I2p is notoriously difficult to setup up for the average user as well as being pretty slow right?
It's also 20 years old, do you think the core i2p code would need to be retackled in order to make it more available to the average user and at that point would it be more advantageous to start from scratch?
There is i2pd written in C++, it's slow because of lack of participants and level of encryption and layers i2p offers in order to provide total anonymity.
Old doesn't mean bad, it's improving, the older better. It's still actively maintained and tried.
I have no doubt that it's well maintained and well tested, but how does this provide benefit to an every day intenet user.
The setup for i2p is virtually impossible for 98% of internet users. How do we apply i2p benefits to the 98%. This is the rewrite that must be done. The code is solid and works yes, but it's the social application that needs to be rethought.
The absolute greatest people you can target with i2p are torrenters as it proves the MOST benefits to them immediately and you generally don't even see any of them adopting i2p. This is the problem that needs to be thought through.
But I must joke a bit: did you ask the same question to the I2P devs when they forked Freenet? :P
Seriously though, they are different projects with different goals and vastly different functionality. The only overlap is "privacy".
There simply isn't anything like Web420, and I think this is a better approach to getting the average Internet user away from the big tech/VC-backed web apps that are ruining the world, that's why.