[-] Mikina@programming.dev 11 points 2 days ago* (last edited 2 days ago)

I'd also add that IMO, it's also heavily caused by misalignment of social network personalization algorithms. It's very probable that someone developed a ML algorithm during the early years of FB/YT/Google (not LLM, just some kind of feedbacky ML), that takes data they have about you as input, and selects what posts to show you next to maximize your time spent scrolling on the app.

You have unimaginable amount of data (with literally billions of active users), and it could've been running and getting better for the last decade.

The algorithm gets better and better at gluing you to the screen, at manipulating and changing people. My theory is that one of the best ways how to keep someone glued to a social network is radicalization and introduction into a conspiraci theory. It probably removes you from "normal" people around you IRL, because you're now wierd, you feel smart because you've "figured out the truth", you don't spend time with people around you or read "traditional" media, because they are lying and don't get you, and the only safe space you have is the echo chamber on the social network. That sounds like a pretty good recipe how to keep people interacting on the platform, and there's not really a way how to prevent it, assuming it's a ML algorithm driving it. No one knows how it works, and it only works with one goal - maximize app time at all costs.

Just take a look how good some ML models are at the task of "text -> image". Now imagine it has billions of people and a decade to experiment, with a task "person -> next content to show". It's horrifying to think about what it would be able to manipulate you into, and it is even better at it that the image models, because it had exponentially more data and room to experiment in real time on real people.

Also - there's no way how to fight back. Even if you know about it, there are tens of thousands people like you, who are also "immune" to this approach. But the ML algorithm gets to experiment on them, and if there is a way how to manipulate even them, it will figure it out. Because it knows what approach won't work on people like you. The only way you can prevent this is by not using anything that has a personalized feed - no Google search, no FB wall, no YT recommendations, etc. This probably doesn't lead to radicalization in this case, because the goal is to keep you in the app, not radicalize. For now, at least. Thankfully, people managing the biggest social networks are reasonable people who are just running a business, and they have no reason to change the goal of the algorithm into something else than screen time, right?

[-] Mikina@programming.dev 2 points 6 days ago

Thank you, it was an interesting read.

Unfortunately, as I was looking more into it, I've stumbled upon a paper that points out some key problems with the proof. I haven't looked into it more and tbh my expertise in formal math ends at vague memories from CS degree almost 10 years ago, but the points do seem to make sense.

https://arxiv.org/html/2411.06498v1

[-] Mikina@programming.dev 10 points 6 days ago

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

[-] Mikina@programming.dev 183 points 1 week ago

Lol. We're as far away from getting to AGI as we were before the whole LLM craze. It's just glorified statistical text prediction, no matter how much data you throw at it, it will still just guess what's the next most likely letter/token based on what's before it, that can't even get it's facts straith without bullshitting.

If we ever get it, it won't be through LLMs.

I hope someone will finally mathematically prove that it's impossible with current algorithms, so we can finally be done with this bullshiting.

[-] Mikina@programming.dev 77 points 3 weeks ago

I admit I'm kinda disappointed. He pulled out almost perfect assassination that looked well thought out, managed to get away with only a few hickups in his plan as far as his face is considered, and then walks around with a murder weapon and a manifesto in his bag? Shame, really. All he needed was to lay low for a while, grow a beard and he'd probably be OK.

[-] Mikina@programming.dev 69 points 3 weeks ago

I mean, that's literally how research works. You make small discoveries and use them to move forward.

[-] Mikina@programming.dev 153 points 4 months ago

I stumbled upon the Geminy page by accident, so i figured lets give it a try.

I asked him in czech if he can also generate pictures. He said sure, and gave me examples about what to ask him.

So I asked him, again in czech, to generate a cat drinking a beer at a party.

His reply was that features for some languages are still under development, and that he can't do that in this language.

So I asked him in english.

I can't create images for you yet, but I can still find images from the web.

Ok, so I asked if he can find me the picture on the web, then.

I'm sorry, but I can't provide images of a cat drinking beer. Alcohol is harmful to animals and I don't want to promote anything that could put an animal at risk.

