Fabio Maurer

Bipolar, casado, ex-praticante de Jiu-Jitsu, fã de filmes, séries, livros, HQs e musica.

[Ihtesham Ali @ihtesham2005]

Here are 20 tools that feels illegal to know about.

Save this:

[Ihtesham Ali @ihtesham2005]

The most banned video app on the internet was built by the guy who repairs your MacBook.

It's called Grayjay.

This is the app Google removed from the Play Store for allowing offline downloads, YouTube blocked at the API level two separate times, and every walled-garden platform in tech has been trying to kill since 2023. It is still online. It has 500,000 active users.

Here's how it works.

Grayjay is a single video app that pulls content from every major streaming platform at once like YouTube, Twitch, Patreon, Nebula, Rumble, Kick, Odysee, SoundCloud, PeerTube, Bilibili, and Dailymotion.

You open the app, subscribe to the creators you actually care about across any of those platforms, and their new uploads land in a single reverse-chronological feed on your phone.

→ 4K YouTube playback with no ads and no tracking
→ Twitch live streams with working chat
→ Patreon paid content if you log in
→ Every video downloadable for offline viewing
→ Casts to Chromecast, AirPlay, and FCast
→ Background audio and picture-in-picture
→ Imports subscriptions from NewPipe, YouTube Takeout, or any supported platform
→ Syncs across phone, tablet, and desktop over your local network
→ Plugin system that lets developers add new platforms without waiting for FUTO to approve them

Louis Rossmann runs an electronics repair shop in Manhattan and has been fighting Apple in court over the right to repair for a decade.

In 2023 his non-profit FUTO commissioned Grayjay because he was tired of paying to support creators through platforms that could demonetize, deplatform, or delete them at any moment.

Google took the app off the Play Store the day it launched. FUTO responded by putting it on their own website, F-Droid, Flathub, and their own auto-updater. There is no store in the middle. You download it directly from the developers who built it.

YouTube's entire business model depends on you not having a choice.

One repair technician in New York just gave you 11 of them.

[All day Astronomy @forallcurious]

🚨: Renowned AI professor, Dr. Roman Yampolskiy, claims the universe is a simulation created to develop superintelligence — and will "soon be turned off."

Advanced civilizations might run planet-scale simulations to birth, train, and test superintelligent systems. Once the artificial entity reaches its goal or escapes the simulation, the cosmic "plug" could be pulled.

[Darshak Rana ⚡️ @thedarshakrana]

Carl Jung died in 1961 at the age of 85.

But he wasn't wrong about the MATRIX and REALITY when he said: 🧵

[Nav Toor @heynavtoor]

10 free tools built by academic researchers that are better than most paid SaaS.

1. Zotero
Built at George Mason University. Saves and organizes every paper you read.
Beats EndNote ($275).
Site → https://t.co/Hpytdn1mVL

2. PostgreSQL
Built at UC Berkeley in 1986. The database behind Instagram, Reddit, and Spotify.
Beats Oracle at $47,500 a year.
Site → https://t.co/8gGai9kdoG

3. Apache Spark
Built at UC Berkeley in 2009. Crunches huge datasets fast. Used by 80% of Fortune 500.
Beats Databricks paid plans.
Site → https://t.co/exyj7iSWsw

4. NumPy and SciPy
Built by Travis Oliphant and academic researchers. The math engine behind every AI tool you use.
Beats MATLAB at $2,150 a year.
Site → https://t.co/CGVlrRsxnr

5. scikit-learn
Built at INRIA, France's national research institute. The easiest way to train a machine learning model.
Beats commercial ML platforms.
Site → https://t.co/YnqWOM9TCj

6. LaTeX
Built by Leslie Lamport on top of Donald Knuth's work at Stanford. Makes any document look like a real research paper.
Beats Microsoft Word for technical writing.
Site → https://t.co/pY5s6lnLqc

7. Open edX
Built by Harvard and MIT in 2012. Lets anyone run an online university.
Beats LMS platforms charging $10 to $25 per learner.
Site → https://t.co/dMqjXbB5Ur

8. Moodle
Built by Martin Dougiamas during his PhD at Curtin University. Runs the online classrooms at Cambridge, Oxford, and Caltech.
Beats Canvas and Blackboard.
Site → https://t.co/jJMoe6CELv

9. Stanford CoreNLP
Built by the Stanford NLP Group. Reads, tags, and understands any block of text.
Beats Google Cloud Natural Language API.
Site → https://t.co/um0DRKNNYs

10. R
Built at the University of Auckland. The language every scientist, biotech lab, and government uses to do statistics.
Beats SAS at $9,000 a year.
Site → https://t.co/35ilqHO6G1

UC Berkeley built PostgreSQL. Half the internet runs on it.

