Hacker News new | ask | show | jobs
by forkerenok 53 days ago
There are tons of entry points available now [0], and I get thousands of announcements every day.

https://rmap.world/

It's so much fun with little pages, message boards and random people hitting you up for a chat. I brought up my own transport node and propagation node too to contribute to the mesh.

2 comments

I'd love to get a node working just for fun. But it also seems like a waste since I'm extremely rural. The closest node is 200+ miles away. The chances of seeing any other device but my own connect to it seem slim.

Is there still a reason to do this?

because the protocol is transport agnostic, there are a lot of interfaces to the public reticulum net that you can access over TCP, I2P, or yggdrasil.

https://github.com/markqvist/Reticulum/wiki/Community-Node-L...

takes away some of the fun of imagining the SHTF-all-corporate-infrastructure-is-gone scenario i guess but i think that for realistic mesh networking applications it’s cool to build out many infrastructure types and enjoy the fact that the mesh will reconfigure itself realtime across a variety of scenarios.

It works over TCP too. No need for radio hijinx.

So you basically eliminate futzing around with the hard parts until you understand the reticulum network itself.

Basically work your way down the OSI model instead of working your way up it.

Perhaps there are others in your neighborhood in the same position, who would only get into it if there were other nodes. So be the first, get your friends into it, and maybe more nodes will follow. It's only $30 or so for a device.

They have a decent range (15 miles or more) so depending on how rural you are, you might be able to create a line of repeaters back to a major population center.

Lol, I'm rural enough that the concept of "neighborhood" has no meaning here. I'd have to have a neighbor first. And friends all live further away than 15 miles.

Your point still stands though.

I literally just put the meshtastic antenna on the roof today, in an old services box. Been in the window for months, had a few weird perfect weather moments show a few nodes and a ping. Put it on the roof, hours ago, nothing yet.

Someone has to start up the area! (I live in nowhere maine).

Set it up, and when family visit, give them little LORA pucks to strap onto their belts when they go out on the property. Boom little property wide messaging network. Send out a text when dinner's ready!
I ended up getting a ham radio license and now I get to use technology that actually works (even if it's a little more janky than meshtastic/reticulum).

My friend is across town and I should be able to hit him with the line of sight meshtastic repeater from my house, but I've never been able to.

OTOH, we can hear each other clear on any of the ham bands.

For hobby usage, ham is fantastic. For decentralized communication for the general public, which seems to be Meshcore/Meshtastic’s goal, it’s a nonstarter. There’s just too big a barrier to entry.
And unfortunately Meshtastic fails miserably at that. Meshcore is better, but maybe not anymore. I'm not even sure Lora is the best technology for this either since you'd really want something that can listen to more than 1 channel at a time.

Lora seems to be a great technology for remote sensors within a 1km of each other that can transmit occasional data. But once that single channel fills up, the channel stops working.

> Meshcore is better, but maybe not anymore.

Why not anymore? I know meshtastic has dumb routing but I thought Meshcore was much better.

I think things like this are at high risk of having perfect be the enemy of good, but I’m not exactly in my comfort zone technically.

Because of the split. But your right, meshtastic does have dumb routing. And I haven't used meshcore, but I probably won't now until the dust settles on this for a while.

I would like to disagree with you here that perfect is the enemy of good for mesh networking. It's not that meshtastic is good, it's not. But the barrier to get to good is far harder than the offerings. There are three primary issues.

1. Lora can typically only receive and listen on one channel at a time. This prevents listening and transmitting on anything but the one channel. If you could have multiple channels, the incidence of radios stomping on each others signals would go down.

2. The FCC limits 900MHz unlicensed operators to 1W of effective radiated power, and Lora really isn't optimized to make that 1W go as far as possible.

3. A good mesh network will have reliable delivery and routing. Meshtastic is more "spray and pray".

FT8 works very well as a digital modulation, and it solves the first two, but it doesn't solve #3 even though it makes it so much easier to design a solution for #3.

For a real life example: FT8 on 5W of RF power can often get my signal from North America to South America, Canada, Australia, Japan, etc.

If you listen to 14.074MHz, that's the channel that primarily is used for FT8 on the 20 meter band. Pick a random Web SDR from this list [0] and tune to that frequency and set it to USB (Upper Side Band). The channel width is only 3Khz, but each one of those squiggly lines is one station transmitting a signal.

I was getting very good signals with this one [1].

[0] http://kiwisdr.com/public/

[1] http://21959.proxy2.kiwisdr.com:8073/

I vaguely remember reading an article where someone had somehow transmitted digital signals over HAM, could feasibly be a transport for a reticulum network, right?
Digital is heavily used in ham radio. For example, FT8.
> There are tons

I'm sorry but are you serious? That map shows 224 nodes in the world, fewer than 30 in the entire Western hemisphere. And only 24 in the world are using LoRa? Meshcore has 38,000 nodes, Meshtastic 10,000. Those two projects can actually be said to have "tons" of nodes.

It hurts your credibility. I trusted you, spent time trying to debug the map, thinking that something was wrong on my end... why am I only seeing 224 when there should be "tons", is there a filter, are these just super nodes....

So I looked into it because of what you said, but you raised expectations so much that I feel nothing but disappointment.

Fair enough and apologies. Justified or not, I took the comment I was replying to out of context of the current (MeshCore, LoRa) topic.

I was referring to the TCP/IP, I2P and yggdrasil endpoints. And regardless, "tons" was an unnecessary exaggeration.

too late. you're burned. time to buy a new passport.
that map on rmap.world is only showing nodes that run dicoverable=yes in their configurations or something like that.

based upon the announce stream coming through my local node, i am seeing around 14k unique identities advertising over 21k unique application endpoints (destinations) over the course of the past month or so that i’ve been tracking it.

It seems odd that such a tiny proportion would run `discoverable=yes`. Any idea why that is? It makes me question the numbers.