josephwegner

> The company said 30% of that amount — about $5.3 billion — will be released to states only if rivals YouTube and TikTok meet two conditions: implementing similar safety features, including a one-hour daily time limit, a nighttime block and age-assurance measures; and paying the same amount, split between the two companies.

This is pretty shady. The $17B figure is much less impressive when a huge chunk of it is hinged on competitors choosing to hurt their business.

I have a silhouette cameo pro 4 (and love it), and have been wondering around similar things. The Cameo can only be controlled via the provided software, or a couple proprietary plugins for popular design software.

The provided software is clunky, and the plugins suck. I’ve been building an app for designing stickers, and what I really want is to either be able to drive the Cameo from the app directly, or output a design file in a format that can be loaded straight into the software to start cutting.

The thing I’m stuck on is… how do people reverse engineer these devices protocols, without fear of bricking the device? My understanding is that reverse engineering like this is a lot of “try shit and see if it works”. It seems to me that a few wrong commands could get the cameo in some irrecoverable state. Obviously that’s not a risk I’m willing to take on a machine I use very regularly.

Along with the hordes of other options people are responding with, I'm a big fan of Perplexity's voice chat. It does back-and-forth well in a way that I missed whenever I tried anything besides ChatGPT.

This assumes there is no added benefit to being able to reach your kids/be reached by your kids easier than it was historically. While I agree it's probably not as critical as many parents might make it seem, there are tangible benefits. Off the top of my head:

- Before cell phones, we were also in an age of far less mass violence in American schools. I completely empathize with parents wanting their kids to have an emergency contact device, given the relative increase in violence at schools.

- There is a long history of kids being abused, sexually or otherwise, by authority figures in their school. Having a lifeline like a quick text to a parent can easily be the escape hatch from a predator convincing a kid to do something unsafe.

Give it time. Networks are a garden that grow over time, and moreso if you cater to them. Some of those starving grad students will be VPs in 10 years.

I will say, there is a Wendy’s near me that is piloting an AI drive-thru experience, and I prefer it 10-to-1 to the human version. It had a clear voice, it didn’t disappear randomly, it understood what I meant the first time (even though I was speaking naturally - I didn’t know at first it was AI), and it asked me for feedback (“what sort of sauce?”) in a very understandable way. Drive-thrus are famously a bad experience - I’m happy to see improvement here.

Pretty sure they lost to rsync ;)

See

Same, actually. I’m feeling much more pro-ollama suddenly!

> coding isn’t about the finished product. its a lot like writing

This doesn't seem applicable in most contexts. Yes, when I'm coding for fun or purely for learning the finished product is less relevant... but I'd guess the vast majority of code that is written is for a business that _only_ cares about the product. Code is an implementation detail.

If (and this is a _big_ if) AI-based coding can increase developer velocity even as little as 50%, no sane business is going to let their engineers ignore it just because it's not as fun as artisanal code.

I become more and more convinced with each of these tweets/blogs/threads that using LLMs well is a skill set akin to using Search well.

It’s been a common mantra - at least in my bubble of technologists - that a good majority of the software engineering skill set is knowing how to search well. Knowing when search is the right tool, how to format a query, how to peruse the results and find the useful ones, what results indicate a bad query you should adjust… these all sort of become second nature the longer you’ve been using Search, but I also have noticed them as an obvious difference between people that are tech-adept vs not.

LLMs seems to have a very similar usability pattern. They’re not always the right tool, and are crippled by bad prompting. Even with good prompting, you need to know how to notice good results vs bad, how to cherry-pick and refine the useful bits, and have a sense for when to start over with a fresh prompt. And none of this is really _hard_ - just like Search, none of us need to go take a course on prompting - IMO folks jusr need to engage with LLMs as a non-perfect tool they are learning how to wield.

The fact that we have to learn a tool doesn’t make it a bad one. The fact that a tool doesn’t always get it 100% on the first try doesn’t make it useless. I strip a lot of screws with my screwdriver, but I don’t blame the screwdriver.

Thank you. It's too late to edit, but goodness am I embarrassed now. Very sorry, Kasey!

But I am actually interested in the economics! The author mentions sending her designs out to a factory - I would expect this is astonishingly expensive for a single prototype! Wouldn’t that be thousands of dollars? Is anyone familiar how to get good factory-made parts like this at DIY budgets?