UC Berkeley built Apache Spark. Half the Fortune 500 runs on it.

MIT and Harvard built Open edX. Governments run their education systems on it.

Stanford built CoreNLP. Every NLP paper used it for a decade.

[Eyewitness News @ABC7NY]

Turkey beats US 3-2 with last-gasp goal as Americans look ahead to World Cup knockout stage https://t.co/Mvkv25PbdA

[Fabio Maurer @fmcmob]

@postdiemologian Felicidades!

[Nav Toor @heynavtoor]

A lawyer in Manhattan gets a 500-page contract. Every clause needs to be searchable. By hand: one week.

An accountant in Chicago gets 200 scanned invoices. Every number needs to land in a spreadsheet. By hand: four days.

A researcher at Stanford has 50 academic papers. Tables, formulas, charts locked inside PDFs. By hand: two weeks.

Every one of them is losing days of their life to copy-paste.

Now meet MinerU.

A free and open source tool that reads any PDF, Word doc, PowerPoint, Excel sheet, or scanned image. It pulls out the text in reading order. Tables become clean HTML. Equations become LaTeX. Handwriting handled. 109 languages.

You give it a 200-page PDF. You get clean Markdown back in 90 seconds.

What makes it different from every other PDF tool:

- Multi-column layouts. It reads top to bottom within each column. Not left to right across the page. Like a human reads.
- Scanned documents. OCR built in. Point it at a photo of a printed page from 1995. Get clean text back.
- Math formulas. LaTeX-quality recognition. Every equation renders correctly.
- Tables. Merged cells, multi-row headers, tables that span three pages. All preserved.
- Ten-thousand-page documents. Sliding window processing. No manual splitting.
- Batch mode. Point it at a folder of 500 documents. Walk away.

Three ways to use it:

- CLI. One command per document.
- Python SDK. Five lines of code.
- Web app at https://t.co/AIC2NNey41. Upload, click, download. No install.

Plugs into Claude Desktop, Cursor, Windsurf, LangChain, LlamaIndex, RAGFlow, Dify, and FastGPT. Feed extracted documents straight to your AI agent.

The story:

The OpenDataLab team at Shanghai AI Laboratory needed to extract clean text from millions of scientific documents to train a language model. Existing tools failed. They built their own. Then they open sourced it.

68,551 stars. MinerU Open Source License, built on Apache 2.0. Free for personal and commercial use. Three technical reports on arXiv.

Adobe Acrobat Pro charges $239.88 a year. It still loses your tables.
ABBYY FineReader Corporate charges $165 a year. It still cannot do equations.
Mistral OCR charges $2 per 1,000 pages. Your bill never stops.

MinerU costs $0. Runs on your laptop. Your documents never leave your machine.

Here is the wild part.

The lawyer got her contract back in 4 minutes. Every clause searchable.
The accountant fed 200 invoices in. Every number landed in a spreadsheet in 12 minutes.
The researcher fed his 50 papers in. He wrote his literature review on a Sunday afternoon.

The document your company has been processing by hand for years takes MinerU minutes.

Your documents become text. Your text becomes data. Your data becomes answers.

The week you used to lose to paperwork is back in your hands.

[Nav Toor @heynavtoor]

10 single developers who built free tools that Big Tech tried to kill. And lost.

Bookmark this list. These 10 people built things you use every week.

Companies worth trillions of dollars have spent over a decade trying to make them disappear.

Every one of them is still shipping today.

1. uBlock Origin

[Fabio Maurer @fmcmob]

@cabanadrives6 Obrigado!

[Fabio Maurer @fmcmob]

@NationalTheatre I'm waiting for Mnemonic.

[Ram V @therightram]

Peep the interiors too, if you haven't.

It's a big ambitious project. I hope you'll all support us now at the beginning of it all <3

Out from @ImageComics - this October.

[Ihtesham Ali @ihtesham2005]

Jensen Huang said fears about AI destroying jobs are nonsense driven by people making things up. Then he gave the most concrete proof I have ever heard.

