Considering the amount of tech and self hosting types that live in cities, it seems like it would be popular to have little mesh intranets all over the place, but I’m not aware of any.
I read about NYC Mesh a while back, and I wonder if there are other similar things already in widespread use that I just haven’t heard about.
in Germany there is Freifunk (https://en.wikipedia.org/wiki/Freifunk), based on gluon (https://github.com/freifunk-gluon/gluon) which is based on openwrt.
Let’s have a look at the Freifunk Map https://www.freifunk-karte.de/
It doesn’t actually mesh tho, most Freifunk hubs are just open WiFi hotspots that provide internet access via the usual internet providers.
That is impressive!
Zooming in, i can see villages where i know people. These are very small villages, and the system is in use.
So I imagine it’s friends, and family.
And that’s ok. A great way to start and build a momentum.
Go Deutschland!
In my village, I equipped the town hall and the fire station with free Wi-Fi nodes, and also the accommodations of refugees.
There seem to be many Meshtastic® nodes around, it’s pretty popular
However, it seems that in order to join the network you are forced to buy hardware that uses patented trademarked radio modules (LoRa®)
The website also says:
Meshtastic® is a registered trademark of Meshtastic LLC.
I don’t have relevant experience, but I’m looking to maybe join such a network, and being forced to buy specific hardware to join a network that is ran (/developed?) by a single LLC company is not very appealing for a network that is supposed to be resilient, off-grid, peer-to-peer, etc
So I’ve started looking into https://reticulum.network/ and Nomad network 🤔
There seems to be more traffic on MeshCore than on Meshtastic, probably due to its greater range. Also, the core library seems to be MIT-licensed.
Besides, given the goals of Meshtastic/MeshCore (low power long range text communication without a radio license), LoRa is a sensible choice: It operates in appropriate radio bands, is power-efficient, and hardware is readily available at reasonable prices.
Sure, something like DASH7 would be more open but it’s also much harder to find hardware for. The most ideologically pure stack in the world won’t do anything if there’s not enough users to actually form a mesh.
WiFi is useless when you’re trying to send messages over long distances without any infrastructure beyond “I tied a few battery-powered transceivers to trees along the way”. It has completely low range and high power draw.
Packet radio gets you a lot of range but may require a license for legal operation. It also has high power requirements; you’re not going to run your radio setup off a 1000 mAh battery for a week.
Reticulum is pretty much developed by a single person. An immensely impressive feat for an individual but meshtastic seems to be developed by a wider group of people.
Also, if you want to use the true mesh aspects of reticulum without just tunneling it over the existing internet infrastructure then you’re going to want to buy some LoRa devices anyway.
you’re going to want to buy some LoRa devices anyway
Yes, but you’re not forced to; you can have nodes in your city that use any radio they want to communicate to each other, and e.g. your local hackerspace can have a node with multiple radios that bridges them to other nodes on the global network
With Meshtastic™®©, if your country bans LoRa™®© radios you simply don’t have any other option, so the whole network is just done. With Reticulum or any other agnostic network they can’t ban all radio modules that can be used
Reticulum is pretty much developed by a single person.
Hmm that’s unfortunate, I didn’t know that 😕 But that’s a chicken-and-egg and network effect problem; we shouldn’t be “forced” (network effect) to use something that is not ideal just because more people use/develop it, otherwise we will never have a better alternative, because no one wants to develop it because no one is using it because no one wants to develop it
At least for me, dedicating energy to build a Meshtastic™®© node would feel like I’m making something that profits LoRa™®© and Meshtastic LLC without trusting the “independence” of its every aspect. It transfers the dependence from the ISP that brings the wire to the home, to the companies that make Meshtastic™®© and LoRa™®©, but it’s still a dependence on one or two external companies instead of an independent community like I’ve seen with other local WMNs over the years
(I don’t have the experience to say that Reticulum is the best option, but it’s the main agnostic network I’ve seen with the little search I’ve done; people reading this feel free to make suggestions! 💚)
Meshtastic’s not really a company exactly. MeshCore exists if you don’t like it.
Semtech does IIRC have patents on and make all the LoRA chips. But the patents will expire eventually, and the LoRA chips they make are well-behaved modules that do what you tell them and not locked-down bits of nonsense that are a pain to work with. The data sheets for the modules are easy to find, and you’re not stuck messing around with firmware blobs you need to load into things. You can get boards with the Semtech radios on them from a whole bunch of manufacturers. You’d be hard-pressed to find a competing radio tech with modems available at a similar price point that we “should” be using instead.
