bitchat!

[jackjackbits]

hi all. thank you for being part of this project. a bit of a background, and some notes on the path forward.

i did this as a weekend project to understand a few technologies: bluetooth mesh, store and forward and relay models, and various encryption models. this was one part of a series of challenges i've given myself: to build something every day that i didn't think i was capable of, and that i didn't think the current set of AI tools were capable of, namely block's tool goose (https://block.github.io/goose/). i'm always surprised at how much...just works. we have a new way of programming now, with an entirely new language and compiler: english and intelligence models.

the original goal for this project was to create a simple app, ala firechat, that allows people to communicate locally without needing to know someone else's phone number, email, or @name. think classrooms, airplanes, town squares, restaurants. it's the side-groupchat. it worked within a day. so i added private messages, channels and private channels the next day. that "worked" (see https://www.supernetworks.org/pages/blog/agentic-insecurity-vibes-on-bitchat#vibe-identity-an-obvious-flaw), but is in no way robust or thought through enough. i thought this was going to be mildly interesting, but it's turned out to hit a nerve, and something that a lot of people want to see actually become real.

so now that we have this community, we're going to do just that. we already have a working android implementation thanks to @callebtc, and lots of good PRs and issues to work through. i'm going to work on converting bitchat to use the noise protocol (http://www.noiseprotocol.org) to ensure much stronger and standard key management and encryption. i should be able to get to this over the weekend. serious PRs and audits always appreciated.

as for the AppStore release, that's blocked on three things: more robust network optimization (good suggestions here: https://x.com/tim_zaman/status/1942052775510335559), noise protocol implementation, and Apple approving. bitchat is currently stuck in review as apple is asking for things that are more suited for social media rather than p2p chat. for now, all we have is TestFlight and android.

after that, it's wi-fi direct implementation, stronger store and forward (with internet sync perhaps), and media.

and of course, bitcoin integration.

bitchat will always be free and open source software in the public domain. this technology needs to be free and accessible to all.

i'll keep updating here as we make progress. thank you all again for the attention on this project, and help along the way.

jack

[IanoNjuguna]

Great vision.

[director-of-chaos]

We're witnessing something that can become more than imagined.

[T756-Tech]

I think Jack you give me the key component i need to create the next internet thank you

[14XIV]

This is a revolutionary innovation 💡

[AIStutteringFix]

Hi @jackjackbits,
I’m Andrea Ceravolo, founder of INFOMYCITY OÜ – an Estonian civic-tech company.
We're building Help Me Now, PulseNet, and FlyLove, real-world mesh communication platforms (privacy-focused).
BitChat is amazing — we’d love to explore a collaboration or protocol integration.
Are you or the team open to a brief technical meeting or async discussion?
worldvanguardgroup@gmail.com
Thanks & congrats again!
— Andrea

[IsaacTalb]

yeh, the bitchat really help for those who have security concerns !! Love your journey!!

[chizy7]

Exciting!

[mzs114]

On andorid there is a way to bypass the app stores and use something like Fdroid(Android).

https://f-droid.org/

[Dylan933]

I want to join the test using the IOS system,but now it shows that it's full.
Where can i apply?

[aarsxx]

10/10

[matesus2]

Hi there, I love the idea of this chat. Bit I have aquestion. I have allowed all permissions and installed it on my friends phone and allowed all permissions but i could not see him in my network. Do i need to add his phone to list of my devices? Best regards, Matevž.

…

On Thu, 10 Jul 2025, 15:34 arsx, ***@***.***> wrote: 10/10 — Reply to this email directly, view it on GitHub <#139 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AJGE5S7VB6ZJFUL42T5LGTL3HZTXZAVCNFSM6AAAAACBGYZ42CVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGNZSGEYDCMA> . You are receiving this because you are subscribed to this thread.Message ID: <permissionlesstech/bitchat/repo-discussions/139/comments/13721010@ github.com>

[jaytechs977]

impressive

[Vinnybrunn00]

activate your cell phone's locator.

[NILAY1556]

Just Great !

[Achidon]

Lovely work Jack

[ComunionIO]

It would change the world again.

