Hacker News Reader

Comment by runeks | Hacker News Reader

I don't disagree with the fact that IPFS has value. I just don't like the way it's presented as a generic transport protocol that's "superior" to HTTP.

View on HN

Just take a look at old torrents of obscure stuff and you'll see just how resilient bit torrent is. (it isn't)

It's more resilliant than HTTP: When you grab a torrent, you can download from any seeder. With HTTP, the host is specified, so if that host goes down, good luck.

In either case, if all the hosts go down and you didn't mirror it, you're screwed. However, IPFS makes it easier to mirror things.

While this sounds like it "should" be true. I think managing that will turn into something of a nightmare. IPFS is making awesome inroads into achieving that, but its not clear of the benefit over something like say freenet (other than freenet is very slow because it prefers privacy and resilience over speed).

I can see it being "really" useful as a backbone for something like Open Cobalt - http://www.opencobalt.net/ And I'm really looking forward to seeing that mature.

But the last time Open Cobalt seems to have been updated is 2010 - that's quite some time ago.

I wonder if it's being held up by patent trolls, but really all of them feels like more like a solution looking for a problem. That can take a very very long time to mature. Which is a shame.

Alpha 22 was actually released just last year, and it's still actively being worked on AFAIK. It's just that the website hasn't been updated, save for the downloads.

? Private torrent trackers do this very well.

Reply on news.ycombinator.com