• 0 Posts
  • 161 Comments
Joined 10 months ago
cake
Cake day: November 22nd, 2023

help-circle

  • I really don’t understand how people use Instagram. I’ve tried, but it’s about 45% ads, 10-15% posts by people I don’t follow, it’s not in chronological order (or any sense of order for that matter), and regardless of whether I was on there yesterday or 2 months ago, it’ll show me about 40 posts before saying “You’re all caught up from the past 3 days!” and then refuse to show me any more.

    I guess this is why I’m here on Lemmy and went crawling back to Tumblr, one of the last vestiges of the old internet. At this point, I’d rather watch a platform die than become marketable to advertisers and shareholders.



  • Tipping is ingrained into our basic economic culture. Restaurant staff (waiters and waitresses in particular) make 80%+ of their money through tips. Federal minimum wage is about $7.25 USD, and almost no states have a minimum wage that low (some places it’s easily double that), but it’s completely legal to pay wait staff $2.25 an hour and expect them to make up the difference to $15-20 per hour in tips almost anywhere. A standard “good” tip at a restaurant is 20%. Even going to a grocery store you’ll often see a tip jar on the counter that people toss their spare change into. Outside of restaurants, no other job is completely dependent on tips to live, but in many service industries it’s still customary to tip as a way to show appreciation for a service rendered (especially if they go above and beyond).





  • Another Millennial here, so take that how you will, but I agree. I think that Gen Z is very tech literate, but only in specific areas that may not translate to other areas of competency that are what we think of when we say “tech savvy” - especially when you start talking about job skills.

    I think Boomers especially see anybody who can work a smartphone as some sort of computer wizard, while the truth is that Gen Z grew up with it and were immersed in the tech, so of course they’re good with it. What they didn’t grow up with was having to type on a physical keyboard and monkey around with the finer points of how a computer works just to get it to do the thing, so of course they’re not as skilled at it.


  • Because we’re talking pattern recognition levels of learning. At best, they’re the equivalent of parrots mimicking human speech. They take inputs and output data based on the statistical averages from their training sets - collaging pieces of their training into what they think is the right answer. And I use the word think here loosely, as this is the exact same process that the Gaussian blur tool in Photoshop uses.

    This matters in the context of the fact that these companies are trying to profit off of the output of these programs. If somebody with an eidetic memory is trying to sell pieces of works that they’ve consumed as their own - or even somebody copy-pasting bits from Clif Notes - then they should get in trouble; the same as these companies.

    Given A and B, we can understand C. But an LLM will only be able to give you AB, A(b), and B(a). And they’ve even been just spitting out A and B wholesale, proving that they retain their training data and will regurgitate the entirety of copyrighted material.



  • The argument that these models learn in a way that’s similar to how humans do is absolutely false, and the idea that they discard their training data and produce new content is demonstrably incorrect. These models can and do regurgitate their training data, including copyrighted characters.

    And these things don’t learn styles, techniques, or concepts. They effectively learn statistical averages and patterns and collage them together. I’ve gotten to the point where I can guess what model of image generator was used based on the same repeated mistakes that they make every time. Take a look at any generated image, and you won’t be able to identify where a light source is because the shadows come from all different directions. These things don’t understand the concept of a shadow or lighting, they just know that statistically lighter pixels are followed by darker pixels of the same hue and that some places have collections of lighter pixels. I recently heard about an ai that scientists had trained to identify pictures of wolves that was working with incredible accuracy. When they went in to figure out how it was identifying wolves from dogs like huskies so well, they found that it wasn’t even looking at the wolves at all. 100% of the images of wolves in its training data had snowy backgrounds, so it was simply searching for concentrations of white pixels (and therefore snow) in the image to determine whether or not a picture was of wolves or not.


  • Yep, they literally cannot work any other way than as a ponzi scheme. Because the people “earning” want to take more money out of the system than they put in, and the company is taking money out as well just to keep the game running and the employees paid, as well as to make a profit. So you need substantially more suckers buying into the system than the money that is being paid out.

    Eventually, somebody is gonna be left holding an empty bag.


  • So the way Tumblr works is that your account is basically a blog, with your home page on the site being populated with posts from the accounts that you follow. You can reblog posts onto your own account and comment on them to create individual conversation threads like this one. At one point, there was a bug in the edit post system that let you edit the entirety of a post when you reblogged it, including what other people had said previously, and even the original post. This would only affect your specific reblog of it, of course, but you could edit a post to say something completely different from the original and create a completely unrelated comment chain.





  • I wouldn’t say that it’s harder to counterfeit so much as that the methodology is radically different due to the untrusted, peer to peer nature of crypto. Because of the way that that works, in order to fake a transaction you need to convince the majority of ledgers that the transaction occurred (even if the wallet that is buying something doesn’t have anything in it). Because the ledger is ultimately decided by majority vote. You can trace the transaction, but wallets are often anonymous, so the trail ends at the wallet. Especially since somebody would use a burner wallet to do such a thing. It’s basically buying something with a hotel keycard with a stolen RFID on it.

    I think governments don’t want anything to do with it because its nature causes it to be too unstable in its value. It would be like tying the value of your country’s currency to the value of day trade stocks. One day, your money is worthless; a week later, it’s skyrocketing in value.

    At the end of the day, currencies are a system of abstraction to simplify the process of trade - whether between people or countries. We agree that the magic paper is worth the same amount because it’s easier than arguing that the magic rock that gave your wife cancer is worth at least 2 goats, not one. It’s always going to be a flawed system in some way. Crypto’s flaws just make it an ideal system for black market dealings compared to traditional fiat currency in its current setup, on top of the energy and computing costs.