That’s the problem of most general-use languages out there, including “safe” ones like Java or Go. They all require manual synchronization for shared mutable state.

They are trying to make money to stay afloat. Postmarketos is a community project so it’s not comparable. And neither Purism nor Pine64 seem to be huge commercial successes just like Jolla, though they seem to be doing a bit better.

They have been owned by a Russian state-owned telecom corporation for a few years until recent events (Russia currently tries to push Sailfish OS fork as its “russian-made” mobile OS). Original Finnish management has split off to a new independent company with the same name last year, and this looks like their last ditch attempt to continue existing. I don’t expect they will last much longer (the reason why they were bought by Russia in the first place was that Jolla failed as a business).

Qt 6 has been out for more than three years now.

It’s not ready yet.

The protocol for apps/games to make use of it is not yet finalized.

Gconf was text though (well XML actually but not binary).

Variable names shouldn’t need comments, period. You don’t want to look it up every time this variable is used in code, just to understand what it holds. Of course there are always exceptions, but generally names should be descriptive enough to not need additional explanation.

And context can also come from names of other things, e.g. name of a class / namespace that holds this variable. For example AccessibilitySettings.HighContrast, where AccessibilitySettings holds all options related to accessibility.

It is scalable but the icons are still drawn against the virtual pixel grid. If an icon is designed to be perfectly pixel-aligned when rasterized at a certain size, then rasterizing it at 1.25 of that size will cause small distortions if it contains small elements (such as 1 px width lines).

Please, someone tell comrade Stalin Xi that this is all just a terrible mistake!

I fear that at these resolutions we will discover that UI frameworks are not really scalable in terms of performance

Even “real” fractional scaling in Plasma with Qt 6 is not much better. Text will look slightly sharper, but icons are still blurry. There is no way for them to look sharp with 1.25 scaling since they are drawn with a pixel grid in mind. Unless you invent some way to stretch svgs so that their individual elements and spaces between them retain their integer-ness while the scale of the whole image is fractional.

The only other solution is monitors with 300+ PPI where blurriness is simply not noticeable (that’s the way Apple went).

It looks mostly the same as XML views but some components look and behave wildly different for no apparent reason (tooltips are one of those).

I have given up on the fight a long time ago. On the desktop the only line I draw is that the app must respect system font configuration and use system-provided file dialogs.

JS by itself is very fast (it’s one of the fastest dynamic languages). It’s interop with platforms APIs that is slow, at the fact that each React app spins up its own instance of Chromium also doesn’t help.

Same with Compose even though it’s ironically considered native in the Android dev community.

The easiest way to tell that the app is not native is tooltips (those that appear when you long press on a button in a toolbar). For some reason UI frameworks just can’t agree to display them in the same way, even if they use material design. Compose’s ones are especially bad (some apps like Play store actually have different kinds of tooltips on different screens, meaning they use multiple UI frameworks in the same app).

That can be true for self-contained command line tools, but not for complex programs with actively development dependencies (especially anything dealing with networking or encryption). For example hexchat uses GTK2 which is likely to be removed from mainstream distro repos in the coming years because it has been obsolete for a long time. Also openssl which is known to change its API occasionally which means that anything that uses it needs to be updated to stay compatible.

I tried cosmic in VM and they have a long way to go. Whatever they release in 2024 will be just barely usable, nothing more. Think of stability of KDE 4.0 with 1% of its features. I’m not saying that they are doing a bad job, quite the opposite. But what they have right now is only nearing the bare minimum, and the road ahead is long.

Wayland is not a drop-in replacement tho. It’s like if glibc developers declared it obsolete and presented a “replacement” that has a completely different API and has 1/100 of glibc functionality and a plugin interface. And then all the dozens of Linux distros have to write all the plugins from scratch to add back missing functionality and do it together in perfect cooperation so that they remain compatible with each other.