Hacker News Reader

Comment by clueless404 | Hacker News Reader

Just exactly how does IPFS solve content distribution?

You still have to serve the content from a server, since you cannot depend on the kindness of strangers to persist your content.

View on HN

The idea is that some fraction of users will be configured to reshare content, which helps distribution for popular content scale. This seems to work in practice for bittorrent.

Yes, but only for suitably popular content, as all the dead torrents will testify. This doesn't make IPFS very appealing to Joe Blow, content maker unextraordinaire.

Joe Blow would just need to find some fans who think that keeping his content available for others is worth their while.

I believe this could work amazingly well for netlabels and indie bands.

Regardless of whether I think this is a bit of a far fetched use case or not, this really does nothing for Joe Blow and his content distribution needs.

But let's roll with it anyway. How does IPFS solve this problem, and more specifically how does IPFS solve this problem any better than just publishing a torrent of Joe Blows collected ramblings?

Sorry, I don't really care about your theoretical questions. IPFS is nice and fun technology. If you don't want to participate, you don't have to.

> theoretical

I don't think that word means what you think it means.

Asking what problem a technology solves is a very practical question. Countering with "IPFS is a nice and fun technology", is the very opposite, i.e. alluding to that there might be some theoretical benefits to it, but that your mainly into it for gits and shiggles.

IPFS isn't meant to provide hosting. It's meant to help you scale distribution. If Joe Blow's content isn't that popular, he doesn't need IPFS anyway.

Ok, so IPFS isn't for hosting and it's not for unpopular content.

What is it for then? For scaling distribution, you say. What problem does it solve? How, what and why? Is it better than BitTorrent for that use case?

it's better than bittorrent because you don't need a centralized tracker.

Also content can be incrementally updated without duplication.

Ok, when or why would you use IPFS instead of something else?

lol, you can troll better than that, c'mon :)

Are you just unable to answer the question or are you just so new to the interwebz that you think that was a troll?

ZFS is less flexible in deduplication because it's not true content addressing, and git has lousier protocol. Rsync and unison also have lousier protocol and don't deduplicate storage.

That's not much of a benefit, as there are tracker less torrents.

Reply on news.ycombinator.com