Oct 6, 2026, 8:08 PMSchöne Geschichte.
"But the VIC-20 is the one that counts. The family has been uprooted twice. The country is still foreign, and the land outside is as hard and clean as a sheet of ice. Economies collapse and people fail you, but the machine is exact. Understand it down to the last register and it will never betray you.
For a boy teaching himself to program alone in the northern dark, correctness isn't a matter of taste. It is shelter."
Oct 6, 2026, 8:12 PMIch habe auch einen kleinen Puffy zu Hause 🥰
The attack depended on a chain. Linux distributions had patched their SSH daemons to talk to systemd, systemd pulled in liblzma, and liblzma carried the poison into the daemon's memory. OpenBSD had never let systemd in.
Wait, does this mean much of Gentoo was similarly unaffected?
"Especially with compressors like XZ," he said, "which had 2 memory system vulnerabilities in the last year…"
I have been complaining about those and more particularly media codec implementations for a while (libwebp was not a surprise to me).
So much C, and much of it not of particularly good quality either.
(And of course they don't do secure C, it's a hassle. But ECL has been around for a while now, if they really need C compatibility.)
Before the ensloppening I had the idea of making an image viewer & video player in pure Java (including all the codecs), but given what's happened to its ecosystem that idea is rather harshly shut down (I was planning on using existing implementations).
And the Linux people will just stick with it and add to it rather than stepping back and saying, 'This is garbage and we should fix it.'"
Quite.
Then there were the blobs. Modern Wi-Fi chips, GPUs and network cards often work only with secret firmware, called a "binary blob," which is loaded into the device and given high privileges without anyone outside the vendor seeing what it does. Linux mostly accepts these blobs so that laptops can connect to the coffee-shop Wi-Fi.
The shaming worked. Several major Taiwanese wireless chipmakers opened their documentation. Free drivers followed, for OpenBSD and for every other open system, Linux included.
I think the article is confused. Firmware != drivers. In theory firmware can be constrained by IOMMU, drivers in monolithic kernels do indeed have ring0 privileges.