Recently I rewrote it in Dart/Flutter and finally implemented RaptorQ codes (way more efficient than Luby used in original Txqr). Testing it internally now, prepareing Appstores/GooglePlay/Web deployment and new article.
I've been noodling on https://qr-send.com which is a slightly more polished version of the "erasure fountain codes + stream of QRs"-idea, inspired by divan's Txqr posts but using Wirehair FEC for the fountain code (basically: you receive ~file size bytes via QR codes and it magically assembles them into the source file regardless of missed codes).
It's an offline-first progressive web app and there are native & wasm builds for the sender. The browser-to-browser transfer falls up to WebRTC when possible because 30 MB/s over wifi beats a 100 kB/s QR stream. The QR scanner is a heavily-optimized WASM build of zbar, scanning at 60 fps on mobile & multiple QRs per frame (but it's finicky! Work in progress.)
This looks like a nice polished implementation of the idea, but when I try and use it, I get to "file complete" but then.. nothing? And I see no way to report a problem or bug.
Thanks for trying it, sounds like a bug I've been running into for a couple days where the wasm stream decoder does a bad memory access. I made some buffer lifetime fixes, next up a more complete stress test and fuzzing to see if I can pinpoint it.
Depends on the zoom. With this setup you can transfer about 0.1 B/s per pixel of 60 FPS video. So a 65" screen and 1080p camera at 10 meters away would max out at 2 kB/s with the normal lens (26mm equiv) or 45 kB/s with the tele lens (120mm equiv.)
I'm cooking something faster but depends on the job situation and funding whether I have time to spend on it.
Napkin math: QR codes encode 0.75 bits per module, each module needs about 3 pixels of camera resolution, and the temporal resolution is quite dodgy as well, maybe 0.25 * min(cameraHz, screenHz). So if everything is perfect, 44 kB/s at 60Hz per a 500x500 pixel patch. I've seen ~250 kB/s when a 1920x1080@60 transfer is working well. At 4k@30, you might reach 0.5 MB/s. If you throw in the 2x subsampled UV channels to transfer data as well, you might get an extra 50%.
Single page file transfer using QR Codes and a browser. Sending device loads a file into the page, gets chunked. Receiver gets all the chunks through a camera, tosses lightly and reassembles, CRC to garnish. Designed to push data from an old phone that had broken comms after it took a swimming lesson in a coffee mug, it's been quite handy.
Not OP, but I'm guessing by running the code on itself, i.e. turning the code into a QR code (or a series of them), then scanning those QR codes on the phone and reassembling them using a text-editing app on the phone.
I've wanted to use this for an air-gapped communication device.
I have a device with a camera and a touch-screen that only uses capacitive charging. I type a message. Bytes are encrypted. I hit send. QR codes flash on my screen. I use my PC or my normal phone to receive the encrypted bytes, and transmit them to you. You have the same device. You have your PC or phone flash encrypted QR codes. You use your device to receive, and then decrypt.
I've daydreamed about also buying several different hardware random noise generators. XOR all of their bits together. Save a huge one time pad to each of our devices. And then also use public key crypto on top of it.
I'm not really sure why I want this. But, it's my answer for how to reduce attack surface as much as possible, and have truly secret messages.
So two parts to a reply - first is, you don't need the encryption per se, but you can add that in the case that you give it some key and then it's encrypted. I don't see the value unless you're using this to generate frames for a video, which isn't current functionality but totally doable.
Second part, Charlie Bennet said "the only entropy source is one you can trust" and the best entropy source is quantum fluctuations, so we built a fully open source phase diffusion QRNG at Quantum Village and released it. Link: https://github.com/QuantumVillage/EntropyLoop
> I have a device with a camera and a touch-screen that only uses capacitive charging. I type a message. Bytes are encrypted. I hit send. QR codes flash on my screen. I use my PC or my normal phone to receive the encrypted bytes, and transmit them to you. You have the same device.
Why do you need a separate device for this and not just an airgapped computer?
Me, in my life, I have a PC that's connected to the internet. I have a phone that's connected to the internet.
