4 ms·

I'm just curious why the likes of ipfs, namecoin, opennic, onion websites are not more popular than they are now among tech people. We dont like the existing in

I'm just curious why the likes of ipfs, namecoin, opennic, onion websites are not more popular than they are now among tech people. We dont like the existing internet. We have alternatives. Why are we not leaving in droves? Why are we all still...here?

gobengo 7y ago
How do I use any of those things to do social link discovery like on HN, reddit, etc?
rakoo 7y ago
Because the problem is not technical but societal/political ? We all know Google/Facebook/Twitter and friends are farming our data, but in return they offer all of us (techies or not) extreme convenience. We have pieces of the technology to replace them, but we don't have a sustainable way of maintaining it in the long term at such vast scales. Everyone is too accustomed to not paying anything on the web and using services with as little hassle as possible, so we won't be able to finance it large scale by making people pay for it, at least not directly. What business model is left ? I'm not desperate, I know something will emerge that is going to put the cursor further in the direction of decentralization and individual control. But nothing clear is there yet.
krageon 7y ago
> We have alternatives We have alternatives for serving flat, dumb content. But not for the things that really matter: Social discovery and discourse.
lidHanteyk 7y ago
They're not fast enough, and disks aren't free. People don't have just a few bytes that they'd like to store content-addressed. We have, as a commons, about 3-4 PiB of https://news.ycombinator.com/item?id=21752320data https://news.ycombinator.com/item?id=21752320data to store, at a minimum. This data cannot possibly live in one place, it cannot possibly be checked all at once, and it cannot possibly be enumerated or searched. These all operate over the existing Internet, by the way. There are proposals like NDN [0] for doing something different from TCP/IP, but odds are good that we don't need NDN, we just need less bad content-addressable systems. I suspect that we'll need either a compression revolution, or an ISP revolution, in order to get more speed. It doesn't seem like existing distributed systems can improve on their use of local CPUs much more. [0] https://named-data.net/ https://named-data.net/

gobengo 7y ago

How do I use any of those things to do social link discovery like on HN, reddit, etc?

rakoo 7y ago

Because the problem is not technical but societal/political ? We all know Google/Facebook/Twitter and friends are farming our data, but in return they offer all of us (techies or not) extreme convenience. We have pieces of the technology to replace them, but we don't have a sustainable way of maintaining it in the long term at such vast scales. Everyone is too accustomed to not paying anything on the web and using services with as little hassle as possible, so we won't be able to finance it large scale by making people pay for it, at least not directly. What business model is left ? I'm not desperate, I know something will emerge that is going to put the cursor further in the direction of decentralization and individual control. But nothing clear is there yet.

krageon 7y ago

> We have alternatives We have alternatives for serving flat, dumb content. But not for the things that really matter: Social discovery and discourse.

lidHanteyk 7y ago

They're not fast enough, and disks aren't free. People don't have just a few bytes that they'd like to store content-addressed. We have, as a commons, about 3-4 PiB of https://news.ycombinator.com/item?id=21752320data https://news.ycombinator.com/item?id=21752320data to store, at a minimum. This data cannot possibly live in one place, it cannot possibly be checked all at once, and it cannot possibly be enumerated or searched. These all operate over the existing Internet, by the way. There are proposals like NDN [0] for doing something different from TCP/IP, but odds are good that we don't need NDN, we just need less bad content-addressable systems. I suspect that we'll need either a compression revolution, or an ISP revolution, in order to get more speed. It doesn't seem like existing distributed systems can improve on their use of local CPUs much more. [0] https://named-data.net/ https://named-data.net/