thanks for the info. i’ll take a look later, it sounds like an interesting idea.

“connect to expressvpn and run the test”… is that actually what you were doing?

i tried with a different VPN provider but yeah. understandable for people to not go to an unknown websites like the demo i posted, so a link to the code on github is provided.

IMO you simply can’t do anything reasonably useful here

i agree… im stretching the definition of what is “reasonable”. id settle some something clunky likes a popup for the user to remind them to be on VPN. its important to understand im putting efforts in determining what the safest and securest way is to use my app. i think it has the potential to be a private and secure chat app. the app works with webrtc, the key attack-surface is individuals and their peers not applying good security practices.

in traditional centralized chat system the central server will typically take responsibility for proxying request and consequently, the personal IP addresses are not needed to be shared. so having something like a toggle for “enforce VPN” might do nothing more than remind the user to turn on their VPN. but sometimes a reminder is what someone needs to use good security practices.

Thanks for the clear info! I’ll investigate this further within the demo.

Not cool. It’s hard enough trying to promote projects like this.

Unless you want to he honeypotted stay away…

I can’t be more transparent than open sourcing it with instructions on selfhosting.

I drew the logo on my phone. I’m not an artist.

this app is in javascript… completely different to java. you should ask in something like chatgpt, it’ll do a better job at explaining the difference than me.

id like to make it clear that the app is using webrtc which requires IP addresses to be exchanged which could result in IP address being exposed when using the public peerjs-server (hosting your own is an option). this app is explicitly NOT for anonymous communication. it explicitly shares IP addresses and data sent/recieved from peers cannot be moderated. the app is using cryptographically random ID’s for profiles to make sure they are unguessable. the connection details have to explicitly be shared.

the app is pretty experimental in how it works so there isnt any useful documentation on it apart from what is mentioned in the post. if you have questions, feel free to ask and i can try to explain the best i can.