> When you dynamically link a library, the entire library is loaded into RAM.
It doesn't. When you dynamic link a library, no part of it is loaded into RAM. It is page-faulted in, as it is used. In the end, only parts that were really used were loaded into RAM.
It is easy to find windows app, that is either blurry (basically almost all installers), or is confused by hidpi (take any qt app and you have 1:1 chance it will get it wrong, like this: https://imgur.com/a/Ad0ZkmX).
They used to be, but not anymore. However, they switched, because even if not pixel perfect, it is good enough. At sizes used by their laptop, most users are not going to notice that it is not pixel perfect.
The system they use gets them something: 1) apps do not have bother with fractional scaling, so they do not get to bother about rounding errors and 2) they get to use output encoder for scaling, which is essentially for free, instead of GPU, which will take its toll on power consumption and GPU load.
> a cyclist ran over a child (ER took the child to hospital) on a busy promenade on the bank of the river.
Promenades happen to be dedicated to pedestrians. So pedestrian was hit in a place, where he was supposed to be safe.
And then...
> Meanwhile there's over 2000 deadly car accidents a year in Poland, 20% of which are pedestrians[1], meaning there's more than one pedestrian killed by a car every day.
These accidents probably didn't happen on promenades, pavements, sidewalks, footpaths or other places dedicated to pedestrians.
So back to my point: comparing apples with apples. This comparison wasn't it.
On the other hand, if you take a book that leans towards ideas and concepts, and you make series with character-driver narrative out of it so people won't tune out, you get something like Foundation, that has only title and names of places and characters common with the original. All that made the original interesting is lost.
You can join a meeting; you cannot share your desktop. You cannot share your desktop, because that software is buggy and without complaining, that bug will be never fixed.
Nobody pays Nvidia. Nvidia has an opportunity to enter the market (on Windows and Linux) and it is up to them to use it. It usually goes easier if it is not against the grain.
On some platforms (Mac) they don't have that opportunity at all.
Yes, Pipewire is at 0.3 today; but that's not a problem. Pipewire has the ambition to replace Pulseaudio, Jack and provide APIs for several other subject, not just screen grabbing. That's the part that is missing toward 1.0.
Screengrabbing during the last year was even better than what X11 could provide, including the ability for the user to limit the grabbing to display or window. That's something that user cannot have under control in X11.
Pipewire is also not limited to Wayland; it will work with X11 too, so the X11 users won't be abandoned just because there's Wayland support too. Firefox doesn't seem to have problem with it, Chrome has it behind flags. For apps with fast moving releases as these, I don't see any problem with them supporting 0.3 release and when/if there is API breaks, updating it.
Beats not supporting screen sharing under Wayland at all.
While I agree, that support for popular apps is a problem for platform, it doesn't mean that the platform will freeze for every snowflake app.
Apple for example regularly uses pressure from users on app authors to drive improvements in their platform. I don't see why Linux distributions should flip backward to keep their platform static, to the detriment of their long term development.
> How much of the market share is Nvidia? A stupid quick search's top result says 80%. So 80% of PCs can't run Wayland?
No.
Long-term Nvidia share is about ~18%, AMD ~12% and Intel ~70%. But that's for the entire market, and since Windows gamers are skewing toward Nvidia, that makes the rest to make up the percentage.
So 80% not being able to run Wayland due to being Nvidia is grossly incorrect.
> ost wayland advocates can't seem to understand the philosopy of "I don't really care about xorg vs wayland, my current xorg-based setup works well enough for me, i'll switch when I can be equally productive on wayland".
Because usually the part "my current xorg-based setup works well enough for me" is not true. It will be said, when it comes to wayland, and they don't want to switch... but once hidpi or mixed-dpi displays or reliable screen-lock or something other comes on topic, then suddenly "linux" (not xorg, but linux) is behind the times, doesn't work out of the box as other systems do and requires hours and hours of configuring.
That's the "my current xorg-based setup works well enough for me" in the full glory.