Turing machine isn't a 'conventional computer' in any way it's a model not asomething that can be implemented as described. Philosophical weight has little value anywhere. Author tried to reframe increasingly cultish bullshit that 'ai' crowd loves to push.
I've always had either problems with schedules or problems with environment because it has it's own. Cron is fine as long as you are lucky. That isn't to say there never were problems with timers, on debian 9, I think, shipped version of systemd had a bug where systemd-analyze parsed schedule correctly and timer just shit itself. But only for some specific case I can't remeber which, but it was extremely frustrating to debug.
I really dislike dumb and entitled takes like this. First he somehow complains about both GNOME and KDE when literally none of the complaint apply to KDE, second switching between different operating systems will break the way you are doing things anyway, and third - who cares.
And then with your explanation it goes: cron job doesn't execute for reason, crontab -l lists not all jobs, someone else put cron job in whatever directory like cron.daily, you've added or removed empty line at the end and now nothing works, debian crond and redhat cronie ahave different quirks, etc., etc.
I've dealt with insane cron problems for so long that switching to timers was like whoa i don't need to suffer?
First, it is usable just not the way Talon requires (and its developer doesn't want to support wayland). Second people can use LTS distributions that will ship X11 for many years, and by that time it will either be good enough or won't.
Which all of them, which cron implementation is simpler and better? Do you even use cron? It's not a joke, explanation on how to use cron takes > 8 pages. It should be in the hall of fame of the least accessible and hard to learn programs. Lennart cancer is one of the greatest things that happened to linux.
It's not so bad like the article author describes, there is work being done to address the input issue e.g. libei and wayland protocols. (also N.Graham answer that says that it should be done on wayland level if they require universal solution is right if a bit unkind)
Rich vampires would even subsidize it, because it will be a hellhole, due to it being an extremely controlled environment with alternative being death.
Npm having a lot of packages at least partially a problem of lacking standard library, no? And partially developer culture where every trivial thing is a package. Anyway, similar thing will happen to rust soon enough (looks at 1 gig of dependencies for a cli program).
Turing machine isn't a 'conventional computer' in any way it's a model not asomething that can be implemented as described. Philosophical weight has little value anywhere. Author tried to reframe increasingly cultish bullshit that 'ai' crowd loves to push.