While I do like the idea behind a P2P E2EE chat, I believe that unless you're willing to invest heavily into OrbitDB and IPFS, this project will stay niche at best.
The performance issues that come along with running OrbitDB/IPFS on a machine, let alone a mobile device, are still significant unfortunately. Adding Electron on top of what is already a heavy-weight application is probably going to make people's devices go brrrrr all the time. Not only that, but I would argue that for instant communication this stack might not be the best idea in terms of performance in first place.
Besides, the way IPFS has been (and still keeps) changing their dozens of libraries doesn't make development particularly smooth either. OrbitDB is always behind the latest IPFS version due to all these changes that are being introduced. Hence unless you're planning to allocate developer time on these two things as well, my guess is that you probably won't have too much fun with your back-end.
The integration with Tor is another thing that will likely be a time drain for developers, as other people here already pointed out, and that will lead to even more performance issues down the line.
Don't get me wrong, I really like the idea behind this project. However, I feel like the aspirations are unrealisticly high and the actual outcome might be realtively frustrating for the average end-user. Having that said, I would love my gut feeling to be proven wrong!
Disclaimer: I'm the developer of Superhighway84 (https://en.wikipedia.org/wiki/InterPlanetary_File_System#App..., https://github.com/mrusme/superhighway84), a USENET-inspired, uncensorable, decentralized internet discussion system running on IPFS & OrbitDB.
The performance issues that come along with running OrbitDB/IPFS on a machine, let alone a mobile device, are still significant unfortunately. Adding Electron on top of what is already a heavy-weight application is probably going to make people's devices go brrrrr all the time. Not only that, but I would argue that for instant communication this stack might not be the best idea in terms of performance in first place.
Besides, the way IPFS has been (and still keeps) changing their dozens of libraries doesn't make development particularly smooth either. OrbitDB is always behind the latest IPFS version due to all these changes that are being introduced. Hence unless you're planning to allocate developer time on these two things as well, my guess is that you probably won't have too much fun with your back-end.
The integration with Tor is another thing that will likely be a time drain for developers, as other people here already pointed out, and that will lead to even more performance issues down the line.
Don't get me wrong, I really like the idea behind this project. However, I feel like the aspirations are unrealisticly high and the actual outcome might be realtively frustrating for the average end-user. Having that said, I would love my gut feeling to be proven wrong!
Disclaimer: I'm the developer of Superhighway84 (https://en.wikipedia.org/wiki/InterPlanetary_File_System#App..., https://github.com/mrusme/superhighway84), a USENET-inspired, uncensorable, decentralized internet discussion system running on IPFS & OrbitDB.