sourcelocation

Developer. Christian.

[sourcelocation @sourceloc]

1 year ago, I wanted to play GeoGuessr. But it was behind a paywall, and I was too cheap to pay.

So I thought, screw it. I'll make my own.

Today, GeoDuels has 23k players, hundreds of maps, and even tournaments.

And it's still 100% free.

Here's how it all happened 🧵

[sourcelocation @sourceloc]

@boredcoder411 but is it memory safe????? does it run blazingly fast????? does it use rust??????????

[sourcelocation @sourceloc]

@thiojoe What was the reply?

[sourcelocation @sourceloc]

@facedevstuff @Anthropic uhhh i think you tagged the wrong dude

[sourcelocation @sourceloc]

@boredcoder411 @grok i just ran this code on my computer and it yielded result "67" in stdout, does this line up?

[sourcelocation @sourceloc]

Who’s gonna tell Asahi Linux, that if you make your code open source, forking it doesn’t mean “exploiting a license” and “hijacking work”?

they’ve lost the plot… 💀

[The Lunduke Journal @LundukeJournal]

Asahi Linux, which works to enable support for Apple Silicon Macs on Linux, is dropping the MIT license… and moving to the GNU GPL.

Why? Because, according to their statement, of “bad faith actors” who “exploit” open source code by… using it in other open source code.

“We were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon.”

This licensing change (MIT to GPL) is happening immediately following Omarchy adding support for Apple Silicon, building upon the Asahi work with the usage of LLMs.

And Asahi does not approve of AI usage. At all.

“We broadly forbid the use of generative Al,” says the Asahi AI policy. “A contributor using an LLM … may be issued with a first and final warning. A contributor found to have concealed extensive LLM use will be banned immediately.”

[sourcelocation @sourceloc]

@dhh However Omarchy probably won't even feel the change:

Have one agent document the original's behavior, then another with fresh context rebuild it from that spec. You can get surprisingly far. WineHQ has essentially done this with Windows libraries for years, just using humans.

[sourcelocation @sourceloc]

> Omarchy releases support for ARM Macs
> 2 days later Asahi Linux conveniently switches to GPL license

rather than improve the project, they sabotage the one that’s actually moving. stagnant for years and still acting like someone cares.

[sysxplore @sysxplore]

Asahi Linux is moving everything from the MIT licence to GPLv2-or-later.

Asahi is the volunteer project that reverse-engineers Apple Silicon Macs so they can run Linux. In today's Linux 7.3 progress report, developer James Calligeros explains why: people have been using the permissive licence to take the team's work. "In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin."

The report points to WINE, which switched from MIT to LGPL in 2002 for the same reason: "Those who do not learn from history are doomed to repeat it."

The same report brings real wins too. M3 Macs now install without the EXPERT flag, Thunderbolt support (USB3 and networking for now) is shipping to users, and the team got GTA IV and Fallout: New Vegas running outside the muvm VM using Google's per-process page size patches.

[sourcelocation @sourceloc]

By switching to GPL, they made it impossible for Omarchy (MIT) to use their code.

Omarchy is made by @dhh, a known developer who’s been a target by many left-leaning people for his beliefs.

(yo dhh give me a rt)

[sourcelocation @sourceloc]

@boredcoder411 @facedevstuff i just yap a lot

[sourcelocation @sourceloc]

@facedevstuff there’s place for one and only one editor

[sourcelocation @sourceloc]

@facedevstuff FaceDev becoming more based by the minute

[sourcelocation @sourceloc]

@boredcoder411 gang don’t make me hide your tweet i need a repost from dhh pleaseeee dhh i need this

[sourcelocation @sourceloc]

@BaconManiaGD i had one very private project leaked too💀✌️

[sourcelocation @sourceloc]

They forgot one training data source:

Your projects as .zip files, “accidentally” sent to a Chinese IP in the background.

[bebu @bebuu_ai]

https://t.co/UStDmf5MNx chief scientist Jie Tang just revealed the training pipeline behind the upcoming GLM-5.5 (32B, 122B, and 256B) during a lecture at Tsinghua University today.

Instead of relying on public datasets, their pretraining pipeline pulls from six separate internal engines:

- Web: 1 trillion pages crawled at ~10B pages/day with PB-scale template clustering
- Documents: 2 billion scientific PDFs parsed through GLM-OCR (4x faster with speculative decoding)
- Books: 50 million books with layout reordering and garbled-text filtering
- Code: 500 million repositories turned into executable tasks with real tests and issue descriptions
- Video: 10 billion videos for long-sequence world knowledge
- Audio: 300 million hours rewritten via ASR into trillions of tokens

[sourcelocation @sourceloc]

lease your RAM to Sam Altman immediately. stop wasting your RAM. make the world a better place.

[Stefan Esser @i0n1c]

Yeah they are right 2 GB of RAM in HomePod mini 2

[sourcelocation @sourceloc]

@withaker0 vibecoding began when gpt-1 was made.

[sourcelocation @sourceloc]

the research: a dude who vibecoded a notch app and charged $4.99 for it since 2021

[Guilherme Rambo @_inside]

Of course I had to virtualize one of the new PROLED MacBook Pros. The first thing I wanted to see: how the Dynamic Island responds to the mouse cursor.

[sourcelocation @sourceloc]

@bomberfish77 just sell the checkmark bro

[sourcelocation @sourceloc]

@bomberfish77 stop larping, elon did not pay you to say this

[sourcelocation @sourceloc]

people keep forgetting just how long GPT-4 and GPT-4 Turbo stayed at the top:

[Anthony Kroeger @kr0der]

is Opus 5.5 the longest time that a model hasn’t been dethroned?