Remember that (I think) C++ race condition in RTG software, that killed people with something like 0.0001% probability and it was a huge deal and a reason to immediately retire the devices (or maybe just fix the bug, the point is that in medical, it's super important to have a high success chance)?

I'm sure AI doing diagnosis will be able to get to a higher success chance, lol.

EDIT: From a quick search, it looks like mis-diagnosis chance in doctors is around 10%. I still don't think AI can do better.

I've seen a lot of technical recommendations, but what I found most fun to experiment with is visual/art/music stuff, so here are some recommendation if that's also your thing. It's not strictly programming, because most of it requires learning more skills than just that, but I see that as an advantage. YMMV, though.

The Book of Shaders is and extremely good introduction to some basic shader stuff. Especially thanks to the interactive editor they have in their tutorials, and web tools like Shadertoy, experimenting with shaders is easier than ever. It was the tutorial that made me finally get past the "super confused" part of learning shaders.

It's kind of math heavy, especially once you get into 3D stuff, but I find it fun to learn, plus it's a rabbit hole and you can do some pretty cool stuff once you get into it. In general, anything technical artist related is interesting.

Another thing I'd recommend is looking into Algoraves. Algoraves are live performances where both visuals and music is performed by people live-coding their tracks and projections in some kind of language that's made for the task. TidalCycles, one of the libraries/languages that's commonly used, has a web editor, and there's also Sonic Pi, although I've never tried that one.

Processing is another language/tool used for making visual art. It also has a web edittor (with a lot of tutorials), and can make some cool visual stuff that can be fun to learn.

And one last recommendation, this time not about art, but about learning/building your CPU, your own assembly language, and learning to do stuff in it! Turing Complete is a puzzle game, where you will learn how to build your own CPU, starting from a single NAND gate, slowly combining them into registers, memory, adders, ALU, up until you have your own, complete and working CPU. You then create your own instruction set and use your CPU to solve a few puzzles.

It's super fun and engaging, and I'd consider learning logic gates and building a CPU as kind of also programming.

Unless you need to work on a solution with more than a few projects, such as Unity games. Then the LSPs go haywire and eat 20+Gb of memory, while not actually working.

Which, ofc, is Microsoft's fault, since it's their analyzer that has had the bug for years now. Rider didn't have that problem, but it shits itself when you change branches. You can't win :(

While I haven't read the paper, the comment's explanation seems to make sense. It supposedly contains a mathematical proof that making AGI from a finite dataset is a NP-hard problem. I have to read it and parse out the reasoning, if true, it would make for a great argument in cases like these.

https://lemmy.world/comment/14174326

Whi getting through college, I was always bummed that we have to learn a lot of stuff that seemed super irelevant to my future carreer, while also being annoying. Stuff like prolog, Phyro, Lisp, Assembly, or bunch of obscure math.

It was only years later when I finally realized why it was important - the school wasn't for teaching me to be the C#/Java programmer, but it taught me to be A programmer. I can pick up and start successfully writing anything I need, in any language, relatively quickly and without issues, nonmatter whether it's functional, objective, or wharever style of language, because I've very probably already had to deal with, learn, understand and pass exams in language that is similar to it, since college made me learn a language from almost every style or flavor of languages there are.

I was surprised when I first saw colleagues struggle with picking up languages other than the ones they work in, and that was when I finally realized why and how sneakily did the college make me a universal programmer without me noticing it. And that's something that's harder to get when self-taught, because you don't get exams and it's easier to miss the point and just skip courses on lisp, prolog or lambda calculus, because it seems irrelevant, but the different point of view and approach used when writing in those languahes is what will teach you the most.

A good reminder to always set your password manager to auto-lock (with PIN for convenience) after 3-5 minutes. The PIN makes it easy to re-log, while not being bruteforceable (AFAIK after few failed attempts it reverts to password), and if someone would get to your PC, either physically or remotely, they won't be able to get all your passwords.

One of the best jackpots I've ever found during Red Teaming engagements was when I RDPd to a server through pass-the-hash, only to find an unlocked password manager with passwords for most of the other servers, service and admin accounts.

Many companies are still using Windows 7 machines or 2008 win servers, without MS17-010 patch. They don't really care about security that much, when it's inconvenient or slightly difficult to mitigate. They won't be switching entire architecture just for a few screenshots

I'm really conflicted about this game. I really love the idea from game design standpoint, it's a perfect mix of genres, while also serves as an amazing case study in how theming is really important for games. Craftopia apparently had exactly the same capture mechanics, but with real animals instead of monsters, and you could also use the animals to work on your base. No one really cared about it too much.

But reskinning it from animals to fantasy monsters/pokemon, suddenly entirely changed the feel of the mechanic and made it much better. And that is a really simple change, but with such a great impact, and a really important lesson in game and theme design.

I also kind of understand that they have given up on Craftopia and started pursuing this project instead - while it sucks for people who enjoyed Craftopia, the idea of mixing monster collecting game with a survival game is genius, and works really, really well. It makes so many simple mechanics much more fun to interact with, once you're doing it with your Pokemon instead of by yourself (or through an automation machine). But implementing it into Craftopia would not be feasible, and I don't really fault them that much for deciding to start mostly from scratch.

