That's pretty cool, I had no idea one could run JavaScript scripts on the FlipperZero 🐬 https://developer.flipper.net/flipperzero/doxygen/js.html and access the CLI to run them via a Web interface, no installation needed https://docs.flipper.net/zero/development/cli#method-1-using-flipper-lab
Remote
Utopiah (Fabien Benetou)
@utopiah@mastodon.pirateparty.be
mastodon 4.5.18+glitchPrototyping - Former European Parliament Innovation lab WebXR consultant and UNICEF Innovation Fund WebXR technical advisor
My website https://fabien.benetou.fr my videos https://video.benetou.fr my code https://git.benetou.fr all self-hosted.
481 Followers
277 Following
50 Posts
Joined February 11, 2019
Website:
Stumbled upon the concept of appropriate technology and its pragmatic encyclopedia https://www.appropedia.org
What a fascinating empowering rabbit hole to go down.
I imagine @trevorflowers@hachyderm.io and @olm_e@tchafia.be are familiar with it but just in case!
@ptrpaws@wetdry.world I was chatting with @lvk@mastodon.online about PmOS on Lynx, boot2webXR, Monado, webkitxr and others ways to be more free with XR. Where do you think the gap currently is? Which is the "free-est" headset, rooted or not, to actually enjoy content (ideally WebXR) on? If you were to get hardware or even provide some via e.g. CrowdSupply how would you do it?
Replying to
@troed@swecyb.com @bagder@mastodon.social wouldn't it be helpful to add resources used to get such high numbers? What I mean is that if radically more resources (energy, money, time, etc) was used to get to that result then it's important to mention it to get a realistic picture.
I think "lots of people" are simply tired of marketing pieces that say "it changes everything" rather than "it's an interesting new way to do this kind of work, it's not perfect or free but it can be useful in this specific context", basically balanced viewpoint rather than sensationalist sale pieces.
repurposed blocks from my tangible calendar : X3 stand
Replying to
@swetland@chaos.social what could be interesting is to have a package manager, e.g. npm or pip, with a AI-free option (that could be toggled on by default) that tracks the presence of traces, e.g. MCP configuration, CLAUDE.md, AGENTS.md, copilot commits and prevent from adding a dependency, ideally proposing an alternative that is AI free based on similar tags.
"Big Tech says AI is an important tool to fight climate change, while virtually none of the generative AI driving the boom is actually being used to fight climate change." https://www.foodandwaterwatch.org/2026/04/10/data-center-greenwashing/
RE: @utopiah@mastodon.pirateparty.be
Imagine this current usage in users too young to have proper self control PLUS fueled by greed strategically leveraging incumbency and dark patterns for addiction.
Think I'm exaggerating? Think again https://www.404media.co/microsoft-wants-to-make-people-addicted-to-scout-its-new-ai-assistant-internal-documents-reveal/
Provocative take : I don't think information overload is a novel phenomenon.
The idea : people looking at me giving XR demo don't have to ponder what it is others are looking at inside the headset, they see on my back.
It's v0.1, after reviewing it the in-VR perspective needs to be way WAY bigger.
Replying to
@maxleibman@beige.party it's just ➡️ code all the way down.
Replying to
@bagder@mastodon.social I voted 10 but in retrospect as you are familiar with Claude and such tooling if you already used any of recent model to done some kind of security review then I'd drop to 1 or even 0. From my understanding of recent posts from https://www.flyingpenguin.com it's mostly tooling around but nothing radical allowing to find genuinely new exploit or full chain.
for the next #WebXR workshop, using my new PinePower (Pine64) for fast charging 5 headsets at once 🔌⚡️ (165W)
Replying to
@impactology@mastodon.social Ah! Was discussing the topic just this morning over breakfast as follow up on recent tinkerings https://git.benetou.fr/utopiah/math-interactables thanks!
