I must say I have little sympathy for those researchers who used the plagiarism machine for their work only to have the operators of the plagiarism machine plagiarise their work and beat them to publish
Remote
Hailey
@hailey@hails.org
mastodon 4.5.18
woof! I am a computer tinkerer, polyglot hacker, music enjoyer, and bike rider. Also a huge nerd. Trans and queer 🏳️⚧️ 🏳️🌈 pfp by https://bsky.app/profile/jankapnoc.bsky.social
3247 Followers
237 Following
50 Posts
Joined November 10, 2022
pronouns:
she/her
home:
naarm / melbourne
rsync was basically done until the maintainer discovered vibecoding
from mine to a friends place in essendon google maps driving directions is a pretty direct rat run through residential streets (which would be quiet if not for all the rat runners), cycling directions is “why not ride on mt alexander rd? look we’ve even highlighted it bike green for you!”
society is currently grappling with such questions as "I want to use the plagiarism machine but in my heart I know it's wrong, how do I justify it to myself?" and "I want to use the plagiarism machine but other people might think I'm bad for plagiarising, how do I justify it to myself?" and don't forgot "there must be a role for the plagiarism machine in education but plagiarism is bad, how do I justify this to my students"
I actually don't have any sex transing prescriptions on the go rn, all my needs are met via the community. Isn't that cool?? You can do this too!
https://mastodon.lubar.me/@ben/117056145634566558
in which the author of xfwl4, XFCE's Wayland implementation, tells us exactly how much his principles are worth https://www.spurint.org/journal/2026/07/llms-and-xfwl4
I got to play with a period accurate pentium 4 machine running windows xp the other night, noodled around in control panel and in ableton live 7 and the thing that really shocked me was how *fast* it was. Obviously loading stuff off the hard drive was quite slow, but once the page cache was warmed up it was a very snappy, very responsive experience.
watching the whole industry jonestown itself is the craziest thing I have ever seen in my career
tech management heads are like "ai is so amazing, this team went from idea to closed beta in three months" like that wasn't just normal velocity for a greenfields project at one point
the other thing about the rsync situation I cannot get my head around is the idea of just throwing out the existing test suite for a stable, mature piece of software. How do you even know that the new test suite checks all the same properties? I cannot see the sense in doing this
working on a theory that the headphone jack being removed from phones was the turning point towards the tech backlash
every time I read someone talking about their "AI adoption journey" I mentally gsub Claude with methamphetamine and the whole text still makes sense
Article posted to discussion site. Commenters suspicious article is AI slop. One person says they like the style. Author shows up, thanks that person for the praise, says they fight LLM accusations most of the time.
On other site, author admits to writing article with LLM.
the way people talk about computer science is like if they thought you get a physics degree to become an electrician
How funny is it that the guy who wrote a book about the god delusion now has chatbot delusion
The latest (yes another) Linux kernel LPE also relies on an exotic socket type, AF_NETLINK. If you lock down allowed socket types to just internet+unix, you are safe.
Unfortunately even rootless podman relies on AF_NETLINK sockets via pasta (user mode networking), so it's trickier if you're using containers. Good time to reconsider if you really need all the extra complexity containers bring.
Replying to
Don't come into my project being like this, ok?
hello I am back I broke my instance's federation in a way where I could post stuff out but not receive anything back! I only clocked it when I was like hmm it seems quiet around here
incredible things are happening in redmond
@osnews@mstdn.social https://mstdn.social/@osnews/116558005801959042
Replying to
The point of most propaganda is not to change what you believe or force a particular point of view on you, but instead to tell you: that thing you know is wrong? There are factors outside of your control. It's not your responsibility that what you're doing is wrong. It's just how the world is. You can keep doing it and that's ok.
The propaganda is there to give people *who want to hear it* an intellectual framework for avoiding cognitive dissonance. And many, many people do want to hear it. It's comforting.
It's small fry in the grand scheme of complaints about GitHub but I really dislike how the compare view has been pushed aside in favour of the new pull request view (which is the compare view with ?expand=true). It's very telling of where the focus has shifted and is yet another nudge in the wrong direction
If you debootstrap a brand new trixie install today, you will get not just a vulnerable kernel but also a vulnerable libssl3t64 missing patches for 6 CVEs including a heap buffer overflow and a use after free.
I'm not really a debian expert but this seems... not.. good?
if you like post-punk you gotta check out the debut album from naarm trans legends LXRP. it is brilliant https://lxrp.bandcamp.com/album/lxrp
bringing this toot back to say that if you have `RestrictAddressFamilies=AF_UNIX AF_INET AF_INET6` in your systemd service (as distros have already been rolling out in services they ship), you're safe from someone chaining an RCE in one of your network services through to today's kernel LPE vuln
Have you checked that your database transactions are doing what you want?
Replying to
@xssfox we need distros for popular lang ecosystems imo. Don't care about being on the absolute latest vite or whatever, 6-12 months old is just fine.
so like what's the story with the copyfail patch on debian? My trixie machine is still vulnerable and there are no updates available
Replying to
@s0@cathode.church check out the service files arch linux ships! They have been on leading edge of systemd for a long time and their config is quite good and less encumbered by legacy than the bigger distros. eg postgresql.service https://gitlab.archlinux.org/archlinux/packaging/packages/postgresql/-/blob/main/postgresql.service
Replying to
@whitequark@social.treehouse.systems firecracker style VMs are pretty alright! You do have the extra kernel boot latency but amortised over a whole job it's not so bad
endless money for bullshit and the already lean community organisations actually doing good work are the ones that have to get leaner. their funding shortfall ($300k) is about the salary of one useless AI guy btw
https://social.chinwag.org/@abc_bot/116684709144339173
Replying to
Replying to
@navi@social.vlhl.dev @dalias@hachyderm.io with systemd you can relax hardening rules for services which need it. Distros are shipping service units with this kind of hardening already and it stops you from chaining a vuln in those services to something like this
Replying to
@s0@cathode.church basically you want to lock as much down as you can for each service without impacting functionality. less kernel surface area exposed = less bugs you have to care about
Replying to
Replying to
@rose we can certainly argue forever about what looks exotic from our personal vantage points
Replying to
@rose ya I said exotic, not obscure. Most programs do not need netlink and it opens up a wide surface area for exploits
Replying to
@xssfox@cloudisland.nz i'm sure 100% of towers would know about it if your phone plan billing had expired
Replying to
@dalias@hachyderm.io hold up, I see the iproute2 shell out. I disagree, I think AF_NETLINK is quite relevant. If you block that address family, you block the exploit. Most programs have no need for that address family, so it's unnecessary exposure. I'll revise my position when I see a poc which does not use AF_NETLINK
Replying to
@timriley@ruby.social is it still possible to access the docs for older versions of the dry-rb gems (like dry-validation 1.0)? The old docs url redirects to the latest on the hanakai website now. One day we'll be on an up-to-date version!!
Replying to
@dalias@hachyderm.io oh there's an even newer one! https://www.openwall.com/lists/oss-security/2026/05/07/8
Replying to
@dalias@hachyderm.io it looks like AF_NETLINK is in the critical path of the exploit poc at least? see add_xfrm_sa, it it responsible for actually writing the shellcode and does so via a netlink socket
Replying to
@rose yes so you relax hardening for programs which need it and keep it locked down for programs which shouldn't ever have a reason to touch it