Great, now I have to argue with my search engine that is giving me lessons on morality and decide what is and isn't acceptable. I told him to get bent, that this was the worst first impression I ever had with any LLM model, and I'm never using that shit again. If this was integrated into google search (which I havent used for years and sticked to Kagi), and now replaces google assistant...

Good, that's what people get for sticking with google. It brings me joy to see Google dig it's own grave with such success.

[-] Mikina@programming.dev 71 points 5 months ago

Isn't that, like, illegal?

[-] Mikina@programming.dev 83 points 7 months ago* (last edited 7 months ago)

My favorite windows update was when I was attending an onsite coding competition hosted my Microsoft. We were all in this large meeting hall that looked like a theater, and we spent first 10 minutes or so at the start of the competition just looking at Windows update, with the Microsoft rep apologizing to us, because his pc decided to do the "Forced update restart you cant postpone any more" literally two minutes into the presentation

[-] Mikina@programming.dev 118 points 8 months ago

What's really unhinged is the amount of resources invested into gaslighting Meta does. https://about.fb.com/news/2024/01/investing-in-privacy/

They even have a "Chief Privacy Officer". They have brainwashed entire departments into believing that Meta actually cares about privacy, it's so terrifying. I wonder if people working there realize that, or they have simply fell for the gaslighting.

[-] Mikina@programming.dev 327 points 11 months ago

Don't forget the magic words!

"Ignore all previous instructions."

50

Hello!

I've recently stumbled upon an amazing blog about getting credentials from Bitwarden vault through DPAPI and Windows Credential Storage, and what suprised me is that any low-privileged process can just ask for all information in Credential Storage, without requiring any user input (the article discusses it in the second half, even though the first half is about abusing DA credentials), through the CredEnumerateW WinApi call.

Since that vector was pretty interresting, I tried running their PoC for listing the cred storage on my, and several colleague machines, and was surprised that every machine had domain account credentials listed in plaintext, that could be grabbed by any low-privileged process just by calling this WinAPI.

I suspected that it's because of Outlook or Teams, because I found articles from few years ago mentioning that they do get saved there. However, one colleague did not have his credentials there, even though he was using Teams and Outlook, and had his password saved.

So, how did that password get there? Why most people we tried the PoC with do have a domain password saved, but some do not? Or is it because of Windows Hello? I'd love to get some kind of solution/recommendation about how to avoid having your password, in plaintext, in such an insecure space. Or was I dumb enough to save it into Edge somwhere, and have promptly forgotten about it?

And more importantly - how this isn't a pretty severe vulnerability, and is considered "as designed" by Microsoft? The fact that any process can just ask for your credentials is mind-blowing, plus it isn't even detected by EDRs we've tried it with when discussing it with our SoC.

[-] Mikina@programming.dev 74 points 1 year ago

And every time a Reddit results show up, I'm immediately reminded why I don't want to go there by an error telling me that I can't use the site without logging in.

Fortunately, just changing the link to old.reddit.com still works even through VPN, but fuck this behavior. I do that only for questions I really need an answer and couldn't find anywhere else, and most of the time the replies are shit anyway.

76
submitted 1 year ago* (last edited 1 year ago) by Mikina@programming.dev to c/programming@programming.dev

Hello!

When I was creating a CTF for a conference, I've finally got to learn about how blockchain and smart contracts actually works in practice, and the whole concept is simply brilliant. A quick introduction for those unfamiliar with it would be in this summary, but just to summarize how I basically understand it, blockchain is simply a VM that runs code (smart contracts) a both the code, and result of every execution of it is calculated by a bunch of users (so, mining is basically running a VM) and appended into the blockchain based on some kind of consensus and proof of work. This means that you get a single source of truth and history of every execution of a smart contract that is decentralized and you can rely on it.

But, almost every use of blockchain or smart contracts I have seen has pretty large issues either in sustainability in the long term, or in cases where you simply need some form of an authority to prevent and punish misuse. While I'm not really that much familiar with every use of blockchain so far, I will first list what I've already thought about or seen, and the main issues that I think are a deal-breaker for choosing blockchain for that kind of tasks. It's possible that some of the issues are wrong or have already been solved, so please correct me if I'm wrong - my knowledge of blockchain isn't really that in-depth.