[mohnda]

Let's go!

[dayanruben]

Thanks for sharing this, Jack. Really inspiring to see the thought and intention behind the project. Excited to follow along and contribute where I can. Appreciate the openness and clarity on what’s next.

[gdiazgranados]

Love seeing this kind of project come to life — simple idea, powerful potential. The fact that it started as a weekend experiment and already resonated with so many says a lot.

Offline, local, peer-to-peer communication is something the world definitely needs more of. Big respect for keeping it free, open source, and pushing through the tough parts like App Store reviews.

Excited to see how Bitchat evolves. Keep going, Man — this is important work. 👏📡✨

[Ange0]

I really appreciate the idea, well done Jack, it's something I'll be following closely and contributing to in some way.

[Esalguero007]

Tech is moving in tremendous speed

[lf32]

Looking for Hybrid Version :)

[icfixx]

Hey Jack, what are your plans for buffering policy in the situation that the mesh changes and recipient is no longer reachable.

Also, how to protect account uniqueness. Ex. I set username to Jack, and someone else also sets the same username ?

[ingFavot]

sorry but why not already use a source code from www.nextvoix.com even if it is old but designed at the time to do this (chat and other APIs in ad hoc mesh networks completely offline in ble*bluetooth and wifi direct between smartphones) it already has a ready superstructure of many APIs to develop even if we are talking about the beginning of 2017, ok it needs to be fixed and worked on and do a porting and compatibility work with the most recent versions of Android and iOS, but the basic algorithms remain more or less the same (if they already worked well with the smartphones of 2016-17 we think today.. once fixed and ported with the recent versions of Android and iOS), you could already have the definitive version if someone worked on it

[paulocoutinhox]

Based on @jackjackbits message about Bitchat on X, he want learn something new and understand how it works, i think that is this because he don't want use in the project start something that is ready yet.

[cmosetick]

@ingFavot

Quote: "why not already use a source code from www.nextvoix.com even if it is old..."

We don't see any public code repositories on your GH profile. If you want to contribute something to this project, that would be a good starting position.

If you think that your nextvoix source code / project is useful or better than whats already happening here, your choices are pretty simple:

01: Release your source code to Open Source Community here on GH, like Jack has done here (Public Domain or some other OSS style license) LINK

02: Continue to try to promote and market your own nextvoix project.
...seems to have gone nowhere, based on what you wrote in this thread earlier:

Quote: "we funded everything out of pocket and eventually had to shelve the project"

Also, I want to mention that I just watched some of the videos that you have hosted on your nextvoid website. Indeed, you have some very impressive mobile Bluetooth-only functionality for nearly ~10 years ago if I understand everything thats happening in the videos correctly.

I appreciate that you demonstrate in the videos that the devices do not have SIM cards and wifi is disabled.

This one is impressive and really stands out (direct link: Connection 'offline' and broadcast message):
https://www.nextvoix.com/?wix-vod-video-id=deeb657da0944952b12e733e4b6bece8&wix-vod-comp-id=comp-k8fr6dyq

Please consider the 2 items I outlined above, before asking more questions in this thread like "why not use my source code"

[ingFavot]

Thanks so much for the kind words and the thoughtful suggestions! 🙏
If you’re curious, have a look at the other demo clips on nextvoix.com—they show many of the extra APIs we built back then (panic-wipe, GPS sync, store-and-forward, mesh extender, gruop chat , light sincro , music share etc..etc.).
Just for context: in 2015-17 there were no AI toolchains, so every line—including the routing algorithm—was handcrafted and refined over a year of tests and bug-fixes. AI today is a fantastic accelerator, but stitching BLE, Bluetooth Classic, and Wi-Fi Direct on phones that weren’t designed for true mesh is still tricky (Multipeer on iOS did make life easier even a decade ago, as you noted today is easy).
About open-sourcing NextVoix: I’d honestly love to, but I have a family to support and the project already cost us quite a bit. “I wholeheartedly embrace open source whenever it’s more than a mere showcase, but if we can’t make a living from it, we’re forced to change work and careers—as I/we did.”
The full codebase is ~30 GB (the APK’s only ~25 MB don t remenber exacly) and lives on an old USB stick rather than a repo. If you think it could genuinely help, feel free to DM me and we’ll figure out how to share it privately—AI could probably modernise it pretty quickly. Thank so much

