Reticulum is the cryptography-based networking stack for building local and wide-area networks with readily available hardware. Reticulum can keep operating even in adverse conditions with very high latency and extremely low bandwidth.

Unlike Meshtastic or MeshCore, Reticulam is not a messaging platform (though you can add messaging on top of it). It’s a full replacement networking stack. You can have many applications on top of it once it’s running. Ordinary TCP/IP applications do not automatically run on it.

In Reticulum, every machine or node is identified by a pair of signing and encryption keys (transferable). I fact a machine can have more than one. This is used for both encryption and routing. Both key pairs use 32-byte keys. Together, the two public keys form a 64-byte public identity. But to identify the node, Reticulum uses an identity hash, which is derived from these public keys.

hash = SHA256(encryption public key + signing public key )
identity_hash = first_16_bytes(hash)

That is 128 bits, or 16 bytes, or 32 hex characters. This is what you usually see everwhere to identify the node.

To route the traffic, Reticulum uses a destination hash, which is an address of an application endpoint in Reticulum. So different applications within the same node get different destination hashes. A normal identity-backed destination is derived from:

• Application name, such as lxmf.
• Aspects that identify the service, such as delivery.
• Identity hash, identifying the key holder.

lxmf is a messaging protocol built on top of Reticulum.

name_hash = SHA256(b"lxmf.delivery")[:10] # First 10 bytes
destination_hash = SHA256(name_hash + identity_hash)[:16]

For rncp (file transfer utility like scp), the destination name is rncp.receive:

• Application: rncp
• Aspect: receive

The calculation is similar:

name_hash = SHA256(b"rncp.receive")[:10]
destination_hash = SHA256(name_hash + identity_hash)[:16]

This is how different applications on the same node can have different destination addresses. Packets are delivered to these destination addresses.

Applications can send an announce for their destination. An announce includes the destination’s public identity information and a signature that others can verify. As it travels through the network, nodes learn which direction that destination can be reached from and how many hops away it is. If a node does not know a path, it can request one.

Reticulum uses Transport nodes to carry traffic between other nodes, sometimes across different interfaces. Not every node needs to be a Transport node. Ideally a stationary, always on and available nodes can be a Transport node. A network needs relatively small number of Transport nodes.

Reticulam doesn’t have IP addresses, DNS names, etc. Everything is handled by cryptographic keys or hashes. Traffic is also encrypted end-to-end. All encryption keys are ephemeral, and communication offers forward secrecy by default. All of this is handled at the network level, and you don’t have to do anything as an application developer. The Reticulam protocol doesn’t support store-and-forward; the packets will be dropped if the destination couldn’t be reached or is down.

The network can use available mediums (like links, called Interfaces) to ship the packets. Your network can be heterogeneous based on your design. It can use SerialInterface (RS-232/485 line, serial radio modem, or XBee-style transceiver), PipeInterface (pipes packets through stdin/stdout), Internet (TCP), Local LAN, Ethernet or WiFi, automatically using IPv6 multicast. Bluetooth or other radio (LoRa, VHF/UHF/HF using KISS, AX.25 ), etc. Basically, it can use any medium as long as they transmit packets. You can configure more than one interface.

To get started, we will install rnsd (Reticulam Network Service Daemon). I installed it using UV, you can do pip if you want. It will install the rnsd and other utilities. Once installed, run rnsd to create a key pair. It usually creates a key pair in your home directory at ~/.reticulum. All your network keys are stored here. You can also edit ~/.reticulum/config to add interfaces. It comes with AutoInterface for local LAN/Wifi connectivity. But you can add one or two TCPClientInterface so you can send packets across the internet. You can start with an Interface provided by the community. It’s the easiest Interface to set up. If you change the config at any time, you need to restart RNS.

# using pip
pip install rns

# using uv
uv tool install rns

# run rnsd to create keys
rnsd

After that, you can create a system service if you’d like to run it as a service. I want it to run all the time, so I set it up that way. If not, you can start it manually whenever you want. You can check the status and traffic flow using `rnstatus`. It should show the status of all Interfaces and the traffic flowing through each one. Now you are on the network and ready.

To test you can pickup a node on Rmap, like this and see the path to it by running the command.

rnpath 5ee1bfcf91d35528c9dc22c083ea0594 # destination hash

It will print.

Path found, destination <5ee1bfcf91d35528c9dc22c083ea0594> is 7 hops away via 
<d7db22f63b453c23bb0688dde565b7c1> on BackboneInterface[PSHP G01/dyn1.intatec.com:4242]

Now you know you are on the network and can reach others on the Reticulam Network.

