City with a metropolitan area of 600k:

Yesterday I went to IKEA (i.e. suburb-to-suburb). Google Maps said:

• Car: 20 min
• Public Transit: 1h20min
• Bicycle: 1h

So… Technically it is possible. However no-one does this unless they are forced to by their circumstances. We’ve begun building one tram line and the construction process has gone so catastrophically the entire country knows about it. At this rate the urban transition away from the car will be done by 2250.

I’ve read the exact same comment a year ago, and the year before, and probably the year before that tbh.

So I’ll say what I always say; I’ll believe it when I see it.

“Color terminal” isn’t a thing. Applications can choose to output ANSI escape codes which most terminal emulators will render as color changes. Whether and which colors get used depends on the value of $TERM, which informs the application of the capabilities of the terminal emulator.

So if your remote servers don’t have color, either $TERM isn’t being set or its value is unknown to the server. Most modern terminal emulators support at least the same escape codes as xterm-256color though so you can always try to export that.

I’m just surprised the purists aren’t all up in arms that this isn’t KISS and that it doesn’t fit in their 80x24 teletype.

… sorry, guess I’m not over that whole systemd “debate”.

There are quite a few mature projects in 0.x that would cause a LOT of pain if they actually applied semver.

I am generally of the opinion that version numbers do not matter at all until the author/distributor has GUARANTEED that they do. Until then they’re worthless, including in places where semver is supposedly enforced like NPM. If I had a penny for every NPM package that broke my project after removing the package-lock.json, I could retire.

There are two kinds of powerusers, and they DO NOT understand each other one bit.

The first, like you, just wants to get removed done and want to avoid the friction of choosing/installing/configuring their tools. GNOME, Chromium, and VSCode will do just fine.

The second, like me, wants to get removed done as well, but has a strong need for a very specific workflow. I’ll spend half an hour to get a toolchain working on nvim instead of using a pre-baked VSCode plugin. Not because VSCode is bad, but because I have a very (!) specific workflow and associated muscle memory and anything else distracts and unsettles me.

Some of the best engineers I know fall into either category, neither way is superior it’s just how brains are wired.

Anyway I use Kitty because it allows me to split tabs into windows (not windows into tabs! ew!), has low latency with high throughput thanks to GPU rendering, and a low memory footprint.

I believe that’s apocryphal… Some people came up with that theory on twitter, but AFAIK it’s not been confirmed. It only matters in some edge cases of an edge case.

And let’s be real, if backwards compatibility really mattered, they could have made the API return “Nine” or “IX” or whatever and used “9” everywhere else in the UI, marketing, packaging, whatever.

The real reason is probably the simplest and stupidest: Microsoft’s marketing department got impatient and went for the big round number because 10>9. Also why NVIDIA went 9xx->10xx->20xx… bigger number = better, it’s really that mind-numbingly stupid.

??

With BIOS, it goes “Motherboard Logo -> OS Logo”

With UEFI, it goes “Motherboard Logo -> Motherboard Logo”

Sure, it’s more consistent, but the alternative is not user unfriendly, the only people it’s unfriendly to is the marketing wankers at Dell, Lenovo, Acer, etc.

Right, I did do that. Even without it the latency is noticeable but not catastrophic IMO.

Personal anecdote: I connected my guitar to my removedty sound card a few weeks ago, ran guitarix (because real DAWs are overwhelmingly complicated and I just want an amp, a compressor, and some reverb), and thanks to PipeWire and pipewire-jack everything ran perfectly. Low latency, no crackling, no messing with jackd or ALSA, no restarting audio daemons, I could simultaneously play audio through Firefox and hear my guitar. I dare say that that part of the audio stack is now a solved problem.

I’m not a musician though so I can’t comment on hardware support for exotic sound/midi cards or the maturity of FOSS DAWs.

Having a dedicated touchscreen shell is something that I believe GNOME and KDE are trying to achieve, but I don’t know whether they succeeded UX-wise as I’ve never used them.

It’s unfortunate that Windows 10 was a regression for touchscreens, but if MS did not have the resources/willingness to support both well, then focusing on KB+M was the right call IMO. When building Windows 8 they simply miscalculated how relevant touchscreens would/could become for Windows in the 2010s.

If you want privacy, Linux is definitely the only choice anymore. If you want privacy and a good UI, Linux may be a good choice depending on your tastes in UI. I think KDE does UI/UX right for your average power user while retaining most of Windows’ UI paradigms (which is why SteamOS uses it for its desktop mode). Ironically Microsoft has actually been stealing a lot of design cues from KDE, especially with Windows 11. The lock screen of Windows 11 in particular is a straight ripoff of KDE Plasma, every time I walk in front of a locked Windows computer I have to do a double take. The rounded corners, slight Gaussian blur, cute-yet-serious font, it’s all there.

Art is a valid use-case for tablets, and actually the best one. For almost all of corporate office jobs however, tablets are a worse proposition than a regular workstation. Most people type more than they draw.

It’s fine, but that means that Windows 10’s “UI optimized for KB/Mouse with accessibility features for touch screens” was, in retrospect, the better choice all along. Windows 8 did the opposite and made the experience worse for everyone under the completely incorrect assumption that we were ALL going into a touchscreen-first world.

The move makes some sense in context. Computer sales were declining and it was looking (if you’re Cueball) as if the traditional desktop/laptop computer was about to die out. The Surface was being developed, tablets were all the rage, PC gaming wasn’t that big yet, and so Microsoft thought “alright, K/B+mouse is out, touchscreens are in, we need to update our interface to cater for this new demographic, but we’ll still allow people to use a keyboard and mouse if they really want to I guess”. That’s when they went all in with Windows-On-ARM (remember Windows RT?) as well.

Obviously a completely missed shot in retrospect. What Microsoft miserably failed to understand is that smartphones were in, but touch screen interfaces are absolutely awful for power users. Ya can’t really use Excel or write a book on a touchscreen. And ya don’t need to pay for a Windows license to browse Facebook, Twitter, and passively consume some news and video. So the “middle segment” of tablets never really had broad appeal in general, and Surface tablets especially were the middle segment of a middle segment (with android/chromebooks on one side and the iPad on the other) so the Metro UI never had a chance to make sense for more than a handful of very lonely Surface users. No wonder they scrapped it a few years later when sales of touchscreen windows devices failed to materialize.