Apple users have been sending text messages interchangeably between their phones and computers/tablets for years.

As have Android users. Microsoft Phone Link/My Phone Companion and KDE Connect have supported this for years on their relevant PC platforms. The Phone Link Android app is even preinstalled on Samsung devices. There's a teensy bit of setup but nothing complicated. KDE Connect even supports stuff like using the phone as a touchpad, remote keyboard, or media/presentation controller.

If your PC is a Chromebook then you don't even need these. If you sign into the phone and Chromebook with the same Google account, the integration just works, much as it does on Apple devices.

Most of your arguments can be boiled down to "everything is really slick if you use an all-Apple ecosystem". Which is fine, but the same can be said about Android - if you use an all-Google ecosystem with Pixels, Chromebooks and Google Workspace then most, if not all of your complaints about Android go away. Pixel Android is more consistent and less buggy than most vendor versions of Android. Integration with Chromebooks works out of the box. Google Workspace MDM is simple and straightforward, and you don't really need to buy a separate MDM solution.

The difference is that Android at least makes a decent effort to cater for a heterogeneous ecosystem. With Apple, if you're not entirely onboard with an all-Apple ecosystem then it starts getting messy quickly.

Bit of a nitpick, but the comparison with the reversing of the MS Office formats is a bit tenuous, and somewhat revisionist.

Competitors and open-source applications were reverse-engineering the Office file formats long before Apple iWork was a thing, and arguably no-one really gets it right because in order to get it perfect you'd have to reproduce the Office application layouting engine exactly, bug-for-bug. Even Microsoft doesn't get it 100% from release to release.

Nvidia drivers have (slightly) more timely support for the latest cards, and more mature support for non-3D uses of the GPU, especially scientific computing. To a large extent they are the same code as the Windows drivers, and that has positives in terms of breadth and maturity of support.

For everything else, the AMD drivers are better. Because they are a separate codebase from the Windows drivers, and are part of the de-facto Linux GPU driver stack Mesa, they integrate much better into the overall Linux experience, especially around support for Wayland. Unless you have an absolutely bleeding-edge card, they "just work" more often than the Nvidia drivers. If you like doing serious tinkering on your Linux system, then the AMD drivers being fully integrated and having the source available is a major win. Also, it used to be that the Nvidia drivers did a much better job of squeezing performance out of the hardware, but today there's very little in it, and the AMD drivers might even be a little more efficient.

I've got both AMD and Nvidia GPUs currently in different machines, and I much prefer the Linux experience with AMD. I don't think I'll be buying another Nvidia GPU unless the driver situation changes significantly.

FWIW I don't stream so I can't comment on the exact situation, but I have used the video encode hardware on AMD cards via VAAPI and it was competent and much faster than x264/x265 on the CPU. I think OBS has a plugin to use VAAPI (which is the "standard" Linux video decode/encode acceleration interface that everyone but Nvidia supports).