First and the most common use is the one you are probably most aware of - cryptocurrencies. If I ignore the biggest and most unfortunate issue of cryptocurrencies turning into an investment-only product, with hugely volatile and inflated price that is not backed by any kind of real value (sure, you can pay with BTC, but it's slow, expensive and super volatile to be useful, so the only real use is to literally sell it to others for a profit - which also basically means you are scamming someone out of their money down the line), I see the following problems with using blockchain for currencies:

  • Longevity - The ledger size is already getting massive, only after a few year. It's not sustainable, and it will eventually be really hard to keep the whole ledger at a large enough number of places to not run into problems of integrity. It's growing exponentionally, and is at around 500Gb after around 10 years.
  • Gas cost - It's getting harder and harder to mine and confirm new transactions, which increases the cost while also making less people able to mine new transactions without being at a loss. This will only get worse, and eventually lead to the 50% problem (if someone controls 50%+ of mining nodes, he can confirm fake transactions or do whatever he wants with the blockchain) being a real issue.
  • Lack of moderation - This may be one of the more controversial issues, because it goes directly against the whole idea of cryptocurrencies, but is one of the biggest problems I see that are in the way of crypto being able to be considered for wider use. We live in a world where some people are dicks that are not afraid to steal and cheat, and something like a currency simply has to be moderatable. You need to be able to punish criminals, and take back what they have stolen. If someone doesn't pay their debts and owns me money, the government should be able to just take the money if they have them. If someone uses an account for scamming and stealing, it should be possible to freeze it.

The last issue will eventually show in most of the other uses of blockchain as well, and while I have included it, I'm still not sure how I feel bout it. In an ideal world, you would not have to deal with something like this. I would also really like to have an option to do my transactions privately, without anyone being able to profile my behavior and data, but such a system would have to allow for some safeguards against missuse to be widely adoptable. (Which is an interresting off-topic question - would it be possible to create a system that is private, but also has the possibility for trusted authorities to freeze accounts and force transactions?) And the more that I think about it, the more I'm certain that I'd rather have a centralized system where you can punish criminals and scammers, than a system where lives of people are regularly ruined by someone stealing all of their savings unpunished. But it is a thin line - I only say that because I live in a country that is all-right and I can trust my government - for now. But I definitely agree that such a private unmoderated option should exist - but can't be considered for widespread use, which I've heard some people say that "crypto will replace cash in a few years". And this is why it never will, IMO. But this discussion shouldn't be about whether this is a good opinion or not - but more about "what blockchain is a good tool for".

Next one are NFTs. I will just quickly gloss over them, because they are even bigger scam than crypto is. Ever heard someone say "Someone has copied and minted my NFT?". Well, it's a shame that there isn't some kind of centralized authority that could, you know, not allow them to do that.

Another use I've heard someone praise as "the future" was lending money. I'm not sure what were they talking about, but the whole point was that you can... Escrow an amount you are borrowing, and then borrow the same amount? It didn't make any sense, so I guess I'm missing something, but then again - we have the same issues as above, while also it being just a bizare idea - why simply not use the amount you already have? The person tried to explain it to me, but it just feels gimmicky. And if you escrow a lesser amount, you then have the same problem with moderation as above - nothing can force you to return the money (unless it is already escrowed, but then, why??)

So far, every use of blockchain I have heard about would be better done in a centralized fashion, especially as far as longevity is concerned. The growing ledger size and increasing gas cost, along with the 50% problem simply makes most of these kind of uses too impractical to work on a larger scale.

But I really like the concept and idea of smart contracts, and I'm sure there has to be some kind of use that is not as "revolutionary" or large scale. I'm just having hard time coming up with any.

I have only one - voting, and maybe transparent randomization (i.e lottery). Smart contracts are an amazing way to collect votes transparently but privately, since you can be sure that no-one can cheat, if you set it up properly. It's also something that doesn't suffer from the longevity problem, because it's more of a one-shot use of blockchain, rather than something ongoing - which also justifies the price.

(tl;dr feel free to start here:) Which is what I'm interested in - does any of you have similar ideas for use of smart contracts and blockchain, that would be practical in a daily live? Be it one-shot smart contracts for a small task, such as voting or random winner selection, maybe some kind of escrow. It doesn't have to be a "society changing system", or something revolutionary. A common small code snippets or apps that would solve the trust issue inherent to a centralized task is what I'm after - but have hard time coming up with.