And while the ability to use any link for e.g. Reticulum is nice, it also means that without coordination you have no idea what link you should use, and so you can never see anyone because you have no idea what technology or even what LoRA channel to look for peers on.
And Meshtastic now can go over UDP anyway.
You don’t have to pay meshtastic any money, so I’m not sure how they’re profiting.
Remember that no man is an island. There is no way to be completely free of dependence on others and that once you’ve bought a LoRa device, there’s nothing they can do to stop you using it as you see fit.
You don’t have to pay meshtastic any money
They can still profit indirectly from providing services etc (which is fine)
But even just the fact that in order to use the word “Meshtastic” ™®© I have to read https://meshtastic.org/docs/legal/licensing-and-trademark/ shows that it does not have “community” vibes but “Meshtastic™®© is ours and we’re just letting you use the source code etc for now” vibes
Again, the fact that it is owned by someone means that the community (probably) does not have control over it, and one day we might need to fork the whole network and migrate every node
there’s nothing they can do to stop you using it as you see fit
If a specific radio is illegal, it’s easy to just find where it’s transmitting from and fine you; they already do this with pirate radio stations
There is no way to be completely free of dependence on others
But why be dependent on 2 companies instead of having the option to buy a radio from any company? Why is competition and diversity bad for an independent and off-grid network that we don’t want it to have a single point of failure? 🤔
Not only it can make the network more resilient (which is supposed to be one of the goals), but it allows for experimentation and innovation in new technologies, which you can’t do if you’re locked into using LoRa™®©
Why lock every user into a single technology just because some users want to have a long-lasting battery? (Which btw is probably important for very remote nodes and not the home and portable nodes that I think are more common).
Power efficiency is extremely important for an off-grid network, because it translates more or less directly into battery and especially solar panel area requirements. You need the final node design to be hoistable up the tree.
Why do I get the feeling that I’m arguing a pointless argument with an LLM? I’m going to put the good-faith effort in for this comment but I’m not going any further.
They can still profit indirectly from providing services etc (which is fine)
Yes, that kind of approach can help ensure that open source projects are sustainably maintained.
But even just the fact that in order to use the word “Meshtastic” ™®© I have to read https://meshtastic.org/docs/legal/licensing-and-trademark/ shows that it does not have “community” vibes but “Meshtastic™®© is ours and we’re just letting you use the source code etc for now” vibes
It’s open source. You always run the risk that you might have to do a hard-fork of any open source software regardless of who maintains it.
If a specific radio is illegal, it’s easy to just find where it’s transmitting from and fine you; they already do this with pirate radio stations
Yeah, but that’s true of virtually all methods of communication. That’s a regulatory problem, not a meshtastic or reticulum problem. There is nothing specific about meshtastic/reticulum that makes it resistant to Government censorship. The best you could possibly do with reticulum is stick some messages on a USB drive and pass them to someone else - but then why not just FAT32 format the drive and send your friend some files? See https://en.wikipedia.org/wiki/Sneakernet
But why be dependent on 2 companies instead of having the option to buy a radio from any company? Why is competition and diversity bad for an independent and off-grid network that we don’t want it to have a single point of failure? 🤔
I feel like you’re arguing in bad faith here. I’ve been trying to help you understand some of the benefits/drawbacks of Reticulum and Meshtastic, not argue that one is inherently better than the other.
LoRa licenses the technology out. You’re not buying the device directly from them. It’s a standard, basically an identifiable brand.
Why lock every user into a single technology just because some users want to have a long-lasting battery?
The reason that meshtastic and reticulum are designed primarily to be work over LoRa is because Governments and businesses have done the hard work of setting standards and legislating free and open portions of the spectrum which end-users don’t have to pay to use. This opened up the realistic possibility of private medium-to-long range mesh networks existing in the first place.
Also, do you know that Meshtastic uses a queue messaging format which can be routed over UDP/TCP just like reticulum?
(Which btw is probably important for very remote nodes and not the home and portable nodes that I think are more common).
The main first use-case for LoRa was IoT devices where low power is a requirement. Think things like monitoring when gates are open/closed, what the soil temperature is, how much Nitrogen is in soil, etc. I think there are likely waaaay more low power nodes out there than nodes in people’s homes.
