• 19 Posts
  • 82 Comments
Joined 2 years ago
cake
Cake day: July 5th, 2023

help-circle
  • No, it wasn’t, and that’s exactly the problem. Car dependent infrastructure is so bad for society and he was proposing a concept that would entrench it to a degree that would make today’s anglosphere look like a utopian Nordic paradise by comparison.

    Ignoring the fact that its already entrenched and not going to change without dedicated infrastructure that happens seperate to the development of individual vehciles, at that point, you’re asking him to make a video on urban planning rather than AI. Its an entirely different field. Might as well ask for him to make a video on vaccine development or something at that point. To be clear, I’m not saying you’re wrong about the importance of pedestrian and bike infrastructure, but it’s importance doesn’t mean that cars can’t or won’t develop further. Frankly, given the fact that Grey has lived in both American and Europe, I doubt he’d disagree, given the first-hand experience.

    Ending it and then saying nothing for months, before having your cohost release a lie stating that it hadn’t ended is unreasonable by any metric. It would have taken so little effort to put a post up on the subreddit saying “thanks for all the good times, but we’ve decided we’ve exhausted what we can talk about, you can keep up with me on my YouTube or my productivity podcast at these links.”

    Given the unscheduled, (relatively) umplanned nature of it, its likely they just didn’t fully know what the plan was. Again, not suprising for what it was. Its entirely possible, for example, that they do hope to come back to it someday. Declaring the indefinite hiatus a lie just because it hasn’t ended is an overreaction. They could very easily decide that in the last five years, they’ve amassed enough new topics to come back for another 100 episodes.

    You don’t get to give your fans cutesy nicknames and invite them to send you postcards en mass to vote on a community flag and then pretend to maintain the sort of faceless transactional relationship you described there. You just don’t, and it’s ridiculous to pretend otherwise.

    I mean, companies and organizations do that sort of thing all the time. Providing a banner for fans to rally to doesn’t make them any less of a buisness. I wouldn’t consider myself a friend of the head of the Nationa Research Counsel just because I was a big supporter of Boaty McBoatface, nor was that their goal.

    I should add one more point to the list though that I just … … … interested in that, Grey and Kurzgesagt ended up leaving.

    I was not aware of the Kerzgizart or Nebula drama. If true, that would be a much bigger concern to me, and yeah, would paint him as much more of a scumbag.


  • I do agree with the criticism against him in terms of Guns, Germs, and Steel, but I think the criticism of his automation and self-driving cars videos is a bit of an overreaction. These are videos trying to predict the future based on current trends. At the time, many were very optimistic about the progression of AI (and not just Tesla). At the same time, whille his predicted timeframe was off, its also not seeming unlikely that his predictions will come true. We’re already starting to see AI automate a bunch of jobs. If I remeber correctly, he also talked about the original video on Hello Internet about five years ago and basiclly said what I am now: his timeframe was optimistic but much of what he predicted still seems feasible if not likely. As for the traffic video specificly, it might be overly simplistic, but the idea is to give an idea of how traffic management might change in the future. It’s focus isn’t on bikes and pedestrian, and while not covered in the video, its not like they don’t fit in to that model. You literally just need a signaled crossing like is used now, or if you insist on ensuring near-maximum efficiently, a bridge, underpass, or otherwise separated infrastructure (which is usually better for pedestrians and bikes where possible anyway).

    I also don’t really think the end of Hello Internet was esspecially out of the blue, or unreasonable. It was very much an improvised, talk-about-a-random-topic type of podcast, so after about 200 hours of talking, its not suprising that they would find it difficult to come up with new topics that they can both engage with and make entertaining. IMO it was very clearly fizzling out well before they stopped, and post covid, it would have only been more difficult. The lack of announcement is annoying, but I would hardly consider it a betrayal.

    I know its very blunt but, it sounds like you were far more attached to him than was warrented. I always read his talk of fans as very clearly viewing them as a seperate group - appreciated customers, not friends. For example, his talk about using the Grey alias to seperate himself from his work, trying to keep himself faceless and mostly anonymous, and the treating of all of his works as a buisness. IMO he never tried to sell it as anything more than a buisness.














  • I can’t say I’ve noticed this, but depending on the design, it might just be about ensuring contrast of different elements. Red contrasts a yellow crown and beige skin pretty well whereas blonde hair would blend with skin and crown, and brown isn’t as bad but isn’t great and doesn’t work with brown skin. Black works, but can be hard to do with any detail in a minimalist style or one using black outlines. This leaves, red, and grey (with grey being them more common option in my experience).




  • Technically, it can and has been done already. The problem is that AI is very bad at creating new ideas and even worse at understanding what it has created (as is required for plots or jokes). As a result, any writting created with heavy AI influence tends to sound like a child’s stream of thought with an adult’s vocabulary, and any jokes rely purely on randomness or on repeating an existing well-known joke. Similarly with art and animation, because the AI doesn’t understand what it is creating, it struggles to keep animation of elements consistant and often can’t figure out how elements should be included in the scene. Voices are probably the strongest part, but even then, it can be buggy and won’t change correctly to match the context of what is being said.

    None of this is to say AI is useless. Its very good at creating a “good enough” quick-fix, or to be used to fill unimportant or trivial work. If used to help clean up scripts or fill in backgrounds, it can speed up the process greatly at minimal cost. It’s a tool to be used by someone who knows the field, not to replace them.


  • I think it is technically possible - with the Valve Index you can read the camera input like a webcam, and I’m sure theres some way to do it with the Quests (although probably not easily). That said, as others have noted, between the bulkyness of the headset, the lower quality of the cameras, the risk of losing tracking, and the natural shakyness of people’s heads, it likely wouldn’t be an improvement. Try watching VR footage from someone who doesn’t stream/video it regularly and you can get an idea of how hard the footage can be to follow, even before the lower camera quality.






  • Relative to cooking a similar meal, absolutely. Getting McDonalds takes like 5 mins and almost no effort. Less if ordering for delivery or pickup. If I want to cook myself a burger its probably going to take me like 40 minutes to makes and fry the burger, and prepare toppings. Im sure a good chef could do it much faster, but thats not me, and esspecially not after a full work day.

    Edit: Plus, less directly measurable and comparable, but the time and work for planning, shopping, and dishes afterwards.