Failed to create WebGPU Context Provider main @ deathray/:9
(anonymous) @ deathray/:113
Uncaught (in promise) TypeError: Failed to execute 'configure' on 'GPUCanvasContext': Failed to read the 'device' property from 'GPUCanvasConfiguration': Required member is undefined. at main (deathray/:17:17)
I switched my primary PC over in 1999, my secondary over in 2001, and haven't looked back since.
You're in a better place today than back then as finding software solutions to common problems is much easier now.
For other OS needs, just virtualize. I started off using VMware Workstation, after a decade (or so) transitioned to Virtualbox, and now today, virt-manager (qemu, kvm through libvirt).
That was about the time of the Redhat 9 to Fedora shift and I felt the pain too. I was using RH9 for the stability, but then switched to Fedora like everyone was supposed to. Well, that was a big bag of something left on the doorstep. I had to spend countless evenings messing with configs to get the system useable; then upgrades would break things. Finally found stability with Ubuntu after a few years, and still use it today, albeit with KDE since Gnome's philosophy of unconfigurable defaults is disappointing.
The account mechanism is just used to scope the devices to „walls“. Otherwise I don‘t mine anything. But I share your sentiment, that this should maybe a self-hostable tool.
My $10 Casio is infinitely a more reliable timepiece. It doesn't need charging so it can be taken off-grid for months at a time, easy to read the digits when half asleep and your eyes aren't working, has a built-in little light if needed, and if for some strange reason I need to replace it, it'll cost me $10.
When I need something more akin to a piece of jewelry, I reach for my Citizen eco-drive ($99), or my Seiko automatic ($150), both of which don't need recharging. ;)
Maybe this pebble will finally fit into a cost-conscious, and reliable model that existing timepieces already fill, but I'm not convinced through the quality of this article.
I like the idea, but in S227(g)(1) - "training shall compensate the server operator for the bandwidth and compute resources consumed" - bandwidth can be defined in finite terms for the size of the data pulled, but "compute resources consumed" is arbitrary.
They were not so poorly paid - I was a senior analyst/programmer (and did some teaching), quite reasonably compensated, and the lecturers would get quite a bit more than me.
But if you want to substitute "established business model" for "corruption", go ahead. I must say that not all of them were bad.
In Chrome on Linux:
WebGPU is experimental on this platform. See https://github.com/gpuweb/gpuweb/wiki/Implementation-Status#... deathray/:9
Failed to create WebGPU Context Provider main @ deathray/:9 (anonymous) @ deathray/:113
Uncaught (in promise) TypeError: Failed to execute 'configure' on 'GPUCanvasContext': Failed to read the 'device' property from 'GPUCanvasConfiguration': Required member is undefined. at main (deathray/:17:17)
reply