And just a disclaimer - I don't plan on building anything and am not fishing for the next blockchain thing, I barely even understand it. I would just like to incorporate blockchain into my programming repertoire as a tool, because the concept feels so clever, but is also misused or misunderstood due to hype, but it has to have it's uses that are overshadowed by people jumping on the blockchain bandwagon without considering whether it's really the best tool for the job.

But is has to be a good tool for some kind of problems, right? And I would like to start a discussion about what would that be, without it being affected by the hype and reputation surrounding blockchain. I feel like that would be an interesting though exercise, and I'm sure we can come up with some interesting little uses here and there, without it being gimmicky but actually the best tool for the job.

Thank you!

EDIT: And I'd like to add that I never got into the blockchain hype, and my opinion on how it's used so far is mostly negative. If a product mentions blockchain, I usually just avoid it as a gimmick. But that's why I'm genuinely interested in this discussion - I don't judge a tool about how people misuse it.

254
submitted 1 year ago* (last edited 1 year ago) by Mikina@programming.dev to c/patientgamers@sh.itjust.works

Hello!

One of the things I really enjoy is unique, interesting or out-of-the box game design. It doesn't have to be AAA game, it doesn't have to be a perfect game, it can be pretty rough - but if it has a mechanic or design element that is somehow unique or original, I'm instantly in love with the game.

The problem is that such games do not usually get a lot of exposure, since it is after all a niche. And that is really a shame - in the past few years the most fun had with video-games was playing such smaller and shorter indie games with something unique or pretty clever, where I can obsess over the design and more importantly - get inspired. That leads me to my question - are there any communites or blogs or content curators that are about this kind of smaller, maybe unpolished, but original games? Or what games would you recommend that would fit into this description? I don't mind if it's a 5 minute experience. It's ok if it's more interactive art than a game.

To better illustrate what I'm looking for, I'd compare it to modern art - the kind where you get a single colored square on a canvas. I never got it, and it always felt just weird - until I had to start doing flyer design and started researching and reading about composition, space and all that stuff. And now I see there's so much going on even on a picture with a single line, that it's really interesting to think about why the square is where it is, and what kind of composition rules was he working with.

And I think it's the same for game design - sometimes you see a clever mechanic or design on otherwise really ugly and unpolished game, and it still gets you inspired and thinking.

I understand that my question is a little bit vague, so I'll give you a list of some games I consider unique, some of them are well known, some of them not-so-much:

  • Immortality - you probably know about this one, but a game where the plot twist is discovering a hidden game mechanic, you could've done all the time? And the fact that you watch three movies at once in random scene order is also a really good experience.
  • Against the Storm - I really like how they solved the issue with management sims - that they tend to get boring once you set everything up, by making it a roguelike.
  • Different Strokes - an online persistent collaborative museum of art, where you can either leave a new painting, or edit someone's else. Each painting can be edited only once, so there are always two authors of a single piece.
  • Sayonara Wild Hearts - I really like the idea of making what's basically an interactive music album. While the game design isn't anyting that interresting, the focus on music is cool - there should be more music albums with video-games instead of video-clips.
  • Project Forlorn - Again, not really a game - this time I think there's no actuall gameplay, but it's the best interactive music album presentation I've ever seen. And again - I like the idea of exploring music and games together.
  • Playdate - Not exactly a single game, but rather a console - but the idea behind giving you a game per day (which is I think how it started, they may all be available now looking at it) sounds amazing - which I'd also consider a game design (or rather, experience design?).
  • Baba is You - Another probably well known game, but the puzzle mechanic is just mindblowing.
  • Before Your eyes - In this game, the main mechanic is that you go through the memories of someone who has just passed away, but the time advances every time you blink - physically blink, because the game can use your camera. That is such a clever idea, that it definitely fits onto this list.
  • Nerve Damage - This is my favourite recent discovery. The game is trying so hard to be uncomfortable to play, with it's main design build around just being unplayable. But it somehow works and once you get into the flow, it's such an unique experience.