[ingFavot]

Imagine we decided—purely hypothetically—to dust off NextVoix year 2016, the mesh stack we built a decade ago (the demos are still up on nextvoix.com), and bring it forward to Android 15 and iOS 18 or other. We might open-source the entire 30 GB codebase of 2016 year along with a serious white paper, then launch a clean branch called for example N.V.x or other name but with other modular APIs, modern crypto, and maybe a Bitcoin plug-in. At that point it wouldn’t just be offline chat: we could tap a sliver of the AI horsepower already sitting in phones and turn it into a massive, pocket-sized distributed computer.
Here are the (still hypothetical) numbers. Today there are about 6.8 billion smartphones; if only 1 % joined in, that’s 68 million devices. The latest NPUs hit roughly 35 TOPS; donating just 5 % of that gives 1.75 TOPS per phone. Multiply it out and you get ≈ 119 exaOPS—roughly fifteen times the power of Frontier, the current flagship supercomputer. In a month, that much muscle could process on the order of 3 × 10²⁶ operations, enough to match the training budgets projected for the next wave of AI models.
Storage tells a similar story: with each volunteer offering just 1 GB or less, you’d have ≈ 70 petabytes spread across the mesh—plenty for a global vector DB, federated model shards, or super-fast store-and-forward.
Of course, real-world hurdles remain—syncing millions of heterogeneous nodes, managing battery drain, guaranteeing privacy and security—but the raw scale is already sitting in our pockets. If that possibility excites you, let me know: with enough interest, a truly decentralised, privacy-first mesh with a dash of edge AI could move from proof-of-concept to everyday tool. Is only “It’s just a hypothesis—an idea for anyone who might be interested.”

[pedro-vicente]

@ingFavot Great idea, I think what everybody wants here on this thread is a global, decentralized, and censorship-resistant social media. Like NOSTR. Not many users, but it has published software and it works just fine. @jackjackbits

[ingFavot]

Unexpected interest has lit a fire under us: we’re keen to revive NextVoix, the mesh SDK that a decade ago already handled offline chat, file sharing and panic wipe. The vision now is an upgraded release that works only offline or in a hybrid mesh + back haul mode, and—when needed—also online over an extended mesh. If enough developers are on board, we could open source the full 30 GB codebase and a lean white paper. Quick math: with just 1 % of today’s phones (~68 M) donating 5 % of their NPU, we’d tap 34–120 exaOPS—many times Frontier—and ~70 PB of storage, enough for multimodal datasets and millions of mid size models. Think encrypted mesh chat, maps, offline payments and even distributed AI fine tuning right in users’ pockets. Tough engineering issues remain, but the scale is real. So here’s the invite: developers, researchers, makers, anyone curious—join the ride. Let’s build an open source model that also pays its contributors through fair licences, plugin revenue sharing and a decentralised marketplace. We proved this once; with NextVoix 10.x we can do it bigger. if interested, ping me at roberto@nextvoix.com. Moral support goes to a beautiful projects like Bitchat. Thank’s bye

[SubatomicPlanets]

Anyone like the idea of X implementing this or something similar? They are just releasing XChat, and I think it would be interesting if it had this functionality too. Since so many people have the X app, the mesh network would be huge. If they added more transports and maybe some relays or similar technology, then anyone could be able to send a message to anyone else without a direct internet connection worldwide!

I think @jackjackbits could contact Elon, and I also think it might be a good decision (if this is all technologically feasible). The problem with smaller messaging apps like this is that pretty much only developers know about them. But a lot of people have X.
Other messenger apps could obviously use this too (maybe even with some standardized protocol so that we could have one app talking to another), but I think X would be most open to it. Of course, it would also be nice as a standalone app, but it won't be as big. What do you guys think?

