anzo

joined 3 years ago
MODERATOR OF
programmer_humor
linux
bioinformatics
yurop
europe
homelab
sorted by: new top controversial old
AstrOS - an immutable, secure-by-default distribution built on Arch Linux and the COSMIC desktop in c/linux@programming.dev
[–] anzo@programming.dev 5 points 2 days ago

Oh! This is exciting, just a few hours ago I wondered if there were any inmutable arch based distros.. I will keep my eye on this project. The idea of having extensions as modules for different user needs has great potential!

I wonder how can someone compatibilize the info from archwiki to work with this mew distro to achieve x-y-z customizations which is, after all, the biggest power arch brings.

How to download Lemmy post, as it is shown in the Lemmy UI in c/webdev@programming.dev
[–] anzo@programming.dev 1 points 2 days ago (1 children)

Fwiw there's SinglePage extension for web browsers. Maybe you can get a headless way to run that... Which frontend would you save from? I'd say i rather prefer to get json data from backend

In a universe of enshitification, Linux is one of the few things that consistently gets better. in c/linuxmemes@lemmy.world
[–] anzo@programming.dev 1 points 2 days ago

Me and rpm-ostree are having a picnic in that landscape 🥰 btw

Is Python really that useless? in c/programming@programming.dev
[–] anzo@programming.dev 1 points 3 days ago

Python is the 2nd best language for any task.

You can learn niche langs for specific cases. Go for it.

Me, a python dev, will find an API to your subroutines.

PS. import antigravity

Recommendations for opensource Jetbrains equivalents? in c/programming@programming.dev
[–] anzo@programming.dev 1 points 3 days ago

There's vscodium too, but the remote extension then is not the same as built in o.g... And oh my pi or any other cli agent would do it.

Else, and keeping it up for having remote editing, I'd suggest emacs doom.

0
new "frontier" model requires 24 GiB NVRAM (research.meta.ai)
 

I don't understand why Meta (of all companies) would do this. But here we are... with it, the ecological footprint of your waifu girlfriend that just so happens to be a professional coder, goes from Y liters of waters and Z Megawatts to the mere cheap use of your local gpu, just like gaming. What a time to be alive.

A cow is more likely to kill you than a shark in c/lemmyshitpost@lemmy.world
[–] anzo@programming.dev 1 points 6 days ago

I would also normalize against their frequency. Because if a tech is not used that often for being considered dangerous then of course it can show a low absolute number of deaths.

But then again is a no brainer to tell that this kind of comparison is ill intended from the get go.

The trope defending nuclear energy has always been "too loud" for my "taste". It's like mainstream politics, missing the point of why something is actually good and then spouting cherry-picking "facts" but indeed doing propaganda nonetheless.

didn't think so in c/comicstrips@lemmy.world
[–] anzo@programming.dev 1 points 1 week ago

The environmental impact can be discussed. Maybe training Kimi 2.7 is like sending a rocket to the moon. But using it locally can be optimized and then is as much as running your GPU. Arguably the impact is similar to gaming then.

(I know this is an unpopular opinion because it reads as defending oligarchs. Sorry!)

The blue collar elite in c/memes@lemmy.world
[–] anzo@programming.dev 2 points 1 week ago

Skill issue, indeed.

A win for privacy in c/programmer_humor@programming.dev
[–] anzo@programming.dev 8 points 1 week ago

Thread locked because it's off-topic to our community. (Although there were some interesting discussions, thanks.)

207
A win for privacy (linux.community)
 

cross-posted from: https://sh.itjust.works/post/64574915

386
The call of CthoolChooChoo 🚆 (programming.dev)
 
508
take a deep underwater breath (programming.dev)
 
A win for privacy in c/programmer_humor@programming.dev
[–] anzo@programming.dev 9 points 1 week ago (1 children)

that already exists, without any LLM. just an extension to the web browser: https://adnauseam.io/

Japanese street photographer, Masayuki Oki, captures the everyday lives of Tokyo’s street cats in funny poses. in c/cats@sh.itjust.works
[–] anzo@programming.dev 1 points 1 week ago

all cats are peautiful

Massive Attack Banned from Singapore After Displaying Palestine Flag Onstage in c/world@lemmy.world
[–] anzo@programming.dev 1 points 1 week ago

Singapore recognized the Palestine nation?! /S

10
Open weight models reviewed (wagtail.org)
 

Apparently Gemma 4 (by deepmind) are very energy-efficient for their capabilities.

The post itself seems to recommend a company (neuralwatt.com) but it's unrelated to project wagtail (an open source CMS made with django) so I consider it legit, they're just vouching for this new approach that shows users their ecological footprint basically.

(Of course, running these at a homelab is possible. hence the relevance of this info for the community)

-3
FitMyLLM — Independent benchmarks for self-hosted AI (www.fitmyllm.com)
 

cross-posted from: https://programming.dev/post/51407459

Check what can you use and at what rate of token per seconds would it be... It has examples of many models and quantization levels. Huge resource!

-11
FitMyLLM — Independent benchmarks for self-hosted AI (www.fitmyllm.com)
 

cross-posted from: https://programming.dev/post/51407459

Check what can you use and at what rate of token per seconds would it be... It has examples of many models and quantization levels. Huge resource!

-2
FitMyLLM — Independent benchmarks for self-hosted AI (www.fitmyllm.com)
 

cross-posted from: https://programming.dev/post/51407459

Check what can you use and at what rate of token per seconds would it be... It has examples of many models and quantization levels. Huge resource!

8
FitMyLLM — Independent benchmarks for self-hosted AI (www.fitmyllm.com)
 

cross-posted from: https://programming.dev/post/51407459

Check what can you use and at what rate of token per seconds would it be... It has examples of many models and quantization levels. Huge resource!

-5
FitMyLLM — Independent benchmarks for self-hosted AI (www.fitmyllm.com)
 

Check what can you use and at what rate of token per seconds would it be... It has examples of many models and quantization levels. Huge resource!

11
GitHub action to quickly decide how much to invest on user's contributions via Pull Request(s) (github.com)
 

Examine a GH user's profile, to help quickly decide how much to invest in their contributions. - ehmatthes/gh-profiler

29
tangled: The next-generation social coding platform (blog.tangled.org)
 

I've just found the project. And it seems very interesting take. I am adding a link to their blog entry discussion on a system to combat LLMs abuse via a web of trust.

From their site:

We envision a place where developers have ownership of their code, communities can freely self-govern and most importantly, coding can be social and fun again.

AT Protocol enables federated code-collaboration. Submit pull-requests or bug-reports to any repository hosted on any server.

view more: next ›