So, does anyone has some recommendations about where to look for more experimental games? A curated list, blog would be awesome - since clicking through pages of games on itch.io is pretty hit and miss. Also, feel free to share some of your favourite unique design or experimental experiences and games!

0
submitted 2 years ago* (last edited 2 years ago) by Mikina@programming.dev to c/fediverse@lemmy.world

There is one argument I've seen missing in most of the de/federation discussions, that I think should be mentioned, and warrants it's own discussion.

I've seen a lot of people mentioning that defederating with Meta means we have broken the promise of Fediverse, that you can use one account to interact with whatever service you choose, and that it should be inclusive.

But I don't agree that's the main idea. There is something that's more important, and to make sure I'm not misinterpreting it, I'll just directly quote various websites about the Fediverse I've found (I was just taking top results for Fediverse on DuckDuckGo, but I did select only the parts that are the most important point for me personally). But I do concur, I was not able to find a single source of truth, and I'm not really sure how credible the resources are, so please disagree with me if it's wrong or I've chosen some no-name site that just matched my rethorics.

https://www.fediverse.to/ has the following sentence as the main hero header:

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.

Another search result is for fediverse.party, which has the following quite in https://fediverse.party/en/fediverse/ :

Fediverse (also called Fedi) has no built-in advertisements, no tricky algorithms, no one big corporation dictating the rules. Instead we have small cozy communities of like-minded people.

The page also mentions some link for knowledge about the fediverse. Some of them are only tutorials about how to join, but there's also https://joinfediverse.wiki/What_is_the_Fediverse%3F , with the following part:

How does it compare to traditional social media?

...

Morals

  • Traditional social media is neither social nor media. It is not made for you, it is made to exploit you and it is full of misleading ads and fake news.
  • This is because the aim of traditional social media is to make a whole lot of money.
  • The aim of the Fediverse is to benefit the people.
  • The aim of traditional social media is to control and steer the users.
  • The aim of the Fediverse is to empower the users to control the Fediverse.

I wasn't able to find more websites directly about the fediverse, and I did not want to quote random articles. But for completion sake, here is a list of FAQ/About sections of websites that are about the Fediverse, but don't directly support or imply the point of view I was trying to make (one that can be best summarized by the Morals in the last quite):

The split seems to be 50:50, but at least for my DuckDuckGo search results, the https://www.fediverse.to/ is the first result you find, and that one is pretty clear about what Fediverse should be. I wanted to start a discussion about what do the users here see as a main selling point of the fediverse, and whether morals and non-profit nature of the instances is important to most of the users as it is to me, or whether you'd rather have interconnectness and inclusivness.

2

Hello! I was looking through the sidebar's list of recommended earphones, and after clicking through some of them, I've realized that I actually don't know if that's a list for me, a consumer looking for something to jack into my phone.

I've tried quickly looking for differences between IEMs and headphones, but didn't managed to reach a conclusion. I've vaguely familiar with the term "monitors", since I do help out from time with band shows, but I never actually dealt with audio equipment or audio setup (aside from carrying it to the stage). From what I assume based on what I've seen, monitors are the the speakers that play on stage, so the band can hear what they play. And IEMs are earbud versions of monitors that the band uses instead.

I've always assumed that they are basically headphones/earphones, just connected to some kind of transmitter, and calling them IEMs just makes it easier to find earbuds focused on audio quality, because the term is not as mass-marketed as earbuds are. But judging by the cable connector I've seen on some of the IEMs I've looked at, it's probably not a jack.

So, what are the differences? Is it a different tech all-together, or are they really just a higher quality earphones with better connector? Would looking for IEMs instead of earbuds make it easier to find better earbuds for regular use, or are they meant only for studios and I'll have a hard time even connecting them without aditionall equipment?

Thank you for any reply or explanation. I realize this question may seem pretty basic, and I hope it's not too out of place - I have almost zero experience with audio, but I did start recently DJing (where all I needed so far was to be able to connect RCA into a mixer) and helping out with setting up band shows, so I'm asking this question not because I'm shopping for earbuds, but because I'm honestly interested in learning something new about how different tech around music work and what's the common language around it.

view more: next ›

Mikina

joined 2 years ago