Mastodon Feed: Posts

glyph ("Glyph") wrote:

@isagalaev because it's not my fault but it is my problem? the slop comes in anyway and I have to deal with it.

on the one hand I don't want to get overconfident in my ability to detect LLM output when there's plenty of evidence that this is very hard and one can easily get lazy and start deciding stuff is AI when it isn't. on the other hand I don't want to reject people, I want to reject methods

Boosted by neatnik@social.lol ("Neatnik :prami:"):
mustastum ("Enes") wrote:

Made a fun little dancing omg.lol (@neatnik) #animation over @codepen with #zdog

Live: https://codepen.io/editor/pavlovsk/pen/01a0d01a-6524-79e0-a7b1-e94cc64c63f8

Attachments:

• gifv: omg.lol's logo (pink smiling heart) is dancing around the clouds in the sky

isagalaev ("Ivan Sagalaev :flag_wbw:") wrote:

@glyph agree. But then again, as a maintainer, why make it your problem and not theirs? If you put it like "I can't accept code you don't understand" instead of "I randomly reject your code editing technology", it should be pretty clear. This is why I think most of them actually should know/suspect why their PRs might be rejected.

Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
kurrikage@meow.social ("Kurrikage :deifirev:") wrote:

Get the good side~

🐾 @aryrinn

#BatPics #Fursuit

glyph ("Glyph") wrote:

@isagalaev so they don't feel like norms are changing, they just see a new product, and not a new, defective, extremely dangerous product, and they treat it like any other new product that might improve their workflow. the nature of the product is that it obscures their own perception of their work, with an obstruction that they cannot perceive and is extremely tedious and subtle to explain to them

Boosted by cstanhope@social.coop ("Your weary 'net denizen"):
freebliss@post.lurk.org ("Simon Repp") wrote:

Incredibly excited to announce:

Faircamp 2 is now officially available!

«Nine months of work. Over a hundred thousand lines of code revised, added or deleted. Version 2 is the biggest Faircamp release to date and brings Faircamp from the console to the desktop, introducing the new graphical site editor, a fundamentally overhauled architecture and dozens of bigger and smaller new features.»

https://faircamp.org/changes/2.0.0/

This release would not have been possible without the kind support of the @netidee grant programme, who funded the majority of this work – thank you so much!

Finally, thank you to all supporters and the over 30 contributors to this release – your support made all the difference as well. <3

Faircamp is a static site generator for audio producers. To learn more or download faircamp, check out: https://faircamp.org

#faircamp

glyph ("Glyph") wrote:

@isagalaev I really don't think that this is the case. I have very strong views on this (obviously, I hope?) but there are also several very different camps here, who have pretty markedly different perspectives. There *is* absolutely an "accelerationist propaganda" element at work, but also, many, many developers see "I submitted work with Claude Code" as much more analogous to "I submitted work with Eclipse IDE" than "I potentially plagiarized this work"

isagalaev ("Ivan Sagalaev :flag_wbw:") wrote:

@glyph basic social norms don't change over four years.

isagalaev ("Ivan Sagalaev :flag_wbw:") wrote:

@glyph it's not a consensus etiquette, it's what accelerations are trying very hard to make you believe it is.

If I were still an active maintainer these days I wouldn't even bother with explaining anything but would just close slop contributions outright. Trust me, they know why, even if they act indignant.

Boosted by cstanhope@social.coop ("Your weary 'net denizen"):
sundogplanets ("Prof. Sam Lawler") wrote:

Currently watching an IAU-CPS talk given by Olivier Hainaut.

His simulations show that 30,000 sats at 350-600km causes the same astronomy research data loss as 6,000 sats at 1200 km. This is because higher altitudes are visible for a larger fraction of the night, and the sats move more slowly across the image so they cause brighter streaks even though they are farther away

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
erininthemorning.com@bsky.brid.gy ("Erin Reed") wrote:

This is what happens when you get loud and demand Dems not drop trans people. We CAN get orgs to realize this.

But we can’t stop with Kaptur. She’s the worst but not the only one.