[moehamade]

Id rather do it a library, so other apps could make use of the chatting system

[cedws]

This app isn't the genuine Bitchat right? If not it would probably be a good idea to put out a message to tell people not to install this.

https://play.google.com/store/apps/details?id=dev.hostingduty.bitchat

[ddries]

What is the maximum throughput that can be achieved using bitchat technologies, a.k.a BLE? Or the order of magnitude? 10mbps? less than 1mbps?

[AIStutteringFix]

Hi Jack and all,

Just a quick suggestion: since there's some confusion around the Android app on Google Play (e.g. https://play.google.com/store/apps/details?id=dev.hostingduty.bitchat), would it be possible to officially confirm if this is the correct version?

Even just linking it in the README or a pinned comment could help avoid misunderstandings or security concerns.

Thanks again for this amazing project – looking forward to collaborating in the future!

— Andrea (INFOMYCITY OÜ)

[AIStutteringFix]

Hi Jack and BitChat team,

First of all, congratulations again on this inspiring project.
My name is Andrea Ceravolo, founder of INFOMYCITY OÜ (Estonia), the civic-tech startup behind platforms like:

• Help Me Now – AI-driven emergency & safety app already online
• PulseNet – offline mesh-based emergency communication app (BLE, Wi-Fi Direct, Store & Forward)
• Different Dating – queer community platform with built-in safety layer

We are actively looking for strategic collaborations to integrate or build on BitChat's technology, especially for:

• Privacy-first communication in crisis/emergency scenarios
• Secure, decentralized routing via BLE & mesh for NGOs or public services
• Federated or hybrid offline-online protocols for civil protection

Would you (or anyone in the BitChat community) be open to a deeper discussion or async technical exchange?

We’re based in Europe, and our systems are already being tested with partners in public safety and emergency response.

Feel free to reply here or reach out via email at:
📩 worldvanguardgroup@gmail.com

Thanks again for all your work and for keeping BitChat open & free.
Looking forward to hearing from you.

— Andrea Ceravolo
INFOMYCITY OÜ
www.infomycityhub.eu

[gdiazgranados]

Hello World...

[sylcvh]

is it possible to access bitchat on desktop?

[T756-Tech]

That's really interesting how about we develop that also so I am actually meant to release the android version I have been testing and making sure I don't realeas something which doesn't work well so if its desktop we can develop that.

…

On Tue, Jul 29, 2025, 5:43 PM Sylco ***@***.***> wrote: is it possible to access bitchat on desktop? — Reply to this email directly, view it on GitHub <#139 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BEAVRDDUJDJPH2VKNCC6JYT3K6CB5AVCNFSM6AAAAACBGYZ42CVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGOJSGM2TGMY> . You are receiving this because you commented.Message ID: ***@***.*** com>

[AIStutteringFix]

Hi BitChat team,

I'm Andrea Ceravolo, founder of INFOMYCITY OÜ (Estonia), a civic-tech company building platforms for public safety, emergency response, and privacy-first communication.

We are developing SafeMesh Connect, a decentralized mesh communication module using Wi-Fi Direct and Bluetooth Mesh. It’s designed for emergency scenarios, natural disasters, or areas without internet – enabling peer-to-peer communication, encrypted message forwarding, and secure broadcasting.

We believe there is strong alignment between our mission and yours.

We would love to explore a technical partnership or open-source collaboration around:

– BLE/Wi-Fi hybrid mesh networking
– Decentralized routing protocols
– Real-world deployment in public transport and safety apps
– UX integration for mobile and (possibly) desktop

We already have three active platforms:

• TravelSafe – AI-based public transport safety
• PulseNet – offline mesh emergency messenger (MVP-ready)
• Help Me Now – global emergency support app

Would you be open to discuss potential synergies?
Happy to share a 1-pager PDF and explore async collaboration (no call needed).

Thanks for your inspiring work – looking forward to hearing from you!

Best regards,
Andrea Ceravolo
Founder – INFOMYCITY OÜ
🌍 www.infomycityhub.eu
📧 worldvanguardgroup@gmail.com
🇪🇺 Estonia | Italy

[CheshireTechnologies]

Great concept! I've been trying to build my own mesh network (currently can't say why lol) but if it's cool, I'd like to try tweaking this to fit my needs