I want another device, which I imagine to be a Pi or Esp32 or something with a camera and a touchscreen display, and capacitive charging. After I program it and give it the public/private keypair and the OTP, I imagine physically breaking off a USB port, or sealing one with some hardening resin.
I don't want an entire airgapped computer. Maybe you do, that's fine. For me, I'd love it to be a credit-card sized doodad.
I have a device with a camera and a touch-screen that only uses capacitive charging. I type a message. Bytes are encrypted. I hit send. QR codes flash on my screen. I use my PC or my normal phone to receive the encrypted bytes, and transmit them to you. You have the same device. You have your PC or phone flash encrypted QR codes. You use your device to receive, and then decrypt.
I specifically want this device to have no input or output hardware that could be used without my knowledge. IrDA could absolutely be used without my knowledge.
IrDA could absolutely be used without my knowledge.
Having actually used IrDA on Sony, Nokia, and Ericsson devices, no it couldn't.
In the real world, two IrDA devices have to be very specifically aligned, and also brought within just a few inches of each other. There's no way data transfer would happen without your knowledge.
I blame the intentionally cheap receiving optics. Researchers have reconstructed a TV signal from a hotel room with closed blinds, by very accurately measuring the ambient light intensity of the room (as shown on the blinds.)
Knowing that, I doubt that someone with even moderate funding would have difficulty receiving a signal from any of the transmitters you mention.
But, in all honesty, if you put a physical cap on the transmitter and receiver, maybe I'm wrong.
But in the other hand, none of the devices I currently own have one of your transmitters, and they all have screens and cameras, so...
Thanks for the dialogue and for sharing your experiences.
Is there a client that needs to be used to be able to receive files this way? I tried the dedicated QR reader on my iPhone, and the Camera app, and couldn’t get it to assemble the chunks. It will show the encodings one after the other, but nothing else.
I love this type of stuff. Some years ago I did something similar, but instead of QR Codes it used a convoluted mess of audio frequency modulation to send data through sound between devices. This is much more practical if you have two cameras.
I guess lmao, but much more rudimentary, less reliable and with loads of issues, as it had to blast piercing sounds through a speaker and then capture those with a microphone. But it was pretty cool when it worked!
Did you explore using frequencies outside the range of human hearing?
Amazon had modems very much like this in its little buttons that you could stick to your refrigerator and automatically order different items. When setting up the device, you could only hear the little clicks as it turned on and off.
I loved the technology. Hated that the prices changed all the time and you never knew what price you were going to pay ahead of time.
Cool! Out of curiosity, since qr-codes can contain binary data -- rather than base64, have you tried inserting the file as-is? That way you could do away with the ASCII separator and have a binary header as well. This would spend less frames for the same amount of data, but I'm not sure if it would be computationally cheaper. The other alternative would be the alphanumeric mode of qr-codes, but then you lose lowercase.
There is a toggle where you can show what chunks have been received. This is also where the 'show specific chunk' function comes in... the receiver can see "oh, I'm only missing chunk 125, so just show me that" etc. etc.
I once heard someone create a QR code scanner to retrieve gigabytes of data, but the biggest problem is that cameras aren't powerful enough to handle it all. Essentially, the QR code needs to be downloaded to the device for loading; relying on the camera to retrieve it is very difficult. Am I wrong about this project? What's your solution?
I've done a POC with the native QR reading code on iOS. The short answer is: it's not a problem at all, and you can drive very large QR codes for more efficient transfer.
I've done this exact approach before. It's a good way to exfiltrate data. Post the software on GitHub pages, or a popular CDN that co-hosts other shared libraries and you've got a very difficult to block method.
Really goes to show that it's very difficult to stop a motivated and informed actor.
I'm likely being overly specific, but blocking npm downloads, installation on corporate devices, etc is trivial in a restrictive corporate environment.
We used to be able to send arbitrary files between phones using Bluetooth. Where did that go? We had a bit of a music piracy ring going at school for a time. Good times.
