> step 5 is untrue over large distances such as across the Sahara
Yep. Focusing on local, social connections makes this less relevant.
> Step 6 is true, but does not necessarily lead to mass adoption.
Yeah, there's more scope for growth in places that don't already have abundant internet, where the pull is practical. But there are also a growing number of enthusiasts for whom the pull is principle, and they can help make mesh networking viable for poor people in rich cities.
> Step 7 is feasible, but costs a LOT of money at scale.
Step 7 is what you and I are doing right now :) — using the existing internet. This point was supposed to mean: the existing internet, can co-exist with and compliment mesh networks. Staltz's [Multi-modal connectivity](https://viewer.scuttlebot.io/%25OryfD6hGpjiqwFkKyFnhVPZMPsZF...) puts it more eloquently.
> it seems highly unlikely that people in developed nations are willing to accept drastic slowdowns in their connections.
Never underestimate the value of cheapness! I know some people who don't have home broadband; they just use their phone's data connection. It's slower, but it's adequate, and it's cheaper.
Mesh networking doesn't have to completely supplant ISPs. I want it to become a viable alternative, suitable for certain people depending on their priorities — like Linux!
> Caching might reduce this a bit, but is by definition not effective for "long tail" content.
If we're talking about data storage here (the flatbed truck — as opposed to data delivery, the series of tubes) then HTTPS can still be a highly-available option of last resort. If you want people to see your content, and no-one else is interest in replicating it, you can host it yourself or pay someone else to. Maybe use hashbase.io; they seem generous. If you mean data delivery, the packets have to traverse the internet somehow; it's just a question of whose box they go via.
Yep. Focusing on local, social connections makes this less relevant.
> Step 6 is true, but does not necessarily lead to mass adoption.
Yeah, there's more scope for growth in places that don't already have abundant internet, where the pull is practical. But there are also a growing number of enthusiasts for whom the pull is principle, and they can help make mesh networking viable for poor people in rich cities.
> Step 7 is feasible, but costs a LOT of money at scale.
Step 7 is what you and I are doing right now :) — using the existing internet. This point was supposed to mean: the existing internet, can co-exist with and compliment mesh networks. Staltz's [Multi-modal connectivity](https://viewer.scuttlebot.io/%25OryfD6hGpjiqwFkKyFnhVPZMPsZF...) puts it more eloquently.
> it seems highly unlikely that people in developed nations are willing to accept drastic slowdowns in their connections.
Never underestimate the value of cheapness! I know some people who don't have home broadband; they just use their phone's data connection. It's slower, but it's adequate, and it's cheaper.
Mesh networking doesn't have to completely supplant ISPs. I want it to become a viable alternative, suitable for certain people depending on their priorities — like Linux!
> Caching might reduce this a bit, but is by definition not effective for "long tail" content.
If we're talking about data storage here (the flatbed truck — as opposed to data delivery, the series of tubes) then HTTPS can still be a highly-available option of last resort. If you want people to see your content, and no-one else is interest in replicating it, you can host it yourself or pay someone else to. Maybe use hashbase.io; they seem generous. If you mean data delivery, the packets have to traverse the internet somehow; it's just a question of whose box they go via.