Hacker News Reader

Comment by bryanlarsen | Hacker News Reader

RedHat/IBM does have an Illinois presence so will likely be compelled to add it to their distro, and will likely do it through systemd. So to avoid it getting into any consumer distro you'd have to ship a patch to remove it from systemd.

This is probably all completely irrelevant to StageX since it's a distro designed to be used in containers, AFAICT.

View on HN

Fortunately patching out such nonsense is as easy as running a few commands.

Project tracking these laws and providing patches as needed: https://agelesslinux.org/

I run stagex on servers and workstations baremetal, however admittedly minimal. It can run lots of ways.

IIRC systemd already started implementing some hooks for age verification API

systemd added a field where users can add their date of birth. It's completely optional.

It's not implementing age implementation, but can be used as a place to store a DOB in an age verification system.

NB that you could already add your own freeform fields; it just added that field to userdbctl and friends.

AFAIK they were implemented by Microsoft staff.

It was not. It was implemented by Dylan Taylor, an outside contributor who works at a fintech company in North Carolina, and has made it his hobby to submit PRs implementing age verification to all kinds of projects. The systemd maintainer who clicked "merge" (Luca Boccassi) does work at Microsoft.

Sounds like Dylan needs a better hobby.

I would love to understand his motives, because my first impression is not a positive one. This is not freedom-loving behavior of a patriotic US citizen.

how would it work in the age of everything virtual, from containers to VMs, even an OS instance running in a browser, with OS/kernels being spanned in sub-seconds and existing even for just one call with no user interaction. Are they going to inherit age bracket from supervisor/parent process? Which of course wouldn't work for AWS/GCP, and many other cases. It looks to me like it would require something similar to security labels implementation.

Simple. If unconfigured, user birthdate should default to 1970-01-01

No. The stupid-ass law [1]:

> "Internet-enabled device" means a smartphone, tablet, or personal laptop or desktop computer that provides cellular or wireless connectivity, is capable of connecting to the Internet, runs an operating system, and is capable of downloading and running applications.

Enumerate all network interfaces. Are any of them cellular or wireless? No? Then the law doesn’t apply.

[1] https://my.ilga.gov/documents/legislation/104/HB/PDF/10400HB...

Finally, my insistence on wired Ethernet pays off.

It sounds like you could even get away with a wireless USB dongle, as it’s not part of the computer.

There's so much room for interpretation in there. I share my laptop with the kids, so not personal. My other computer is a tower / htpc / 19" / any other form factor than a desktop. It can only connect to a WLAN (which need not have a gateway to the Internet, who knows). It will not download and run applications by itself unless hacked, certainly not any binaries (lol gentoo) because the user-writable filesystems are noexec.

"Runs an operating system" is about the only one that has its box ticked >99% of the time.

Probably stuff it KMS and expose via EC2 metadata...somehow

Maintaining forks is pretty trivial now with AI.

Reply on news.ycombinator.com