Next, install a useful app that can do something on this network, perhaps short messaging?

I am making some small changes to these yearly posts. From now on, I will stop publishing the chart that I used to publish. I don’t think it makes sense to me anymore.

Anyway, this year has been good, I think. Overall, I’ve been working hard and smart at the same time, focusing on what I call being nice to my future self, in a way that my future self feels good. That includes working out today so my future self is healthy, saving a bit today so my future self is wealthy, and building relationships today so my future self is happy. Things like that. I’m not overly focusing on my future self, but not ignoring him either.

It’s kind of about building or doing things for my current self while also helping my future self. It also kind of works on a day to day basis. Like, I try to sleep well so my tomorrow is better. I try to eat well so my tomorrow is better. I try to finish one extra task so tomorrow is better. Things like that. Small things. That’s been the idea of this year, and I think I’ve done well in the shorter term. But in the longer term, I still have to see the results.

We are doing okay as a family. We are enjoying our parenting. Echo is getting older and older. It’s not easy to watch him get old, but that’s happening. Paatu has become an adult, but she still thinks she’s the youngest one. Uma is actually the youngest one, but she acts like a grown-up.

Anju and I are trying to figure out some things together. Generally, both my parents and Anju’s are getting old. So we’re seeing how to handle that, how to manage that along with our currently growing responsibilities at home, with the home and kids. I think we are doing okay overall. In terms of money, we are living comfortably and saving, though I would like to save a bit more than usual, given how the economy is going. We’re also trying to figure out how to manage our savings and investments better in the current situations.

I have been trying to take time out to meet friends this year. I have been quite successful. It’s been great; I want to keep doing that. I’m also very grateful to have them in my life.

I don’t think the future will be easy in terms of work or time. I can only say that I can prepare myself for it now that I know it’s going to be tough.

I’ve been quite healthy compared to last year. Anju also has been quite okay. I’ve settled down to a rhythm in terms of exercises and how my schedule is organized around that, and I’m paying attention to my health. So that way, it has been good.

In terms of work, I’ve been doing well. I’ve been working hard and smart, as I said before, but I’m also trying a lot of different things to stay engaged in this new work world.

I think I’ve been overworking a bit this year, which I’ll probably reduce a bit next year. But it’s been good; only because the teams have been good to me, and the work has been good, interesting, and challenging. Otherwise, I think it would have gotten to me by now.

That is all. Overall, it’s been a good year.

Hopefully, next year will be great because I have big plans, especially for motorcycling.

Hopefully, I’ll be able to write about it in detail soon, probably by December. I should have a plan for the ride worked out by then; that’s a big dream and target for next year. Some friends are hopefully joining me. Let’s see how it goes.

Otherwise, I’m grateful for the year. I hope to do better next year. Thank you and love you.

• I have been building many of my projects in public for a long time, but not all my work can be public, and not every team I work with works that way. Still, I’ve found a pattern that works for most teams and for me: something like open office hours, except we’re all working. I call it Katte – The Open Desk. We meet online at a specific time, in one or two hour blocks, ideally with 2 to 8 people, and work together or separately, talking only when it helps. It’s mainly meant for teams, but it can also work with strangers as a form of accountability. This week I documented it.
• A friend, Bharathi, at NPTEL/CODE turned 50. It seems like she is just getting started, even after achieving so much. She is such an inspiration to me.
• I attended a lecture by Peggy Mohan at APU today. It was brilliant and also fun. She is such a brilliant storyteller.
• We are making great progress at Dristi V2. The last two weeks were spent designing. It’s been build and build week. The next few weeks will be the same, I guess. Once it reaches some stage, I will share more details.
• Uthara and Dave from BuzzWomen wrote a book, The Buzz Way, covering how they work. Congratulations, Uthara and Dave.
• Tomorrow, October 10th, is a big deal; I am guessing based on how far the Delhi police will go to prevent the peaceful protests. If they were this proactive in general, Delhi wouldn’t have become the crime capital. I am very optimistic about the protests; they are going to be large, effective, and, above all, peaceful.
• NCF is 30 years old now. They are celebrating with lots of programming; attend if you can. Aleph Alpha has built Savanna, a model factory that implements the entire training pipeline in code so that models can be built collaboratively. They have also released their open-weights model, Kolibri. I haven’t tried it yet, but it looks interesting. I came across RoadStr, decentralized road-event reporting using Nostr and MeshCore. Web and Android. It’s very close to the project I had proposed at FOSSHack/2026. I am trying it in a browser now. I will definitely try it on OSMAnd using their plugin. I need to see if I can collect historical data for a specific location for research. Otherwise, it looks great. I am also studying and exploring the Reticulum Network. It seems useful and stable, especially when the bandwidth needs are small.

