elteto

Local models will never compete with large SOTA models, in the same way an iPhone doesn't compete with supercomputers doing nuclear simulations.

They paths will differentiate and split. Probably SOTA models will eventually be locked down and only accessible to state actors because of how expensive they will be to run (already started with Mythos).

The image editing demo was fun... the model is not very well censored.

Then you sound like the type of client that perfectly fits Bambu's profile. Now, don't come complaining when Bambu decides you need a monthly subscription to use the slicer, or whatever rent-seeking they come up with in the future. Remember, you are paying extra for the privilege of not thinking, and you bought into that arrangement fully aware.

Hi Cliff,

You captured my imagination as a teenager when I read The Cuckoo’s Egg. I’m so sorry to hear of these terrible news but I appreciate the update. Here’s hoping to a _prompt_ recovery!

And you say this as a prolific open source contributor I’m sure.

Frontier has humans that you can talk to, for a fee. They charge $25 for going to the counter at the airport.

Competition for non-monetary resources is absolutely a thing. Developer time is scarce and other projects can absolutely see others as competitors in this regard. We have plenty of stories of project forks sprouting because of frustration/disagreement/etc and the new fork starts gathering more attention/contributions because of better governance, better devx, saner environment, etc.

Big players use CATIA and Siemens NX almost exclusively. I don’t know many using Autocad, maybe architectural firms.

I think you might agree with the article if you read it.

Only after working through Rudin’s Analysis first.

If DOGE served for anything at all it was for showing that there isn’t even that much “waste” per se. If there’s any waste it’s in the Pentagon which can’t even audit itself, but of course DOGE didn’t even get close to that. It was all performative for them.

Propaganda aside (which exists), the world is just an extremely complex place and the people writing these things are taking guesses a lot of the time. That’s it.

You are watching astroturfing in real time.

I think this is the first article that truly gave me “slop nausea”. So many “It’s not X. It’s Y.” Do people not realize how awful this reads? It’s not a novel either, just a few thousand words, just fucking write it and edit it yourself.

About the same amount you had in your comment, to be honest.

I think you are splitting hairs. Yes those models “exist”, if by exist you mean they have dual-licensing setups with different tiers (community, professional, etc).

The point is that most individuals who open source their code do so without expecting financial returns from it. In that context, whether Carmack has a $1 or $1e9 doesn’t make a difference.

Attack the argument not the man. Whether he is set for life or not has nothing to do _in this context_, since, presumably, people who open source their code do not care about profit.

How very typical of non-Americans to misrepresent Americans!

You don't need to explain anything to the government, that's why we have the 5th amendment. It is the government's job to bring charges against you and prove them beyond reasonable doubt. The government is right to investigate and ask questions to accomplish that and I am right to refuse to answer anything.

It's basically "innocent until proven guilty". Red light cameras turn that assumption around since if your car gets ticketed it is assumed you are "guilty until proven innocent".

That’s really cool. Scotch tape is just weird, peeling it also generates x-rays (in a vacuum):

https://www.technologyreview.com/2008/10/23/217918/x-rays-ma...

It was a demo mission and that change was actually consulted and approved by NASA.

I think the main difference was political: for Apollo you had the most powerful nation in history throw their economic and political will into pushing a project forward.

NASA programs today are mainly about creating/maintaining jobs and keeping private industry contractors busy. They lost the political agency and freedom to move fast that they had in the 60s.

You don't test the nozzle on _launch day_, what kind of ridiculous statement is that? You think the Air Force is paying SpaceX so they can test things the day it flies?

All components go through several test campaigns on the ground, while iterating on the design to address issues. These campaigns take months/years. That's why changes are stacked into "blocks", which are the equivalent of rocket versions. Each block must be certified by the Air Force and NASA to be deemed worthy of flying their payloads.

> When SpaceX launches a rocket, they think it will work. When NASA launches a rocket they know it will work.

That is such an ignorant thing to say. You think Falcon 9 has had 500+ successful launches because they _think_ it will work?

The difference is that SpaceX is a private company that has the ability to iterate fast. NASA is a jobs program and Artemis/SLS a barrel of pork, simple as that.

The Taking of Christ is absolutely stunning. The way light reflects off the dark armor is incredible.

Wonderful! Thanks for the quick turnaround.

Very clean and polished! I love how smooth the seconds hand move.

I work in an environment where we look at the time across many different timezones around the world. A couple of feature requests if you are ever in the mood:

  1. Make it possible to specify the timezone.
  2. Make it possible to create a grid of clocks, each with different timezones.
  3. Persist the grid/timezone state in the URL so links can be easily shared.

A common pattern is dynamic linking for development and static linking for production-ready releases.

As unsafe as C or C++. In fact, safer, because only the ABI surface is unsafe, the rust code behind it can be as safe or unsafe as you want it to be.

I was addressing this portion of your comment: "C's ABI and dynamic linking are the thing that enables the software to get huge". If the C ABI is what enables software to get huge then Rust is already there.

There is a second claim in your comment about a "safe ABI", but that is something that neither C or C++ offers right now.

> Until that happens, it'll be hard to adopt Rust in a lot of C/C++ strongholds where C's ABI and dynamic linking are the thing that enables the software to get huge.

Wait, Rust can already communicate using the C ABI. In fact, it offers exactly the same capabilities as C++ in this regard (dynamic linking).