About Internet

Probably....many people thought that IPv6, which we’ve been fighting to implement for about a decade, will solve the Internet’s scalability and security problems.
The experts say that is wrong, and how we might go about really fixing these problems may be critical to the future of the Internet. We’re not doing very well so far. The Internet Engineering Task Force (IETF) hasn’t reached agreement on one of the issues, and it isn’t even involved in the other.
Let’s start with scalability. Many of those experts believe that we’re not running out of IPv4 addresses but out of router table space. The currently favored solution is called the “location/ID split,” and what it does is create a new hierarchical space that uses one set of addresses to define “hosts” and “applications” and another independent set of addresses to define how to get to one of those hosts or applications.
By creating a separate address space for the “where-it-is” part of Website identification, you can assign addresses based on the geography of the Internet, so routers don’t have to know where every individual IP subnet is located -- just the area that contains the right one. This is the same principle used in telephone networks, and it could simplify routing tables and the already-too-complicated Border Gateway Protocol (BGP) used among ISPs.
But some experts object to the location/ID split (in whatever form you deploy it) because it builds on the “old” Internet rather than taking a fresh perspective. The Pouzin Society (named for Internet pioneer Louis Pouzin, but seemingly driven by John Day, another Internet pioneer) would largely rewrite the protocols of the Internet around a computer-to-computer “inter-process communications” model, where applications establish relationships with each other but traffic for these relationships is routed over a network with its own private system of addressing.
This is a very cloud-friendly, multimedia-friendly approach, but it’s definitely radical, requiring we rethink (and often re-execute) how the Internet works.
On the security side, IPv6 doesn’t address the unfortunate fact that cloud-hosted data and other private data is subject to theft if the repository is hacked. Encrypting the stored data isn’t helpful, because traditional encryption protects against theft of the drive but not hacking of the software. (That's because the hacking uses the application that knows how to decrypt the data and gain access.)
But now a new development from an IBM expert lets applications operate on encrypted data without decrypting it at all. The data is never visible, even to an application, in decrypted form.
One problem with this new system is that it’s pretty CPU-intensive; it looks to me as if you’d need a special microprocessor to make it work. At the server side of the cloud application, that’s not necessarily a problem, though there are no such processors today.
Another problem is that, while you may be able to encrypt the data, it’s also likely that in some cases you’ll need to display it. I can encrypt an account number with the new system, but if I have to display it to the user it will have to be decrypted, period. To avoid compromising security by sending it in the clear, I’d have to equip client devices with the new system, too. Since we can hardly throw out PCs and smartphones in use today and start over, some more incremental approach is needed here, and nobody has suggested one.
Just these two issues demonstrate that there are a lot of proposals that hope to advance the Internet to the next level, but not much agreement on any of them. The location/ID split debate has already created a half-dozen competing approaches, and the security/encryption solution hasn’t even been launched as an Internet proposal. Couple this with a more-than-a-decade delay on IPv6, and you have to start to wonder whether we can still implement good Internet ideas.
That’s a real problem. Telephony became the inflexible monolith, not because of superior technology, but because of the inability to effectively promote change and growth. The IETF needs to consider taking steps to prevent the Internet from sharing that fate. Perhaps an IETF-sponsored focus on getting a single proposal for address scalability and data security could fix both problems decisively, and show us how to fix more that will surely arise in the future.
I hope this article useful for you.