• Sunday, I went to FOSSIndia/2026. I met a bunch of people, did my BoF session on microgrants, and then also got Cardputer Adv (with Mesthstatic) from Kiran. Also met lots of other folks: Tanvi, Adhavan, Vivek, Aman, Rukmini, Arun, Noopur, Padmini, Pranesh etc. I have missed mentioning a bunch of people here. Sorry about that. But I might write different blog posts just covering the people and projects at FOSSIndia/2026.
• Tuesday, I left for Goa for a work offsite with Pucar. It was an amazing three days; I learned a lot. It’s fun to be in a room when people are debating. Just to watch, listen, and absorb. This was my first face-to-face meeting with the Pucar team. I really enjoyed hanging out with them.
• It was a short trip. I couldn’t do much other than work. I played a bit and went for a walk; that’s all. The tourist season has not started yet, so it was peaceful and fun.
• I also met Ajay Shah and Susan Thomas. We had a bunch of fun but enriching conversations. I will have to make plans to go to Bombay now, just to have longer conversations. Given the work I have now, it’s probably only possible in early 2027.
• It’s been two weeks of back-to-back travel. I don’t know how people do it.
• I played one game of pickleball this week; no runs, but I walked quite a bit and also did one P.T. session.
• Everyone is using AI at work. I think it is okay, but the negligence. is NOT OKAY. I found an AI chat prompt in a judgment by a Bangalore District Court. I have been exploring decentralized mesh networks. Meshtastic on LoRa is popular and seems very useful, but it’s limited to messaging. Reticulam Network is a network stack to build decentralized wide-area networks using heterogeneous communication modes. Currently, I am trying an app called Sideband that runs on top of Reticulam Network, which runs on top of the Internet. I will try other interfaces soon.
• Happy Mahatma Gandhi Jayanti to all of you. It was a long weekend for all of us. We took a short family trip to Hesaraghatta. It’s a do-nothing holiday. Pathu is enjoying her stay with other dogs; Echo is with us.

My idea is how to have more such grants. What might be stopping them, especially if they want to? The discussion was informal, like all BoFs. I made small notes. But here are the takeaways from the discussion.

1. A starter pack or playbook for new micro grant-givers, covering the problems I ran into, not just the process I settled on. The amount of work it takes, taxes, and legal questions are what worry people most. This was the biggest ask, and I think I can work on it.
2. A 10th-year report: an optional, light email to past recipients, then a publication of short updates, blurbs, or photos to encourage others. I don’t ask folks for updates, but a lot of them do update me anyway. Folks think some grant makers are especially encouraged by the effect. So this kind of report would encourage them. I am considering this.
3. Grants on specific themes and sponsors backing a theme, as FOSSUnited did. But in general, there could be theme-specific grant makers, and how do we make it possible for them?
4. Publish unfunded proposals so others can support them, and use the email list actively. I am trying this; there is more to do.

Other similar funds that came up during the discussion were Helium Grants,  Escape Velocity Grants,  Sudha Mahesh Grants, and Bellare Grants; most of them are covered in this wiki.

If you think there are other ways to encourage more small grant makers to come forward and make grants. Do share them.

This week has been tough but interesting overall. Lots of good conversations, both technical and non-technical. I hope to find more time to write next week.

• Uma wanted to get on a digger; she asked me. I said I don’t own it; if you want, you need to ask driver uncle. She called him loudly and asked him if she could come inside. He said, “Yes.” After a while, she asked, ” Can you start and move the hand? He said, “No.” I was just watching the conversation.
• This week’s P.T. was at odd times because I was traveling; thanks to Bhumika for accommodating. I couldn’t play this week, but I was able to run wherever I was, which is a good thing.
• The Indian Express has done a good series on the CEC and ECI of India. For me, the most worrying factor was that technology blocked what the law allowed; hence, 97 voters couldn’t be added to the voter list, and I think it’s a crime that should be punishable. Also, what was decentralized in process and pre-tech is now centralized by technology; i.e., voter list creation is decentralized but centralized technically, and storage and RBAC are centralized. Ideally, each state should own the platform with a separate installation and database. They should sync to the central one only for reporting purposes, as it used to work before. We are kind of losing it due to the centralized architecture design, and it goes directly against the Constitution.
• I was in Chennai to meet the team. It was a very short but hectic trip. I couldn’t get a train ticket, so I took a private electric bus. I must say KSRTC (Karnataka) has the best buses in the country.
• I like that Anand includes “Mistakes I made” and “Questions I was asked” in his weekly updates. Lume is a small, crazy-fast hybrid search engine written in Rust; you can run it locally or on your server.  