Replying to
@gwilymgj@mastodon.social @tojiro@mastodon.social 100%
In fact this and an earlier pinned post I had "Naive heuristic : avoiding gadgets or services (free or not) that increase inequality by design, through technology or business model or both, would be a good starting point." makes me think WE should specific list those technologies to explicitly relying on them, boring (to VCs) is the new sexy!
RE: @utopiah@mastodon.pirateparty.be
Yeay, we march forward regardless!
Global growth in solar “the largest ever observed for any source” https://arstechnica.com/science/2026/04/global-growth-in-solar-the-largest-ever-observed-for-any-source/
#solarpunk today!🌞
Replying to
Side-by-side with my already tiny I-CLIP wallet. The X3 is slightly larger but also twice thinner.
It just does not add up.
Those AI companies are supposedly full of 1000x engineers thanks to their powerful tool. The brightest researchers with supposedly STOA security models. Yet they f*ck just like everybody else, not replying to disclosure. Why?
Replying to
Donated https://app.royalty.dev/crosspoint-reader/crosspoint-reader support free software and open source developers to make your devices truly yours.
AI industry in 1 comment : "The thought behind it was great, but the execution was proving difficult"
From https://finance.yahoo.com/sectors/technology/articles/starbucks-quietly-retired-its-ai-agent-just-months-after-deployment-after-it-miscounted-coffee-shop-inventories-and-slowed-down-baristas-192259953.html as an "anecdote" of a deployment of a Nasdaq-100 company present in 87 country.
No big deal.
Replying to
A romantic view : "power no longer lies solely in invention, but in controlling the channels of passage."🤔
It NEVER did. That's the founding myth of the Silicon Valley where innovation trumps control.
Sadly it has ALWAYS been about capturing resources, funding projects consolidating that process.🔐
https://www.middleeastmonitor.com/20260503-apple-the-digital-strait-of-hormuz/
Between "The Man Who Sold the World" or "Saviour Machine" Bowie was onto something, more than half a century ago.
SpaceX and the ‘enshittification’ of markets https://www.ft.com/content/f724d500-fd45-4f38-86b8-549b5cae88ba?syn-25a6b1a6=1 "The IPO of Elon Musk’s company is a big risk for the millions of investors who have put savings into passive investing"
By the very lefist FT. ¯\_ (ツ)_/¯
Communicating from the browser to the operating system using URLs and their hash properties.
Replying to
Ah, even works in #WebXR now, really shows the power of interoperability thanks to open source and standards!
Manipulating filename straight from the minuscule e-reader right from the XR headset. Amazing to be able to incorporate a brand new device under 2 hours.
https://video.benetou.fr/w/9ETwbcazGgJUQMNdvftmct
Replying to
@neauoire@merveilles.town nice, reminds me of Turing Tumble
"Within 10 minutes, it had generated a fully functional site for less than 5 yuan (70 cents)."
Crazy right? When you read this you can't help but ponder how "advanced" that whole tech stack is.
Well... except one could do that for free in like 1999 with Geocities. 🤷♂️
https://apnews.com/article/china-ai-us-tech-openclaw-0126a120113a92fa450ecb2e464b35bc
I really get triggered when a Web site is used as an example.
A welcome page is a Web site. A "in construction" page is a Web site. Google also is a Web site. Some of the UI on a space ship it's a Web site (see Dragon).
It doesn't mean anything to help assess complexity of the task.
A kid learning HTML can do a fully functional Web site whereas the (theoretical) 10x software engineer can't make alone the most complex Web site because it takes entire teams to do it.
Replying to
The worst part being... I'm not even ashamed of myself, imagine that!
Replying to
@hugo@social.okeso.eu indeed, it's definitely traveling faster but also it goes hand in hand with better tooling and processes, including mass education.
I think we were always overwhelmed because even "just" being in the forest to hunt or collect berries IS about filtering the noise to get the signal.
So I don't think there ever was a period when we, humans, or in fact any organism, didn't have more information than they could process at ANY moment in the history of Life.
So true it hurts "Having to manually register new domains was something that was really holding back my ability to commit fraud."