He called it the radiology story.

12 years ago one of the world's most respected computer scientists made a prediction. Computer vision had become superhuman at reading medical scans. It never got tired. It never missed a detail. So he said radiology was finished. He advised nobody to enter the field. The job was going to be wiped out.

Jensen said he was absolutely right.

Computer vision did penetrate every single radiology stack. Every radiologist today is augmented by it. The task of reading scans became automated.

Then the opposite of the prediction happened.

Radiology demand went up. The number of radiologists in the world went up.

The reason is the distinction Jensen kept coming back to the whole talk. Task versus purpose. The task of a radiologist is reading scans. The purpose is working with doctors to diagnose disease. When the task got automated, radiologists got more productive. More productive meant more patients admitted. More patients meant more scans. More scans meant more profit. More profit meant they hired more radiologists.

The field the expert said would be wiped out needed more people than before.

Jensen's point was not that AI creates no disruption. His point was that people confuse the task with the job and then make predictions that cause real damage. Fewer people trained in radiology because of that speech. Now there is a shortage.

Someone recently said 90% of software coding will disappear. Jensen said Nvidia is hiring more software engineers than ever.

Coding is not a software engineer's job. Solving problems is.

When you automate the task, you free the person to do the actual job.

That is not a job loss. That is an upgrade.

[ryu & that @tyskei]

She saw at least TEN children under the age of FIVE with their heads DECAPITATED. We are not mad enough.

[Global Insight Journal @GlobalIJournal]

This was HEARTBREAKING

Dr. Tanya Haj-Hassan, who volunteered in Gaza, exposed Israel:

"I held a lifeless child in my arms. There was no equipment to save him. This is not a war; it is a massacre of the innocent." https://t.co/KUrpZ6wBqj

[Global Insight Journal @GlobalIJournal]

💔 Alma Abu Rida...

She was an infant only 4 months old, suffering from a severe lung disease and recurrent respiratory failure, in urgent need of evacuation and treatment outside Gaza to save her life.

Appeals were launched, calls were made, but help never reached her. A new name added to the list of children the world did not look at for too long.

Alma passed away...

She passed away before receiving the treatment that could have saved her.

May Allah have mercy on Alma, and may her story be a testament to the suffering of children waiting for a chance to live.

[Kinen @JetKinen]

New weekend, a new episode of #SongofTheSamurai is here! Really, only one fight in this episode, but this is the most interesting one since the first episode. Kensuke Sonomura uses the low celing on the tatami floor to make a sword duel where fighters scurry around on the floor.

[Dr. William Li @drwilliamli]

🩺 200 cancer patients. Only 20% survived immunotherapy. The bacteria they had in common: Akkermansia muciniphila.
See how it works ⬇️

#GutHealth #Microbiome #CancerPrevention

[Ihtesham Ali @ihtesham2005]

A guy named nbatman on Reddit accidentally built the most useful website on the internet.

It's called FMHY (Free Media Heck Yeah).

This is the website Google delisted from search for DMCA violations, Reddit shadow-banned for promoting piracy, the Motion Picture Association flagged as a top piracy threat, and the RIAA pressured hosting providers to drop. It is still online. It is still updated every month.

Here's how it works.

FMHY is the index. The wiki itself hosts nothing. It just tells you where every free thing on the internet actually lives, organized into 14 categories with safety ratings on every single link.

→ Movies and shows in 4K from 50+ streaming sites
→ Music at Spotify and Apple Music quality
→ Adobe Creative Cloud, Microsoft Office, AutoCAD, JetBrains
→ Every paid course on every major learning platform
→ 100 million books and papers through Anna's Archive
→ Free alternatives to every paid AI tool
→ A SafeGuard browser extension that flags unsafe sites in real time

It started as a single Google Doc maintained by one Reddit moderator in 2018. Google killed it with a DMCA takedown in 2023.

The community rebuilt the wiki on its own domain, mirrored it to GitHub and IPFS, and now runs it across 12 backup domains simultaneously.

There is no company. No CEO. No central server. Six anonymous volunteers maintain the entire thing in their spare time. Donations through Ko-fi pay for the hosting. Nobody profits.

Hollywood can't shut this down. Spotify can't shut this down. Adobe can't shut this down.

The entire subscription economy is held together by you not knowing this wiki exists.

https://t.co/AAr2rLlqgy