I’m the administrator of kbin.life, a general purpose/tech orientated kbin instance.

  • 2 Posts
  • 654 Comments
Joined 1 year ago
cake
Cake day: June 29th, 2023

help-circle
  • I would have thought so, but I think it depends on how thin the skin of the pipe is. I would also have expected a breaker to trip under that much load. But, based on that happening, I’d not be surprised if there are bypasses and/or broken breakers.

    When we moved into the house we’re in now, the RCD (GFCI) didn’t work at all. I pressed test, nothing. Had the electrician over to change it. He tested the actual actuation using earth leakage. Nothing. So, faults can happen too.

    I want to be wrong, though. Because that’s a pretty bad state to get into, I think.


  • The only way that immediately springs to mind is so unlikely to happen. It requires multiple faults/mistakes.

    1: The chassis of one of the two units became live (connected to “hot” for you Americans) but was also not grounded in any way.
    2: The chassis of the other WAS grounded and created a circuit for the current to flow.
    3: There was no RCD (GFCD or whatever you guys call it) on the circuit.

    In this way, that pipe would be the only thing connecting the two devices, and the resistance is causing a huge amount of heat (just like an incandescent bulb, or a heating element does by design).

    Probably other possibilities, but it’s just the first thing I could think of that could potentially produce this result. But, that’s a lot of safety features to have either failed or just simply not been in place for this to be possible. So, frankly I hope I’m totally wrong.


  • The problem with wifi is that things will go downhill quickly once you have too many stations online. Even if they’re not actively browsing, the normal amount of chatter that a network has will often just slow things right down. It would need to be split into smaller wifi networks linked somehow and that means someone needs to be in a central location that is easily traced.

    In theory I guess someone with a very fast connection could run a layer 2 VPN. Then you could all run a routing protocol over that network which is accessed over the internet.

    Lot’s of ways to do it really. Wifi alone is probably the worst though.



  • I mean you could have an open wifi mesh and/or a network of either cheap fibre/ethernet with open switches. Then using OSPF or a similar routing protocol that supports routing over LAN networks you could handle the routing between all the remote networks.

    I think you’d need to break the network up at some points to break down the broadcast domains. You could do a similar thing to defederating, by not accepting certain routes, or routes from certain OSPF nodes.

    Issues with LANs that get too big without splitting into a new LAN (limiting broadcast domains) and definitely even the most modern wifi becomes problematic with a number of active stations online (wifi is half duplex in operation). So multiple channels and some backbone either over point to point radio links, or cable to connect wifi zones and alternate channels would improve things somewhat.

    Not sure why you’d want to do something like this. But the tech to do it is fairly inexpensive.




  • I’ve used IPv6 at home for over 20 years now. Initially via tunnels by hurricane electric and sixxs. But, around 10 years ago, my ISP enabled IPv6 and I’ve had it running alongside IPv4 since then.

    As soon as server providers offered IPv6 I’ve operated it (including DNS servers, serving the domains over IPv6).

    I run 3 NTP servers (one is stratum 1) in ntppool.org, and all three are also on ipv6.

    I don’t know what’s going on elsewhere in the world where they’re apparently making it very hard to gain accesss to ipv6.



  • The shorts algorithm is really weird I find. The ones it advertises are usually essentially thinly veiled smut (whereas most videos I watch are tech/science/aviation stuff). But (and I think this is important), if I start watching shorts, most of them are normal type shorts in line with what I’d want to watch. So it feels like it puts some stuff up it thinks will grab your attention but doesn’t push it once you’re watching. I’m sure there’s a good reason for it. But, just comes across as weird to me.

    I don’t get right wing stuff in shorts, but I do sometimes get it in the main videos suggested. How I think it happens is, I watch some videos about computer games, or perhaps a video on a games/science topic which is made by a channel that either also has right wing content OR their viewers also watch such stuff. So it figures, well if they viewed video A, and viewers that watched video A also watched video B, let’s push video B to that user too. The algorithm doesn’t really understand left or right wing, just pattern matching from views and other metrics.

    I find if you start telling youtube that you don’t want to see them, they will go away, at least for a while.









  • The only thing stopping them doing this right now, is that they know it would get regulatory pushback. It has a real chance of causing laws to be made about when and how advertising is appropriate, and those laws might stop some of the things they’re doing now. So they sit as close to that line as they can without crossing it so they can keep self-regulation.

    The moment they believe world governments wouldn’t stop them doing it, is the moment they’ll do it.

    And in terms of benefit for the advertisers and service providers, it’s a no-brainer. Advertisers know that a large percentage of people tune out, or even leave the room when an advert is on. I think it’s part of the reason they kept them so short on youtube, because if they showed you that there’s 1:30 ad break you might go to the toilet, get a drink, or anything else that takes you away from the ad. If they show you 15seconds, well you’ll probably just sit that one out.

    An advert they know people are actually watching is worth a LOT more to advertisers.