3 ms·

So the way I think of IPFS (disclosure: haven't deepdived) is something like the world we could have if we had a broadly supported CDN protocol for the Internet

So the way I think of IPFS (disclosure: haven't deepdived) is something like the world we could have if we had a broadly supported CDN protocol for the Internet (this seems really overdue). This very naturally answers the repeated questions from people about the why of IPFS. Yes you COULD serve this via 1 or 5 of the other 20-30 incredibly individually complex stacks with CDN functionality as well. But why would you if there was a common CDN protocol? What would the upsides be if your ISP could offer you an opt in localish IPFS instance? IPFS is a functional expression of the long term academic research into creating an internet oriented around content blocks vs server socket oriented as it is today - https://en.wikipedia.org/wiki/Content_centric_networking https://en.wikipedia.org/wiki/Content_centric_networking We, of course, need both so its not an either or. We are backing our way into this world via javascript SRI - https://en.wikipedia.org/wiki/Subresource_Integrity https://en.wikipedia.org/wiki/Subresource_Integrity Look at that, we need blocks of code that we know is what we want but we want flexibility in where it comes from.

acdha 7y ago
I agree this is an interesting research area and it’s likely to stay one for awhile until the underlying economic blockers are well solved: storage, bandwidth, and support all cost money. If you host it yourself you have easy answers for that but if you’re relying on generous strangers you are going to hit limits: how many people are going to mirror large amounts of content, deal with DMCA takedowns and other legal requirements, etc. before giving up? Performance could in theory be competitive but there’s no guarantee and it’s notoriously hard to deal with unreliable nodes — which again can be dealt with if you have skilled support staff available. You’ll get some volunteers for the right cause but that only gets you so far – that’s why there’s a a couple decades of these projects dreaming big and failing once they get popular.

acdha 7y ago

I agree this is an interesting research area and it’s likely to stay one for awhile until the underlying economic blockers are well solved: storage, bandwidth, and support all cost money. If you host it yourself you have easy answers for that but if you’re relying on generous strangers you are going to hit limits: how many people are going to mirror large amounts of content, deal with DMCA takedowns and other legal requirements, etc. before giving up? Performance could in theory be competitive but there’s no guarantee and it’s notoriously hard to deal with unreliable nodes — which again can be dealt with if you have skilled support staff available. You’ll get some volunteers for the right cause but that only gets you so far – that’s why there’s a a couple decades of these projects dreaming big and failing once they get popular.