Hacker News Reader

Comment by brian_herman__ | Hacker News Reader

My brother worked on freenet it was cool until tor came up and supplanted it

Freenet still does a lot of things better than Tor. For example, no server is involved so there's nothing to attack or track down, making anonymity guarantees seem to be much better. Yet they've still managed to pull off some forms of interactive communications.

IPFS is closer really, but doesn't move in an anonymity friendly direction at all.

> For example, no server is involved so there's nothing to attack or track down, making anonymity guarantees seem to be much better.

Not really "no server". I mean, Freenet users collectively are the servers. As in BitTorrent, I2P and IPFS.

What Freenet does better than Tor is keep content available, even if the provider goes away. As long as it's popular content, that is. Although Tor .onion services are perhaps harder to find, once they're gone, they're usually just gone. There are exceptions, of course, such as the The Hidden Wiki. But any mirroring that happens is entirely ad hoc. In Freenet, it's automatic.

The point is, Tor relies on establishing a connection with a server which will provide content, Freenet simply distributes data and that data can come from any client/server on the network, Tor's hidden services are modeled off the traditional client server model, while Freenet operates more like a P2P network.

That difference in model is what provides a potential anonymity benefit as there's no longer a path to the origin of the content

IPFS is indeed more similar. Tor is more of a compliment to a DFS such as Freenet (though Tor can very much works as a DFS itself)

No, not tor. I guess ipfs would be somewhat similar today, but it has its own problems and I don't see it succeeding either.

Care to elaborate problems of IPFS? I just recently set up a node (which I run while I heat my apartment with GPU mining) and am wondering if there's some other project I should give my spare cycles to.

Well, for one thing, they don't care about anonymity at all. And already implemented a mechanism to blacklist content hashes on the demand of law-enforcement agencies. Both these things essentially defeat the purpose of many enthusiasts who would use it -- and its only function remains that of a volunteer-driven p2p network for making sure certain data doesn't die. And that's it, nothing else. Even that is somewhat ruined by paid pinning services.

For many IPFS is enough for what it does. But in a world where WikiLeaks and Edward Snowden are facts of life, it falls vastly behind on what's needed out there.

I'm not an expert on IPFS, but the things I personally don't like about it is the lack of anonymity (although I think it is possible to run it over Tor), and also it seems to only be useful as a store of static files.

I don't think it's possible to do a dynamic site, eg something that needs to read from/modify a database, using IPFS.

Not sure what project is better however. Dat looks pretty cool but also doesn't have anonymity.

Reply on news.ycombinator.com