codexb

I think it's more that attestation can require a specific "approved" authenticator, and that authenticator is what would enforce all of the required OS/Software/Hardware requirements, even if it didn't get included in the attestation standard.

Yeah, it's mind blowing to me that GH still doesn't have a branch viewer in each repo.

Does this support using only private LLM instances? Does it integrate with Amazon Bedrock?

"Whistleblowing" requires something illegal to have occurred. It doesn't appear any of the disclosures being made about Facebook allege anything illegal. They are just disparaging insider information. Anyone who has worked in tech for any amount of time has signed an NDA. They are not nefarious.

This is essentially just MTG limited -- draft or maybe team draft. Draft is my favorite part. Though I don't necessarily agree with the "things that make magic fun". Mill/Discard/Land Destro decks are fun, janky decks that rarely come together and it's fun to try and make them work.

Did those original thoughts not build upon all the original thoughts that came before them?

All innovation is theft. It builds directly on top of what came before.

"Good artists copy, great artists steal."

It's always been true. AI just makes it available to more people faster.

Anthropic didn't lose because they scraped (read) copyrighted works. They lost because they distributed copyrighted works directly via torrents. Those aren't the same.

It uses the phrase “regulate commerce between the states” which effectively has the same meaning.

The controlling case is Wickard v Filburn (1942).

A farmer was told he could only grow X acres of feed on his own land; feed that he had no intention of selling and was being fed entirely to his own livestock on the same land.

This seems to overturn that in part, but until Wickard is overturned, and the interstate commerce clause reigned in, there will be weird side effects of it like this.

Because those changes might depend on those other changes. Git merges aren't linear. They're branched. And PR reviews are meant to just examine the merge of 1 branch back into master. They're not really meant to review multiple steps along the same branch to make it easier to review.

There are so many times where I want to create 3-4 Merge requests that all build on each other along the same branch instead of creating one giant MR but the UI for reviewing them doesn't really work that way.

I find LLM's particularly good at filtering and distilling a large rambling idea that I have into a well-formatted and coherent paragraph, and also removing any statements that would be perceived as overly argumentative or rude.

In essense, LLM's are a much better spell check.

NASA hasn't had a proper goal or mission for decades. That's their problem. And the spaceflight goal that everyone wants -- making things cheaper -- is not something that government agencies are particularly good at producing.

Well, in the sense that it's possible to be eating zero calories in the time between meals. You still need the meals. If you're just looking at brief snapshots, it doesn't tell you much.

People get doctorates in these fields and post studies in journals that get picked up by thinktanks and media outlets. It's "science" for all intents and purposes; they're used as a source of authority based on data and analysis and formal papers.

I think part of the problem is that what is often referred to as "science" these days is very different from the hard sciences of yesteryear.

There are a lot of "soft" sciences that get increasingly softer every year. Social sciences, gender and women's studies, political science, some of the fast and loose use of "economics" these days.

There are a lot of "studies" these days that are little more than slanted questionnaires or selective correlational studies with wild unsupported theories as to the results.

Germany and other EU countries that offer free tuition begin segregating students into vocational and university tracks starting around age 10. Only about 30% of students qualify for university. The rest stop school around 15-16 and go into vocational training. These aren't student choices, their dictated by the school system.

The US would never approve of a school system that told parents that their children weren't allowed to go to university and had to go into vocational training.

Ok, but forgiving student loans doesn't do that. It signals to borrowers that they don't have to repay high loans if their career can't support it. It tells borrowers that they can make risky loans without a chance of default. It tells universities that they can keep charging exorbitant tuitions because kids can still get loans to pay them.

The solution is to allow judges the discretion to default them in bankruptcy after X number of years after graduation. Lenders need to accept the risk. With no risk, they can loan as much as they want and have guaranteed repayment. This drives tuition higher and higher.

3d from moving 2d images has been a thing for decades.

I'm always surprised how even the people I consider incredibly intelligent get pulled into bad ideas.

There will always be value in doing work that other people don't want to do themselves or that requires expertise and skill that isn't conveyed all that well through books or pictures. The economy used to be full of stable masters for horses and carriages, and manual typists, and street lamp lighters, and television repairmen, and any number of jobs that don't exist anymore.

I'm pretty sure we'll survive.

I think it's a stretch to call having to make a living in a career other than your preferred job "suffering". Even before AI, there were surely millions of people who grew up wanting to be an artist, or an astronaut, or an architect, or any number of things that they never had the skills or the work ethic or the resources to achieve. I'm sure before cars there were people who loved maintaining stables of horses and carriages, and lamented the decline of the stable master profession. It's no different now.

At one point, there was a case for preventing scammy and fraudulent apps. For a long time, the ios App store had a much higher quality than android.

But now? There are tons of scammy and fraudulent apps on the app store. If you try to search for any popular app, you'll be presented with a dozen apps that look similar with similar names and logos.

License plate holders that obscure the license plate on private property.

Look at any hobby and there are lots of beginners and casuals and far fewer people who are very skilled at it. The Maker hobby is no different. It's certainly not a problem of the microcontrollers available. Arduino is the simplest, but there are plenty of others.

The "blinky LED" roadblock is really just a result of the fact that more complex "maker" projects require some amount of electrical or engineering or fabrication knowledge and skill, which takes some trial and error and practice -- the same thing that limits progress in lots of other hobbies.

The real "Maker" movement is the demand that drives so many consumer level fabrication tools and components that were only available as expensive industrial and commercial orders in the past -- 3d printers, laser cutters, microcontrollers, IC sensors, brushless motors -- there are so many options now that just weren't available at all 20 years ago.

Wow, I can't believe this is still around! I'm glad to see artifacts from the past like this are still out there on the internet.

Makes me miss Google CodeJam though.

Sounds more like people need to de-UK. It's going to be a problem with any company or technology.

I agree with you, but that only works if people value it and are willing to pay for it.

Look at email. It’s technically open, but in reality there are a few large players who control the majority of it.

The only way open source phone software succeeds is if there is real money behind it and there is an attractiveness to it that makes people pay for it.

They were, but the actual cuts needed (to entitlements) are politically impossible to make.

That's basically compilers these days. It used to be that you could try and optimize your code, inline things here and there, but these days, you're not going to beat the compiler optimization.