What would make this truly portable is being able to generate this consistently with a short prompt and generate with a local LLM. That way no network calls or file hash can prevent this
[1] https://divan.dev/posts/animatedqr/
[2] https://divan.dev/posts/fountaincodes/
Recently I rewrote it in Dart/Flutter and finally implemented RaptorQ codes (way more efficient than Luby used in original Txqr). Testing it internally now, prepareing Appstores/GooglePlay/Web deployment and new article.
I've been noodling on https://qr-send.com which is a slightly more polished version of the "erasure fountain codes + stream of QRs"-idea, inspired by divan's Txqr posts but using Wirehair FEC for the fountain code (basically: you receive ~file size bytes via QR codes and it magically assembles them into the source file regardless of missed codes).
It's an offline-first progressive web app and there are native & wasm builds for the sender. The browser-to-browser transfer falls up to WebRTC when possible because 30 MB/s over wifi beats a 100 kB/s QR stream. The QR scanner is a heavily-optimized WASM build of zbar, scanning at 60 fps on mobile & multiple QRs per frame (but it's finicky! Work in progress.)
The WebRTC "fallback" basically means the QR code is just a handshake when both devices are on the same network?
I'm cooking something faster but depends on the job situation and funding whether I have time to spend on it.
Napkin math: QR codes encode 0.75 bits per module, each module needs about 3 pixels of camera resolution, and the temporal resolution is quite dodgy as well, maybe 0.25 * min(cameraHz, screenHz). So if everything is perfect, 44 kB/s at 60Hz per a 500x500 pixel patch. I've seen ~250 kB/s when a 1920x1080@60 transfer is working well. At 4k@30, you might reach 0.5 MB/s. If you throw in the 2x subsampled UV channels to transfer data as well, you might get an extra 50%.
I have a device with a camera and a touch-screen that only uses capacitive charging. I type a message. Bytes are encrypted. I hit send. QR codes flash on my screen. I use my PC or my normal phone to receive the encrypted bytes, and transmit them to you. You have the same device. You have your PC or phone flash encrypted QR codes. You use your device to receive, and then decrypt.
I've daydreamed about also buying several different hardware random noise generators. XOR all of their bits together. Save a huge one time pad to each of our devices. And then also use public key crypto on top of it.
I'm not really sure why I want this. But, it's my answer for how to reduce attack surface as much as possible, and have truly secret messages.
Second part, Charlie Bennet said "the only entropy source is one you can trust" and the best entropy source is quantum fluctuations, so we built a fully open source phase diffusion QRNG at Quantum Village and released it. Link: https://github.com/QuantumVillage/EntropyLoop
Why do you need a separate device for this and not just an airgapped computer?
I want another device, which I imagine to be a Pi or Esp32 or something with a camera and a touchscreen display, and capacitive charging. After I program it and give it the public/private keypair and the OTP, I imagine physically breaking off a USB port, or sealing one with some hardening resin.
I don't want an entire airgapped computer. Maybe you do, that's fine. For me, I'd love it to be a credit-card sized doodad.
Congratulations. You just invented IrDA: https://en.wikipedia.org/wiki/IrDA
Having actually used IrDA on Sony, Nokia, and Ericsson devices, no it couldn't.
In the real world, two IrDA devices have to be very specifically aligned, and also brought within just a few inches of each other. There's no way data transfer would happen without your knowledge.
Knowing that, I doubt that someone with even moderate funding would have difficulty receiving a signal from any of the transmitters you mention.
But, in all honesty, if you put a physical cap on the transmitter and receiver, maybe I'm wrong.
But in the other hand, none of the devices I currently own have one of your transmitters, and they all have screens and cameras, so...
Thanks for the dialogue and for sharing your experiences.
Like a modem
Amazon had modems very much like this in its little buttons that you could stick to your refrigerator and automatically order different items. When setting up the device, you could only hear the little clicks as it turned on and off.
I loved the technology. Hated that the prices changed all the time and you never knew what price you were going to pay ahead of time.
Really goes to show that it's very difficult to stop a motivated and informed actor.
Downloading a tiny JS from a CDN, or accessing a GitHub page is mostly noise, especially if obfuscated well.