There is NO acceptable anti-trans vote or statement or ad from a Democrat. None.

GET LOUDER!

RE: https://bsky.app/profile/did:plc:m65ifh7vn5zdgs7izcmht4gy/post/3mwbprwldic2q

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
heyheymomo@socel.net wrote:

Better be safe than slightly thirsty

cwebber@social.coop ("Christine Lemmer-Webber") wrote:

I'm sorry, goose. We do value you, you know.

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
joeyh@sunbeam.city ("see shy jo") wrote:

@be @cwebber debian.net is for web sites that are ran by single debian developers, not by the project as a whole. Any DD can make a debian.net site. Hell, I still have a debian.net site I happen to control and I left the project over a decade ago ... oops.

However, salsa integration I think speaks to this being more widely part of debian than some random developer's site. debian.net is also sometimes used to make things not look "too" official.

Boosted by cstanhope@social.coop ("Your weary 'net denizen"):
chessert@mastodon.online ("Chris Hessert") wrote:

At the time (1970s), I never imagined I'd become nostalgic for the days of synthesizers.

They burst the doors open for fantastic audio effects, along with all the electronic innovation involved in creating various guitar pedals and effects. But the synthesizer seemed to roam in the range of advanced keyboardists?

Has anyone written a history of these innovations, and the music they helped create? I'd be fascinated.

#Music #Synthesizer #Electronics #Effects #1970s

dysfun@treehouse.systems ("gaytabase") wrote:

Is really impossible?

...

we tried VERY hard

glyph ("Glyph") wrote:

I realize I'm in a strong minority here but I just _cannot_ perceive submitting LLM-generated work to a volunteer project as anything but an extremely rude and thoughtless thing to do.

I _know_ that the consensus social etiquette is that I'm supposed to have a prominently-posted policy statement that this now-generally-accepted-as-normal thing is actually considered bad in this specific place, but it remains jarring for me every single time.

dysfun@treehouse.systems ("gaytabase") wrote:

holy shit someone made windows boot virtualised under sel4. so like that's an option.

jscalzi@threads.net ("John Scalzi") wrote:

Hey, kids! Don't trust a single goddamn poll! If you're a US citizen of voting age, register to vote if you haven't, check your registration if you have, and make a plan to vote, preferably in early and preferably in person if you can. Don't take anything for granted this year.

https://www.cnn.com/2026/09/23/politics/cnn-poll-donald-trump-economy-midterms-vis

dysfun@treehouse.systems ("gaytabase") wrote:

what if MovieOS but secure?

dysfun@treehouse.systems ("gaytabase") wrote:

can you imagine a desktop operating system with native capability-based security? i'm not sure i can

dysfun@treehouse.systems ("gaytabase") wrote:

honestly wondering what it would take to make this a desktop operating system. i think the main obstacle is graphics. you can run linux drivers unmodified, but it will be slower than native.

dysfun@treehouse.systems ("gaytabase") wrote:

i can't be fucked to reproduce all the graphs but suffice to say the reason it does better is because it's using less cpu.

dysfun@treehouse.systems ("gaytabase") wrote:

just for having the safety of a microkernel, you just have to sacrifice a bit of performance - that's actually not true

pixxl wrote:

@dysfun @niconiconi what the php

fribbledom ("muesli") wrote:

"I’ve written 160k lines of code in the last month and I haven’t read a single line of it."

Dude, that's really not the flex you think it is 🫣 We can tell.

Boosted by andrewnez ("Andrew Nesbitt"):
andrewnez ("Andrew Nesbitt") wrote:

Package Manager Sandboxing

https://nesbitt.io/2026/09/24/package-manager-sandboxing.html

andrewnez ("Andrew Nesbitt") wrote:

Making plans for the @fosdem 2027 package manager dev room 📦️

dysfun@treehouse.systems ("gaytabase") wrote:

so we can plug in these linux driver VMs transparently and the rest of the system doesn't notice anything except performance will be worse

dysfun@treehouse.systems ("gaytabase") wrote:

i dunno that kinda feels like the future. sooner or later surely we will get over this monolithic kernel thing.