• I did a long walk/run with friends on Sunday, along with a huge kaalu soup (paaya) breakfast. Otherwise, it was a regular week with one game and one P.T. session. I should be back to two sessions next week. Also wrote a post about why I chose the Forerunner 965.
• The Hyphen Journal is here. I subscribed a long time ago, but their system somehow didn’t register my payment. So I had to ask customer support. Thanks to Pradeep for reminding me to subscribe. It was on my list, but as it often happens, I had made a mental note of it and promptly forgot about it. The journal looks pretty; I haven’t read it yet, maybe this weekend.
• Many times, I try to remember the sites I’ve already visited and struggle to do it. I have used bookmarks with tags, but it hasn’t worked because I forget to bookmark them. What I need is a search engine that remembers every site/page I visit and indexes it for searching. Turns out Hister is exactly that. It’s FOSS and self-hosted. Ducklake looks simple and powerful at the same time. It uses SQLite/PostgreSQL for the catalog and Parquet files for the actual data to build a data lake. WebMCP looks like a very interesting addition; it could be useful for websites that don’t support APIs or where the user wants to see things as they happen. Jeff sent me ebook2audiobook, which seems to support Kannada through Facebook’s mms-tts. I haven’t tried it yet; it’s on my list. Tante’s article on Power Grab and Fascism in FOSS is a must-read. There is a lot of fear-mongering about open-weight models, especially from the V.C. and S.V. community. I remember Bill Gates calling FOSS developers communists almost two decades ago. In that regard, open-weight models have already won.
• Kerala women MLAs walk out of a workshop where Hindi was the only language used. Why is it so difficult to run a bilingual event for a government organization when they are forcing kids to learn three languages?

Requirements

• Known and proven brands that have supported the hardware for at least a decade. And not known for dropping products. No Google, Apple, or Samsung.
• It’s meant for outdoor use, so at least semi-rugged. Buttons and not just touch screen. Good readable screen.
• Good battery life
• It should work without a phone for at least a few days. I.e., good offline functionality. It should work without a network for sure.
• GPS and Offline Maps. Ability to run GPS/Maps for at least 24 hours.
• Good community, openness
• Cost. I had a fairly high budget of 50K.

Solution

I looked into Garmin, Coros, and Casio.

I chose the Garmin Forerunner 965. Other watches I considered were the Garmin Instinct 3, the Garmin Fenix® 7X Pro, and the Coros Nomad. I considered the Casio ProTek (WSD-F20) series, but neither the community nor the OS seemed interesting enough. But I do love how rugged they are. I will keep an eye on them.

Garmin Instinct 3 has everything except offline maps. It has a cool screen and GPS. But it has no routeable maps. It’s also expensive. Garmin Fenix® 7X Pro is better in every way, but it’s double the cost and weight. Coros Nomad is the closest competitor; it has all the features I wanted, at a similar price, and it’s rugged and pretty. But I don’t know anything about Coros; no one I know has it. Also, I don’t know much about the community around it. It’s also the brand that is on my list.

I was able to load Indian OSM maps onto my Forerunner 965. I have used it while running and for casual walking. I haven’t done much routing on the watch, though it’s possible. I could follow a GPX track easily. I will blog more. Overall, these features seem to work.

Everything on the phone works offline and also works without the phone. Of course, you can send messages without the phone (it doesn’t have a network), but it can connect to Wi-Fi if needed. It can also store music and play it through my Bluetooth earbuds. So you can go out without the phone.

I have not charged the watch much. I am quite active every day, so it’s not bad. I haven’t taken it out for a full day of motorcycling or trekking with maps/routing on. When I do that, I will write more.

The community around it is good, though they are more focused on outdoor activities and mapping. I haven’t explored their apps yet.

As I said, this is not a review. This is just the list of things that I looked for while buying this watch, held up against them. For how it performs in those areas, you will have to wait. And it won’t be a single blog post. For now, there are no red flags with the Forerunner 965.