ghxst

> I'm worried these technologies may take my job away

The way I look at this is that with the release of something like deepseek the possibility of running a model offline and locally to work _for_ you while you are sleeping, doing groceries, spending time with your kids / family is coming closer to a reality.

If AI is able to replace me one day I'll be taking advantage of that way more efficiently than any of my employee(s).

The countries I am familiar with in Europe (NL to name one) you can buy sim cards without any ID. Additionally there's at least 1 provider I know of that's giving them away for free while for the majority you pay 1-5 EUR but get some data after activation. There's no limit on how many you can purchase at once either.

Fair point! I should have clarified I'm disappointed that Microsoft removed the capability from the api.

> no custom SSML

I believe this used to be available for edge tts, very sad to see they removed it.

If anyone knows of comparable projects that implement something like SSML please do share.

Was this a case of something along the lines of an isolated function that had a bunch of bit shifting magic for some hyper optimization that was required, or was it just regular code?

Not saying it's acceptable, but the first example is maybe worth a thoughtful discussion while the latter would make me lose hope.

Heard positive things about Caddy before, do you know if it works with ip adresses as well?

In principle yes. The implementation however is rarely defined (if you have examples, feel free to share). In my experience it's by choice that companies use compliance to turn their dev environments into a micromanaged hell that engineers hate working in (our company went through multiple compliance certifications including SOC1 and SOC2).

Is it typical that anyone gets compensated for a temporary road closure? From what I understand this is one of the safest areas for space debris to re-enter, so likely it's justified and just part of having both spaceflight and aviation industries.

I would imagine a lack of experience with gear and not knowing how to react to sudden weather changes might be 2 big reasons why things go wrong. These environments are not forgiving when panic sets in or you make a long series of minor mistakes, many stories like these when it comes to scuba diving as well (especially cave diving).

If that were to be solved (if at all possible, and feasible / competitive) I can definitely see "LLM mining" be a historic milestone. Also much closer to the spirit of F@H in some sense, depending how you look at it. Would there be a financial incentive? And how would it be distributed? Could you receive a stake in the LLM proportional to the contribution you did? Would that be similar in some sense to purchasing stock in an AI company, or mining tokens for a crypto currency? Potentially a lot of opportunity here.

I'm not sure how accurate it is but my gut feeling is that the level of meaningful compression is somehow correlated to the level of intelligence behind a model, I wouldn't be surprised if it ends up being a major focus in general intelligence.

What's an insurance focused model in this context? Sorry if it's a stupid question.

> just $65 can provide a child with six weeks' worth of this therapeutic food.

That comes down to a little over $1.50 a day, that's great!

Seems like it comes in packets of about 500 calories each, containing 40grams of carbs, 30.3 grams of fat and 12.8gram of protein.

Unsure what fiber contents are like but Wikipedia does list that it includes vitamins and minerals.

> as of 2018 Plumpy'nut patents have expired in the US, UK and the European Union

Also says it's manufactured in the US, I wonder if these are available at food banks and such, maybe even in schools and if parents know about it.

Edit: additional fiber is included as well according to Wikipedia.

My main concern is people having to skip meals and kids facing developmental issues due to malnutrition. I get that the idea of feeding people something like Soylent isn’t exactly appealing, but if we could bring the cost down to nearly free for those in need, it could be a really effective safety net, (maybe not Soylent as it exists now but a product like that).

Something I wish more site owners would consider is that if you expose endpoints to the internet, expect users to interact with them however they choose. Instead of adding client-side challenges that disrupt the user experience, focus on building a secure backend. And please, stop shipping business logic to the frontend - especially if you're going to obfuscate it so badly that it ends up breaking on non-Chrome browsers because that's the only browser you test with.

Of course, there are exceptions. If you genuinely need to use a WAF or add client-side challenges, please test your settings properly. There are websites out there that completely break on Linux simply because they are using Akamai with settings that just don't match the real world and were only tested on Mac or Windows. A little more care in testing could go a long way toward making your site accessible to everyone.

