It would take around 1.1 million acres of solar panels to (clean) power the world. That's less than a quarter of the number of acres devoted globally to golf courses. You can see where this is going, right...?

There is zero value in the existence of an entirely AI generated video social network. Literally only bad things will come of this.

NEW: 18,500 children have been killed by Israeli forces in Gaza. The Washington Post has just published every single known name. A seminal moment.

Se llama Quantus Skin y forma parte de una inversión de 1,6 millones de euros en automatizaciones de Osakidetza. Los especialistas la critican por sus resultados "pobres" y "peligrosos". Además fue entrenada solo con pacientes blancos. ‼️ No detecta 1 de cada 3 melanomas.

🚨 70 million US workers are about to face their biggest workplace transmission due to AI agents. But nobody’s asking them what they want. While AI R&D races to automate everything, we took a different approach: auditing what workers want vs. what AI can deliver across the US workforce.🧵

hi everyone! i am looking for my next role as a design engineer / frontend / web stuff. i am experienced in nextjs, and particularly, sveltekit. specializing in novel, fast and fun interfaces. please feel free to DM! you can check my work in my website at jesirgb[dot]com ↓

@simonwillison.net in case you weren't aware, many of your posts are duplicated on people's timeline. Prolly some automation thing...

@simonwillison.net in case you weren't aware, many of your posts are duplicated on people's timeline. Prolly some automation thing...

It's weird how everyone in the world knows a Nazi salute when we see it, with the notable exception of journalists and their editors.

Wow: "our screening analysis demonstrates that driven by the growing demand for AI, the U.S. data centers could contribute to, among others, approximately 600,000 asthma symptom cases and 1,300 premature deaths in 2030, exceeding 1/3 of asthma deaths in the U.S. each year." arxiv.org/pdf/2412.06288

Really smart and good piece by @janus.bsky.social about steps you can take during this time to build power that are not posting www.404media.co/you-cant-pos...

OK, I thought this was pretty clear, but to spell it out: we've gone from gatekeepers that, for all their blind spots, operate in public & are subject to some accountability... to algorithmic gatekeepers that are opaque, unaccountable, & designed by a tiny, blinkered class of dudebro dipshits.

La semana que viene (miércoles 22, a las 19h, Ateneo de Madrid), estaremos contando por qué -y para qué- en @civio.es hacemos lo que hacemos. ¡Venidse!

I've made this citation network of ~700 papers mentioning "churn" and "customers". I've also included the top 100 papers citing or cited by those papers: dev-embeds.graphext.com/5a5ca9660dab.... You can search and filter in any of the 54 variables related to each paper #AppliedDS

apart from the insane work from Thomas gathering and tidying all this info, how insane is it that you can check all this data ON YOUR PHONE? no lag whatsoever experience what truly good engineering feels like @graphext.bsky.social

Ecosystem of 350k models and datasets on @hf.co's hub, connected to other models or datasets they were derived from (where that information exists, which is not the case for another 1M entities) #databs public.graphext.com/d9275e17cc44... Color, filter and search within each of the 30 variables...

Figured it would probably reach more people if I actually tagged the #databs community 🤦‍♂️ If you know anybody working in and posting about #AppliedDS (real-world use cases rather than tooling) please let me know and I'll add them to the starter pack

This post by Simon & @mehdio.com features lots of community members who are building things! @ssp.sh @jakthom.bsky.social @davidgasquez.com & others. If you are building something cool on top of bsky data, would love to hear from you #databs

we had this ongoing problem of going to [LLM of choice] to ask "google search queries for this product, please return in this format and do not do this" why not bake it directly in here with a couple of options? #webdev #svelte

Of course, you can explore the graph and find yourself in it! I've only included folks that have sent at least one #databs post (that excluded me 😜). Will probably redo it in a week or so! Any thoughts / feedback? public.graphext.com/2b808d92830c...

Is there something like an "Applied ML" community, of people not working on tools (whether it's AI, "agents", data warehouses, etc.), but solving specific real-world business problems? #databs

#dataBS In case there are still people doing ML... If you model something like churn as a classification problem, what tricks do you know of to deal with potentially changing covariate distributions? Apart from using the obvious windowed/lagged features, time-base train/test splits or CV etc.

Write a small and rough post on learnings and the process I took to export 5000 datasets (200GB) from the Spanish National Statistic Institute into 500MB worth of Parquet files on HuggingFace. davidgasquez.com/exporting-in...

This paragraph was a hell of a ride

We need a @duckdb.org community extension that reads `at://` protocol, please :) Or is there a workaround, anyone? :) Protocol: docs.bsky.app/docs/advance...

This week on Counting Stuff, I couldn't help but write about how Data Bluesky... DataBS hit some magical tipping point and may be a thing now... also with lots of links to resources and stuff www.counting-stuff.com/data-twitter...

Time for some shitposting about dashboards, since that’s in the zeitgeist: I call many years ago an eager IT manager telling me we should deprecate all of our traditional SSIS reports. Because all that info could be revealed in a dashboard. And dashboards allow “insight discovery” which is 🔥 …

I’ve created a small Python script to extract the social graph of followers and followings for any public Bluesky account. You can upload the output CSV to a free @graphext.bsky.social account and create an interactive project like this one of my followings public.graphext.com/c8f47a5e19c3...

Seems the 𝖧𝟤𝖮.𝖺𝗂 database benchmarks have been updated! duckdblabs.github.io/db-benchmark/ As also, impressed by ClickHouse and @duckdb.org speeds.

I made this starter pack as well. Let's combine them? go.bsky.app/7iwLQq5

Please like or reply and I’ll add you to this list Trying to build an AI/ML list Also share! Trying to get everyone on here bsky.app/profile/did:...

A couple new-to-me 🦋 features that rock: + Ability to interleave results from your custom feeds into your following feed (Settings --> Follow Feed Preferences --> Show Posts from My Feed) + DMs! + Ability to mute words and tags (Settings --> Moderation --> Mute words and tags) + Thread preferences

“We shall spend an incredible amount of money building a data warehouse and a data science team to generate business insights. Everyone will do this.” —Sir, I assume these data teams will be trusted and their data used for important decisions “No, decisions will be made primarily on vibes”

Lots of Data and AI people joining Bluesky, I am afraid I will have to split the starter pack as it is already crowded with 150 people go.bsky.app/7D4NApV #rstats #ai #datascience #bayesian

this is a bloody great name for a climate report www.unep.org/resources/em...

When oil firms lead climate talks 🤦: phys.org/news/2023-10...

"Companies fucking around leaping blindly into the AI hype bubble and finding out they don't have even a tiny fraction of the twisted imaginations necessary to understand what they're unleashing" is my new favorite headline genre

Nope.

It must be really strange to be a kid learning about the Civil Rights movement and see America doing the same thing to trans people but your teachers just ignore the connection in classes.

TLDR: We are excited to share that @graphext.bsky.social has just closed a $4.6M seed round, lead from Hoxton Ventures and 80 angels, bringing our total seed funding to $7M 🎉 https://www.graphext.com/post/graphext-raised-4M-seed-round