It’s great that you’re doing some research and have some ideas that you want to try out, but I think you could probably do with doing a bit more research to shore up your reasoning.
I know meshtastic devices are definitely floating around a ridiculous amount of places
Meshcore is pretty impressive, meshtastic is neat too. Not much in terms of bandwidth but the reach is incredible.
After having tinkered with both, I think I prefer MeshNode because it tries to reduce chatter and IMO provide a better structure, however Meshtastic is easier to get into as all nodes will act as a repeater too, reducing the amount of setup required to have a functional network. If you’re in a medium sized city, there is likely some MeshCore repeaters out there, so you can try it only with only a companion device.
MeshCore seems to have the ability to reach much further away by reducing chatter (64 hops vs 3 hops in Meshtastic)
I know the network as a whole goes a long way, but how far can a single message actually be sent? Don’t both have a maximum number of hops and a hop is only a few km?
It certainly seems interesting, but I am not sure if it could be useful for me. Especially when no one I know is going to buy hardware for it. Is there a reason these don’t extend off the internet? Always thought some kind of internet/LoRa hybrid could lower the barrier to entry as most have internet connected devices already, and exposing them to the network some will realise they could pay less by switching to LoRa instead.
Meshtastic gives you three hops by default (and strongly advises against going beyond that, even if the maximum is seven). The furthest node I see right now is about 200 km away.
MeshCore gives you 64 hops. They can afford that because MeshCore devices send way less telemetry by default; Meshtastic assumes that you want to broadcast e.g. your GPS position regularly. The furthest node I see right now is about 50 km away.
By the way, while LoRaWAN is a thing, neither Meshtastic not MeshCore are really designed for data traffic. Owing to the low (and shared) bandwidth, they’re more like IRC over radio.
Oh yeah I wasn’t really thinking of downloading porn over it, but internet at least gives a global network to connect into. Text communication is still a form of data and if it can be used normally day to day then you have more infrastructure that people are used to using and is still usable in a disaster as well, even if you lose access to all WAN nodes you should have local communication.
Meshtastic is one. It requires nodes you connect your self to with Bluetooth. Your mobile is the UI then. Nodes use 868Mhz and connect to each other in ranges between 10-20km. Building an own node costs about 70-90€. Check out https://meshmap.net/ to find one nearby.
Freifunk is another, which uses wifi only. It has entry/exit points to the internet on every node (but not required), using VPN tunnels to obscure the clients identity. Nodes can be linked to another in order to create a mesh network. I don’t know how popular this technology is outside Germany. https://freifunk.net/
Building a node can be done for under 20€ with a Heltec V3. Not the most power efficient, but not an issue if you’re running it from usb power anyway.
Also a note: the map doesn’t show all nodes, my node can see a lot more devices around me than are listed on the map. Maybe some issues with reporting or something (which is optional in the first place)
If you want a power efficient device with an enclosure, Seeed Studio make some decent stuff. I run a pair of Wio L1s (one on Meshtastic, the other on MeshCore), which cost around 40 € complete with antenna, battery, and enclosure.
Sure, more expensive than a bare V3 but more convenient for actually taking them somewhere.
How do those networks compare for you? I’ve considered both but in my area meshtastic seems to have a bigger community
I haven’t been on there a lot and due to time constraints I’m mostly just lurking. It seems that I can see more people on Tastic but Core seems to be more active in terms of messages.
Both are almost absurdly easy to set up, especially on an nRF5-based device where you can flash a firmware by just mounting the thing as an USB drive and copying a file over. You don’t really need to buy two devices to try out both unless you really want to use them simultaneously.
I know there are a few mesh routing protocols for openWRT.
Has anyone tried Batman or 802.11s?Hmm, just had a thought about getting this sort (meshtastic et.al.) of thing adopted more widely. People generally don’t need the full speed of their internet all, or even most of the time, they mostly buy a speed they can afford that makes their big downloads (e.g. games, updates, perhaps streaming) more tolerable. If a mesh system could aggregate a bunch of users connections they could probably buy cheaper plans knowing that when they need it they can have the speed.
Anyone knowledgeable in the ways of meshtastic or other such, does it, or anything else you know of fill this niche ? The wallet is always a good motivator…
Presumably ISPs etc would quickly make it against terms of service, but if there’s some bucks in it a lot of people won’t care…,
