Internet 2.0
Computer networks used to be exclusive to large organisations and they used to be very inaccessible and unreliable due to the circuit switching principle that meant a single point of failure in the central system could bring the entire network down.
As time went pass a packet switching system was introduced and no single central hub was required, information could flow freely though any best path that it could find. This was revolutionary at the time, especially for sending messages and emails directly between two computers, because it avoided central congestion and failure points. But now that the problem of how the information FLOWS across the network has been solved, another big problem that needs to be solved is how information is STORED across the network. This is still mainly done centrally, with points of failures, with congestion, vulnerability to censorship and bandwidth usage that multiplies with every user that streams the same content. There are elegant solutions to these problems.
The biggest part of the Internet that is still centralised is the "address book" that stores all the IP addresses for websites, namely ICANN. It is controlled by just a few so it is also a dangerous censorship risk. This can be distributed with a blockchain based DNS system like Namecoin.
Despite the DNS system, most of the data is also stored centrally when you visit a Web page or watch a video. It is a problem because the amount of bandwidth used is copied for each user along the entire path right up to the central server. It is very inefficient and wasteful and not to mention unreliable. The information goes through many nodes to reach you and often the data that you are getting is already on the cache of a device near you, so why not just copy that by getting it directly from the device closest to you that already has it?
Up to now, the reason for this has been security,
If you go to internet explorer options, content, certificates, Trusted Root Certification Authorities, select Verisign, View, Details and finally Public Key, you can see the Windows preloaded key that is used to check digital certificates issued to websites for authenticity. These certificates can be chained together in a hierarchical fashion so that certificates issues to intermediate organisations can be used to create certificates for other websites so that one central authority would not have to issue all the certificates to websites themselves. So, the question here is, if you can do this for websites, can’t you do it for devices too? Can’t you chain devices together too so that once device can verify the integrity of the data on the next device and so on?
Of course you can, this is basically what IPFS does and IPFS is the solution to the problem with the centralised data storage and distribution on the internet. It is the same technique used by other P2P technologies too: Git source control, Torrents, Bitcoin, to name just a few.
Another advantage of this kind of system is that not all the devices need to be online all the time