Not that that takes away from the article at all. This project has many merits, and although cost may not be one of them, it’s still interesting!

Pretty sure that's exactly what Clearbit does. Although now their site has some AI hype on it and I can't tell if that's just the obligatory AI marketing, or if they actually changed their product.

https://clearbit.com/

Curious - I was watching the live stream, and noticed as the ship's rockets were on, there were a bunch of these "floaters" in the view.

https://imgur.com/a/OrZTYxP

I know nearly nothing about rocketry, so ELI5 (or give me a wikipedia link!)... anyone know what those floaters are? Is it some byproduct of the fuel? An artifact of the cameras? An illusion coming through the haze of the burning rocket?

That’s not how markets work, right? And I don’t think it’s true in either example. If you can charge more, you will. Price should match value.

OpenAI is charging a _really_ high monthly fee for ChatGPT, and it’s quite popular. It’s very limited - there’s no way the cost incurred for usage is near that. Obviously their costs include the R&D that has already happened, but I still think it’s priced way over that.

Drug manufacturers are famously making money hand over fist. Patients aren’t their customers. Insurance companies are. They are surely gouging insurance companies to their fullest capacity.

> traditional concept of marriage

While it is correct that marriage has traditionally been a union of people of two different sexes, I think the root of the issue is that is not how many people perceive the tradition in reality. I, along with many people, view marriage as a union between _two people_ - regardless of gender. The fact that marriage has mostly always been across sexes is a side effect of centuries of homophobia/heteronormativity.

So saying that marriage is meant for a man and a woman does not compute for me. Trying to enforce that is _taking away_ from my concept of marriage.

This is the news that came out last week, but as a person entirely outside of this industry - where do you all follow for updates on this sort of thing? I don't want to parse every single paper that mentions superconductors, but I'd be glad to follow news around this specific material. I'm glad to see less mass hysteria than with LK-99, I just personally am interested and would love to stay close to it for updates!

How in the world are they going to make money on this? Perhaps the hardware is cheap to manufacture, but with no subscription I feel like they are going to get taken to the cleaner on LLM API fees. Even if they're running the LLM themselves, it's not cheap. Could they really be getting enough margin on the device to pay their staff _and_ all of that infra?

Or maybe this is another VC-backed sale price :)

Really great article, and the site looks great! One small piece of feedback:

> the server-rendered HTML will always default to light mode. This creates a flicker for night owls

You might consider switching this - render the dark-mode version by default, and have the flicker be from dark-to-light. For users operating mostly in dark mode, a bright flash of white can be painful. The same is not true for users operating in light mode - they will barely notice a moment of dark.

Yep, I moderate a sub with ~85k subscribers, and pay $5/month to host a better auto mod than Reddit has built. It does an insane amount of moderator work for us. We will not survive without it, but it would cost a crazy amount with the new API fees.

I will turn it off on the 12th, and we will just suffer without it until we all quit and the sub dies. Oh well.

Anyone have a sense for system requirements on running this locally? StableDiffusion was too heavy for either of my machines (which are admittedly not very powerful) - I'm not seeing much in the docs to indicate whether or not this is more or less intensive to run locally.

Curious what the draw is to use a Chrome extension for this? Why not a desktop app?

I think cute little robots or even just animated characters would land better than this realistic avatars. I don't want the AI to feel human, but it can still have a "personality". WALL-E vibes come to mind.

Seems like we need a new meme. `current_year` will be the year of web 3.0!

Yes, this lines up with the "Critical Project" concept that has been floating around in the past year. It is... contentious to say the last. Previous HN discussion: https://news.ycombinator.com/item?id=32111738

seniority * title…

The parts that are inaccessible right now seem incredibly easy to overcome.

Using a CLI-based tool is inaccessible for most people... but building a GUI around this would be very easy. I'm too lazy to google it, but I would bet someone already has a GUI, or is working on one.

12GB of VRAM may not be accessible on most computers, but there's nothing innovative about offloading that task to an EC2 instance. It just requires an opportunistic developer to tie the pieces together.

I would be monumentally surprised if Figma/Canva/InVision/Adobe are not already working on this.