At that point in my weekends when I look at the source code, Git history, and old Bugzilla bugs, and shout "what were you thinking!?" and "you are smart people, you should have known better!" at my screen
Is it worth spending lots of money on a really good computer mouse? Or is that just mouse snobbery by people who want to be able to configure button 9 to shuffle their windows into alphabetical order and the like?
@sil it mainly depends on whether you need some specialised mouse for RSI reasons or not—and even there, basic vertical mice or trackballs are not that expensive these days; it’s better to pay for build quality and reliability than features
@sil yeah, that is a problem in these days of SEO crap; I ended up searching on Reddit and random forums for comparisons between devices, when I had to decide which trackball was good enough (without going into specialised hardware territory), but it’s definitely gotten harder to avoid websites that are just poor excuses for serving ads
@lw64@jorge fractional scaling has not been a net improvement, to be honest; it introduced a ton of issues inside toolkits and applications, with a poorly defined Wayland protocol, rounding errors inherent to how floating point values work, and compositor or toolkit-specific behaviour. All of that to save some money for the manufacturers, which didn't even really translate to savings for the consumers.
@lw64 the new renderer improved the behaviour, but it also uncovered a lot of issues now that rendering happens at the fractional scale instead of the next integral factor and then scaled down; once you start doing that, you'll get gaps and blurriness because of misalignments to the pixel grid. The compositor is also involved, because now windows get positioned and scaled at odd pixels and rounding errors start compounding on each other.
@lw64 fractional scaling was a mistake. It only ever works if you control the whole display and application pipeline, and if you compromise to specific scaling ratios driven by the results, not by user visible factors: you ask the user what "looks best", and then you translate that to a weird scaling factor, instead of telling the user to select between 1, 1.25, 1.5, 1.75, 2, ... values, which cannot possibly work.
@lw64 the whole concept was, not simply GTK’s choices; the alternative is to downscale things, which can be made look good-ish if you control the scale factors, but it is more expensive, and makes text look worse. It’s all trade-offs.
@Junicast unfortunately, GNOME has nothing to do with it: every application is responsible for adding those shortcuts, except Alt+F4, which is provided by the window manager. At most, we can recommend standard shortcuts: https://developer.gnome.org/hig/reference/keyboard.html
It also depends on the app: if an application is tabbed or multi-windowed, you need two separate shortcuts.
If you find inconsistencies with the HIG, make sure to file a bug in the application's issue tracker.
Sometimes I hack around with the transparency & blur implementations for #GNOME, create a #GtkWebView2 with a tranparent background enabled, and just am in awe for a minute at what is possible on this platform
@panpantepan@pojntfx GTK isn't really involved: blurring the contents of the screen behind a window requires the compositor to do that, because it's the only component that knows what's behind a window. All that GTK has to do is to make the background of that window transparent, and it already allows you to do that.
@tko yes, they restarted the numbering. It's the result of Disney getting into the production, the return of Russell T Davies as the show runner, and trying to turn the page over the Chibnall tenure.
Since we’re all onboard with telling off Linux evangelists in this moment, Linux has privacy issues too! If your screen goes to sleep, when you come back you’re greeted by your screen contents - not the Lock Screen for a solid two seconds or so (at least in GNOME Shell, the default desktop environment for many distros). #gnome#linux#security
@bdiederik@c0dec0dec0de that's really an integration issue, and it should just not happen on anything that uses logind: the compositor is responsible for telling the system to lock the session and as far as I know, GNOME Shell does that.