Something articles like these make me wonder is what would a single "perfect" meal look like? Perfect as in a meal that provides all the essential nutrients the average person needs to stay healthy over the long term? The goal would be to make it as cheap as possible, easy to mass-produce, and have a decent shelf life. It might not be the most exciting food, but it could address serious issues like malnutrition and food insecurity. Food banks could always have something reliable to offer, combine it with a supplement to make up for any lack of micro nutrients and no one would have to worry about not having access to basic nutrition. It feels like a solvable problem, but it's likely harder than I imagine.

Frozen fruits like berries are typically really budget friendly, as long as they get frozen shortly after harvesting they contain the same amount of anti-oxidants and taste fine in my experience, although might require a bit more planning.

I'd assume physics are the biggest constraint here. Just look at the machinery and equipment needed to raise a falcon 9 which is "only" ~230 ft tall.

That's partly what I assume people star repos for, it really doesn't defeat my use case. It's a metric that indicates interest and popularity more so than approval or endorsements.

To be fair, Firefox out of the box prevents against font fingerprinting more than Chrome, it's considerably easier to get Firefox to run in a docker container and pass all the client side challenges than Chrome in my experience, you still have a valid point though.

This is great for bypassing the server side bot detection but not the client side one, where it will attempt to verify the integrity of your browser environment.

I sort by most amount of stars quite frequently when I am learning a new language and want to know what the most popular package is for something. What do you think would be a better metric for a use case like that?

Wiki's are still doing well in the gaming space I think (https://oldschool.runescape.wiki to name one, has a dedicated community of thousands of players behind it).

I agree with you in general though, the GitHub wikis really don't feel the same.

That's a cool idea and I think a fair argument. I learned a lot from watching Ben Eater on YouTube, he manages to explain (and demonstrate!) how CPUs and many other components work really well.

Out of curiosity, what language did you write your toy debugger in?

Oh yeah I should have clarified my post wasn't necessarily aimed at Roblox but rather generally speaking.

This to me on a surface level this seems totally fine, however I can't help to feel conflicted when that money is being spent on some of the more "predatory" games like gachas, however I'm not sure if it's really too different from kids spending their money on pokemon cards either. Do you feel different about those or would you say it's the same thing?

If you're curious about learning how to use debuggers and getting more familiar with them, I came across https://microcorruption.com/debugger/Tutorial and found it to be quite fun and engaging.

For those who use debuggers regularly, would you be willing to share how you learned to use them or any tips and resources that helped you?

I also want to say that the effort behind articles like these is really appreciated. It’s valuable knowledge to share, especially since it seems like the friends I teach are more overwhelmed by them nowadays (maybe I just suck at teaching, but I'll keep trying).

You might find the journey of Tinycorp's Tinybox interesting, it's a machine with 6 to 8 4090 GPUs and you should be able to track down a lot of their hardware choices including pictures on their Twitter and other info on George his livestreams.

> so much so that I think it makes more sense to think of exploits as subscription services.

I think this makes sense, especially given the uncertainty of when an exploit gets patched.

To my original argument of political power vs valuations you can probably say that having those same people you'd otherwise try to influence on your board with a financial incentive allows you to achieve the same thing, I'm not sure why I didn't consider that before.

That's probably a fair assumption too.

I was mostly thinking that the customers / clients you have and services you have to offer can be largely dependent by people in positions of power where having the right connections and influence might be the key difference between a service or product being viable.

For example - although not related to NSO - something like operation Trojan Shield required both Australian and Lithuanian cooperation due to fourth amendment interpretations.

Having a zero day in such cases is only part of the work and everything beyond that might be very much dependant on the strings you can pull.

But I can also see the argument that that would be something the government can figure out after they buy the product or service, so maybe I'm wrong on that and it's less important than I thought.