I also don't really mind that they've decided to use the Pokemon aesthetics - it's a look that fits this genre well, and it's kind of OK to just go with it - I mean, almost every souls-like game tries to look like Dark Souls and give the same vibe, because it just fits so well with the overall theme. So, IMO going for the same look that has been proven is fine. But, and this is why I'm conflicted about it, it's really really not good when you have members of your team tweeting how you've figured out a way how to bypass copyrights using AI, giving example on pictures of Pokemon, boasting how they look different yet you probably won't be able to tell which ones are original, and which are generated.

And now that I've seen this video, it's way worse than I thought. I literally couldn't tell the difference between the game I've spent playing last two days, and the Pokemon mod. The monsters look almost exactly the same, making it seem like they really did just go with "lets feed Pokemon into AI". And that sucks.

On the other hand, I think the developers are getting too much flak for liking AI in games. I've seen them being criticized for their other game, that is a party game where you give prompts to AI to generate pictures, and then vote which of the players didn't know what theme the pictures should be about. That's a pretty funny and great idea, and I don't see why it's shown around as a proof that this developer is evil.

Are they? I don't know. The game designer in me hopes that they got excited for the Palworld idea, cut some corners (which is suspicious and pretty sad), but abandoned Craftophia just so they can work on a project that's amazing idea that wasn't done before, and that they will focus on it in the future. Unfortunately, it's starting to look more and more that it's really just a quick cash-grab, where they just asset-flipped their game to make a quick buck, and will forget about it once the hype dies down.

I hope that's not the case, and that they really do enjoy working for the game, and are as excited as I am about the potential of the concept. I guess we'll see in the follwing months and years.

I think the headline is missleading, if I understand it correctly.

ChatControl is already possible, and implemented for major communication service providers that most of the people use. It's just not mantadory.

Currently a regulation is in place allowing providers to scan communications voluntarily (so-called “Chat Control 1.0”). So far only some unencrypted US communications services such as GMail, Facebook/Instagram Messenger, Skype, Snapchat, iCloud email and X-Box apply chat control voluntarily (more details here). source

~~The article states that they decided that they will not blanketly require it, but I don't think it says anything about rolling back the first version of ChatControl that's already in effect.~~

EDIT: I was wrong, the article actually does mention it, even though on pretty vague terms:

The current voluntary chat control of private messages (not social networks) by US internet companies is being phased out. Targeted telecommunication surveillance and searches will only be permitted with a judicial warrant and only limited to persons or groups of persons suspected of being linked to child sexual abuse material."

My home camera was exactly what prompt me to learn how to reflash IoT devices, and set up a Home Assistant network that's entierly self-hosted. It's way easier than I though.

I've just stumbled upon something that I think makes a pretty solid argument that federating with Meta goes directly against the idea of Fediverse (because I was actually intrigued about whether I'm not just projecting my dislike for Meta into it).

Take a look at https://www.fediverse.to/ (which I'm actually not sure if it's the official main page, but it is the first search result), this is literally the first selling point, written in (absolutely hideous :D) large font on the landing page:

The fediverse is a collection of community-owned, ad-free, decentralised, and privacy-centric social networks.

Each fediverse instance is managed by a human admin. You can find fediverse instances dedicated to art, music, technology, culture, or politics.

Join the growing community and experience the web as it was meant to be.

I think that with that in mind, there's no way how we should even consider federating with them. That is, of course, unless it's what majority of people wants.

I also have the same problem, and then figured that hey, I'm sure there's a startup that's using ChatGPT to summarize videos.

Apparently there is, so here you go. Note that I didn't watch it, and the quality seems to be pretty off :D The emojis were in the summary, I can't really tell why - I've literally just taken first google result, installed an extension to throwaway browser, ran it, and will probably never use it again :D.

But, it's kinda funny, trying to parse out what is actually true and what's the model just being derp.

Summary

Twitter has implemented a rate limit and paywall to combat the excessive use of Bots and web scrapers, highlighting the company's financial struggles and potential impact on advertising revenue.

💔 Twitter implemented a rate limit where non-blue check users can only view 600 tweets per day, while paying members can view up to 6,000 tweets. 00:00

💡 Twitter has implemented new features, including a rate limit and a paywall, to combat the excessive use of Bots and web scrapers. 00:18

😞 Twitter went down due to bad JavaScript code that caused a self-inflicted DDOS attack, and web scraping became a big problem after Twitter shut down access to its free API. 00:48

💔 Twitter's rate limit changes have led to the death of third-party apps like Apollo, despite a failed protest, and it's hard to believe that web scraping is the main reason. 01:25

💔 Twitter's refusal to pay bills, including its Google Cloud bill, and its eviction from offices over unpaid rent, highlights the company's financial struggles and potential impact on advertising revenue. 01:49

💰 The Google Cloud contract, worth hundreds of millions of dollars per year, expired on June 30th, and it is speculated that these new features are related to it. 02:13

💡 Twitter implemented a rate limit to manage their infrastructure migration to Raspberry Pi. [What? :D] 02:28

🤔 Elon Musk is supposedly fighting against the CIA and NSA, who are using Twitter for censorship, and there is anticipation for his upcoming battle against Mark Zuckerberg.

Key insights

🌐Being locked into a cloud platform can give the provider significant control and leverage over the user.