julianlam

Why Tangled instead of something more established like Codeberg, or if f/loss, Forgejo or Gitea?

Just because ATProto vibes?

Forever soup isn't new, of course.

Poor families do it to cheaply make long lasting meals.

My late maternal grandmother used to have a pot of forever soup on the stove, and she would put whatever she had on hand in.

Spare ribs, check. Leftover veg, check. McDonalds fries, gross, but check.

The floating ring of space debris around Earth would like to have a word with you.

Every time this sort of news pops up, I think "why aren't they incentivising and subsidizing fertility treatment"?

You want to raise the fertility rate, yes?

You could convince an entire population to have more babies the hard way (social or economic pressure).

Or literally just let the people who are already trying, but can't for some physiological reason, have one. They already want children, if they're pursuing fertility treatment, they're already decently well off, too.

To me it seems like a very obvious, targeted solution!

Yeah Hopper didn't start off as a booking site. Just a private group not affiliated with any airline group.

:)

Sounds like Hopper, but AI. Why not just use Hopper?

You wouldn't distill a car.

[citation needed]

What do you need a frontier model for, really.

I think the bogeyman would be making possession of a local AI a felony.

A small 4-9B model would be able to run cheaply for this sort of work.

Then we'll amend the policy to instruct LLMs to author PRs as Fred Flintstone, yabba dabba doo!

Very surprised an Apple device can have some atrocious ventilation design.

I'm running this model on a Framework 13 and the chassis barely heats up at all while running full tilt.

As a FUTO keyboard user for about a year I'm quite excited for this!

Their local voice transcription is top notch and proper swipe gestures would be icing on the cake.

Now to fix that issue there tapping "Tuesd" sometimes suggests "Thursday"

I think it's interesting that people write off open weight models because they're "a few months behind" proprietary models.

I know LLMs move at the speed of light (especially these past few quarters), but if Opus and GPT "a few months ago" were really like open weight models, then there's really no reason to not switch, especially for those who were using these models a few months ago.

Your codebase didn't change, so use the open weight model. Don't move the goalposts.

I think a particular failing with developers embracing AI is fighting the sunk cost fallacy. While you might not have spent as much time putting together a non-working solution, you still did spend time working with the agent to slap together a non-working solution.

Being able to step back and say "this was a failure and we need to discard the day's work and start over" is still hard with LLMs.

That is a cute sentiment, because I imagine it pales in comparison to the loathing these groups feel towards each other in North America.

Of course.

Qwen 3.6 35B-A3B on a Framework 13 with 32GB of memory.

Running llama.cpp, 15 tokens per second. Outputs code and text faster than I can parse.

"Memory Wall"? My LLM tells me about those all the time!

6 months is a joke.

Companies a thousandth their size are giving free or at-cost access for OSS projects.

s/JS/PHP

Watch the Last Week Tonight segment.

Basically, Ticketmaster owns all the concert halls too.

enough to make it work via profit reinvested from sales.

Yes, this is just called running a business.

Are you sure? Prior to today the sub seems to be pretty partial to Qwen.

Last time I tried Gemma 4 (26B-A4B) its memory usage would balloon and consume all of my swap until my machine died.

Qwen 3.6 on the other hand barely uses any memory at all for its KV cache.

Respectfully, as an OSS maintainer (not to the scale of nginx or valkey, of course)... if a third-party used an AI agent to rewrite my software in a different language, that gives me absolutely no reason to support that new project.

It is in all respects foreign code in a language I may or may not be familiar with, and worse yet, if I were to take over, I'd be responsible for maintaining the whole black box forever more?

Thank you but no thanks.

Model-in-the-middle LLC

Checks out, what's the problem? /s

This type of laziness isn't novel.

Check out left pad or the two dozen other "utility" packages that could be done in a single line of code.

Does this mean a single pair can be adjusted over time to adapt to a changing prescription?

Buying lenses is often the most expensive part, especially for those with astigmatism, second only to the frames themselves, which is another racket altogether.

The reason MCP isn't dead is because practically ~every company on the planet is building an MCP server. I know this because we interact with all of them.

Wow if that's not an echo-chamber comment I don't know what is.