From now on, anytime someone asks me why I put so much time and effort into this arguably dumb account, I will simply tell them that I do it for Tim.
@lowqualityfacts
And for me...thank you
Showing activities from people in this circle
Stefan Bohacek
boosted
RE: https://mas.to/@markwyner/116360220302747675
My offer stands. If anyone needs a hand finding work, let me know. I’ve already reviewed 9 websites and 4 CVs. Which, by the way, were rad.
I keep seeing those layoffs and it makes me sad. Capitalism is punching down right now. Forcefully. Maybe I can help someone survive it. Hit me up if that’s you.
#FediHire #FediHired #OpenToWork #JobSearch #Design #Development #Community
Are you looking for work? How can I help? If you want feedback on a portfolio, résumé, or whatever…you can call on me. Pro bono.
Lots of folks are trying to find work. It can be a lonely and exhausting process. If I can do anything to make that process easier or more productive and help get you hired, it I’d be honored.
My DMs are open. I got you.
🧵1/2
#FediHire #FediHired #OpenToWork #JobSearch #Design #Development #Community
Mastodon Migration
boosted
Big news folks - we just passed the 1,000 users mark! This is a BFD for a small, yet might platform.
Be sure to invite your friends and let’s head to 10,000!
hypebot
boosted
Blatant abuse of power. They act like we are in Russia, not the U.S.
Djoerd Hiemstra 🍉
boosted
“When it gets down to having to use violence, then you are playing the system’s game. The establishment will irritate you – pull your beard, flick your face – to make you fight. Because once they’ve got you violent, then they know how to handle you. The only thing they don’t know how to handle is non-violence and humor.”
― John Lennon
Anthropic has been growing revenue 10x each year since founded.
- End of 2023 ($100M ARR)
- End of 2024 ($1B ARR)
- End of 2025 ($9B ARR)
and it now looks set to grow revenue by 80x this year which has exceeded their capacity estimates leading to the huge deals with Google and SpaceX/xAI announced this year.
They are actually delivering on the AI revenue hype that OpenAI has been claiming.
https://www.nytimes.com/2026/05/06/technology/anthropic-ceo-ai-growth.html
@carnage4life That's revenue, not profit, and a counterpoint:
Manuel Matuzović
boosted
New client project that was supposed to start this month unexpectedly fell through, so now I have a last-minute opening in my calendar.
If you need a dataviz accessibility, web accessibility, or web/dataviz dev freelancer, I'm available 👋🏻👋🏻
Lisa Melton
boosted
People who voted for Hungary's new government broadly support LGBTQ+ rights - LGBTQ Nation
For auditory output we could play low-quality audio samples to spell the text out, though in that case wherever possible our bootloader should use SFX or prerecorded audio.
Since I'm dreaming lets imagine we voicecast Michaela Swee for this!
As for visual output... We'd need a font mapping char(s) to relatively-positioned images. Making this a monochrome raster font with a single size & style would simplify things.
I see no reason not to include ligatures & proportional text.
2/3!
Under the hood we could make the font a program running on our string-centric CPU, provided we know the available width ahead of time. The trick is to compress it well enough that we can afford to fit as complete a font as possible!
I'm imagining a run-length encoding similar to what the early Pokemon games used.
I've designed GPUs that can some the relative offsets & another to render those glyphs onscreen.
Meanwhile most devs simplify by ignoring all but ASCII, I object!
3/3 Fin!
Manuel Matuzović
boosted
Am I really doing this? Maybe... Not even sure if I can pull this off in about a month or so before CSS Day.
Wouldn't it be cool if you could play cssDOOM during the breaks and try to beat other attendees in a two-player Death Match game.
And I still have to write my talk too...
I found a zero day in a security vendor's firewall software that allows you to remotely crash the entire system by sending it a single malicious packet. Since the firewall is responsible for inspecting traffic prior to the operating system handling it, no ports even need to be open for it to work.
Get yourself paid for your discovery. Microsoft pays people who finds exploits. Often people report this in the wild, and lose out being paid.
Replies: "This model supposedly is safe"
*Looks inside*: Huge pile of web scraped data without consent and ignoring licenses.
Seriously, this is a tragedy. LLMs are useful. You might not be able to train ChatGPT without stealing, but you can totally train a useful base model for task specific fine tuning without stealing. Why is almost nobody even trying????
@lina Which tasks can one make work this way? Serious question.
I like how eyeglasses are really expensive and the official way to adjust them to fit your face is to bend them by squeezing them really really hard, but, not too hard
@StefanThinks 😂 good call.
Recently had one of those immersive "read the entire webcomic from the beginning" experiences with "Drive" https://www.drivecomic.com/ - space opera, often with jokes but sometimes straightforwardly moving, started in 2009 and NOT yet complete but the author posts very consistently. If you were fond of Memnon Vanderbeam in "Starslip Crisis" you may get a kick out of the Fillipods.
@brainwane Thanks for recommending this! I've been binging it since you did, and I'm almost caught up.
d@nny disc@ mc²
boosted
Dude! You have a sweet body!
I wanna open a body shop of my own,
but this town's already got one.
d@nny disc@ mc²
boosted
tell your
audience!
Thanks and
ENJOY!
Alex Akselrod
boosted
LLMs aren't harmful as in terms of the technology is inherently evil. The harm comes from a larger socioeconomic construction. We have no idea what LLM usage would be like right now if this wasn't being forcefully thrust upon people.
I use a deep neural network with Blender's OpenImageDenoiser plugin. That plugin doesn't harm my creativity, it doesn't take away my agency. It *is* a deep neural network.
The problem is we *can't* be divorced from our current socioecomic structure. "Technology isn't inherently good or bad"? Well yes and no: technology may be neutral, but it is suffused in a social environment which is anything but.
Loops has received another NLnet grant! 🥳
We’re thrilled to announce we’ve signed the MoU with NLnet to ship the following milestones:
- Live Streaming
- Username changes
- Creator Collaborations
- Project Möbius
- Direct Messages
- Account Migration
- Embeddable Player
- Accessibility Improvements
- Content Scheduling
- Unlisted & Private Videos
- Video Reposts
& more!
Measuring the Higgs field can be done (indirectly) through the measurement of the interaction of the Higgs boson with itself and has links to the evolution of the universe. This is why di-Higgs production is one of the most important targets of the LHC in the coming years, and this #CMSPaper summarises where ATLAS and CMS together (yes, together 🤲) stand with the data collected up to 2018 https://arxiv.org/abs/2602.23991