From https://news.ycombinator.com/item?id=48031684
Again it's not about the 'potential' of the technology, it's how it's ACTUALLY being used. Today. Against us.
There’s a Class 100 semiconductor cleanroom inside this backyard shed. https://www.youtube.com/watch?v=HfSO-LCKmrA by Dr.Semiconductor
There’s a Class 100 semiconductor cleanroom inside this backyard shed.
Replying to
Also R.I.P. my Corne-ish Zen v2, we had 4 wonderful years together.
RE: @utopiah@mastodon.pirateparty.be
the process of relying on technology boring to venture capitalists, because they are already "too democratized", could prove a good strategy.
What it means to daily drive Linux on a PC, namely a computer that is genuinely yours :
- this morning I made a tool, including its UI, without realizing it.
The console has a background tab to download using a CLI tool (yt-dlp) & the visual interface is using watch to check a file and directory.
Leiden Declaration on Artificial Intelligence and Mathematics https://leidendeclaration.ai
" Mathematicians have a choice about whether and how to adopt artificial intelligence in the conduct of their research. They also have a responsibility to ensure the continued flourishing of the discipline. This Declaration calls upon mathematicians to exercise this responsibility, and provides recommendations for individuals, institutions, government, and industry.
Although we adopt the perspective of mathematical research, much of what we write applies equally to other aspects of mathematics. This includes work in the broader mathematical sciences, education, mentoring, publishing, funding, science policy, and use of mathematics in the wider world. "
Conversation with a 12 y/o today:
Me: what's your favorite App?
Them: AI
- what for?
- homework (shows me how to snap a picture of homework to get it "solved")
- looks like a robot going to the gym on your behalf, are you actually learning?
- sure, the AI teaches me
- does your teacher know?
- nope
DIY amusement park, no entry fee, no age limit, scalable difficulty, ...
Replying to
@dries@social.roeckoe.be yep, had to go that route too. I was even tinkering with the unlocker but with so many warnings and UI in Chinese I went on Discord, tried the right button combination and voila, minutes after it was working flawlessly and in English, way easier ;)
Journée internationale de la femme en mathématiques. A creuser avec
"Matheuses - Les filles, avenir des mathématiques"
https://www.cnrseditions.fr/catalogue/mathematiques/matheuses/
@cnrs@social.numerique.gouv.fr editions.
"club mindset is the product. It’s not technical, it’s not about the model." Davi Ottenheimer https://www.flyingpenguin.com/the-boy-that-cried-mythos-verification-is-collapsing-trust-in-anthropic/
Or how scaremongering from snake oil salesmen has been AI bread and butter for nearly a decade now. Former OpenAI staff rinse & repeat, how unoriginal for masterminds.
The Fear of Dying Before You Become Yourself https://www.dailicle.com/read/the-fear-of-dying-before-you-become-yourself
"People say they're afraid of death, but often what they really mean is dying as the wrong person. What do we think we’re racing toward when we worry we’ll run out of time?"
The Dailicle Desk · June 22, 2026
Replying to
@pozorvlak@mathstodon.xyz That's not my understanding, he does mention AISLE indeed but not solely that, he compares with Opus 4.6 and not spontaneously, simply because Claude itself said it was the process.
Looking for Foucault's Discipline and Punish (1975) adapted, maybe as comics, boardgame or video game, to very young kids, e.g. 7yo. Any recommendation?
" the “primary driver” of these abnormally high failing rates is due to a “vast increase in academic dishonesty” due to students’ usage of large language models, such as Claude, ChatGPT and Google Gemini." https://www.dailycal.org/news/campus/academics/failing-grades-soar-as-professors-see-greater-ai-usage-dwindling-math-skills-in-uc-berkeley/article_16fad0bf-02cb-4b8c-8d88-888ffd9f8608.html
Damn, what a useful technology, we just need MORE of it! /s
"we urge Anthropic and other LLM developers that train models using huge datasets downloaded from the Internet to provide these LLMs to their users in freedom." 👌