• kudra@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    2
    ·
    5 days ago

    Absolutely love this idea. Am concerned anywhere with enough Meshtastic nodes for this to be busy enough to be useful would congest the mesh too much though, as I’ve been reading Meshtastic falls over completely once you have too many nodes communicating …? 🤔

    • Salamander@mander.xyzOPM
      link
      fedilink
      arrow-up
      2
      ·
      5 days ago

      I’m not an expert by any means, but from what I have learned so far I can see that there is some logic to how the nodes process and route the messages to avoid congestion, and there are some options that one can make (like selecting specific frequency sub-bands) to make a more private mesh run smoothly.

      I suspect that a lot of the issues that you read about are related to mis-configuration.

      Examples:

      • You are setting up your first meshtastic device and you want to find other people quickly, and so you of course set up your device to downlink messages from the most popular MQTT server and make use of the default settings - like hundreds of other people around the world. This overloads your device because it is downlinking MQTT messages faster than it can process them, and it can become non-responsive.

      • You purchase a node and want to contribute to your local mesh, and you are lucky enough to live in an area that already has a dense mesh network. Since you want to contribute, you set up your node and make it a “router”, and use a common configuration. But your node is not actually very well positioned, and the “router” type has preference over other nodes. Your node can stop nearby stronger nodes from broadcasting. To make it worse, at some point there were problems because there was a Router Client setting (see: [Feature Request]: Admit Router Client was a mistake) many users did not understand the behavior of the router node and this lead to some meshes running into trouble. So, as the mesh grows, it becomes more likely that you will introduce nodes with sub-optimal configuration into the mesh.

      There is possibly some threshold at which a mesh really does grow so large that even if properly configured it may be problematic, but, honestly, I don’t know if this has happened. In my area we are certainly not close enough to having a functional mesh, so I have not experienced these limits.

      • kudra@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        2
        ·
        5 days ago

        The worst case I’ve read about is BurningMan 2023, once enough people were on the mesh it completely stopped working and they had to write a special version of the app so it wouldn’t fail so badly… so it looks like there are ways to mitigate congested networks, but that is not stock. I admit I’m a complete newbie but there are some scathing comments about how poor Meshtastic is, though I’ve read others that say it’s fine for what it tries to do but not great for other purposes. So my gut tells me Reticulum would probably be a better long term protocol which you could develop a stand alone app for BBS, instead of piggybacking on Meshtastic app.

        • Salamander@mander.xyzOPM
          link
          fedilink
          arrow-up
          2
          ·
          5 days ago

          Ah! I can see problems happening if you do reach those high densities. If an app change was able to fix it somewhat, then it may have still been some configuration problem. Too many people trying to join the same mesh with sub-optimal settings. But, at some point you might indeed get just too much radio interference from different devices trying to communicate at once. Certainly possible!

          I just learned about Reticulum, I still have to give it a try!