[Development Project] A New Era: IPv7

[bardi.harborow]

I recently moved house and organised an Internet connection and was reminded of the litany of associated protocols. The phone/internet system is a botch of systems all working together. Think ADSL1/2+, Naked DSL, VOIP, Fibre, Cable, etc. Another example of this protocol mess is the internet. We have HTTP, FTP, SSL, DNS, TLS etc. and on the p2p side: BitTorrent. Delve deeper you get thousands of old protocols. If you go and look at the Wikipedia page for Push technology, there are about twenty different ways to longpoll.

The point I am trying to make is that computers are becoming increasingly complex and maybe it is time to rebuild from a clean slate. Vint Cerf (Chief Internet Evangelist for Google) proposes something like this in an article here, where he suggests an "InterPlanetary Internet", a delay-and-disruption-tolerant network.

I'm starting a new project, a new operating system that will attempt to standardize all aspects of computers, and be adaptable to any device. Specifically, I'm interested in:

* More Secure DNS (maybe based on NameCoin)
* Standardised Protocols for Voice, IM and Email communication
* All communication is encrypted.
* Etc.

If you want to look at this another way, there have always been those who have stayed two steps ahead of the authorities. A few years ago, those users were Tor users. Now it is time to move on and pioneer a new network.

Comments, criticism, assistance and suggestions needed. I also need developers. The purpose of this post is to attract such help.

<hr>

EDIT:

After some thinking I've decided to try and re-explain what I'm aiming for.

Design a new internet that has a emphasis:
    on P2P technologies; including an inbuilt DHT system and P2P DNS system (NameCoin maybe?)
    end to end encryption for all connections
    P2P email, IM, VoIP and video conferencing (all encrypted)

I've also had an idea for a new type of login system:

Step 1: The service signs your PGP (for example) key with a special signature that contains your username.
Step 2: You can authenticate in a similar way to client side certificates.

We may be able to use plain IPv6 for this, can somebody think about this?

The end product I'm thinking of is a modified version of Linux that supports this. Eventually, we may want to create a new operation system for this.

[jasinlee]

I recently moved house and organised an Internet connection and was reminded of the litany of associated protocols. The phone/internet system is a botch of systems all working together. Think ADSL1/2+, Naked DSL, VOIP, Fibre, Cable, etc. Another example of this protocol mess is the internet. We have HTTP, FTP, SSL, DNS, TLS etc. and on the p2p side: BitTorrent. Delve deeper you get thousands of old protocols. If you go and look at the Wikipedia page for Push technology, there are about twenty different ways to longpoll.

The point I am trying to make is that computers are becoming increasingly complex and maybe it is time to rebuild from a clean slate. Vint Cerf (Chief Internet Evangelist for Google) proposes something like this in an article here, where he suggests an "InterPlanetary Internet", a delay-and-disruption-tolerant network.

I'm starting a new project, a new operating system that will attempt to standardize all aspects of computers, and be adaptable to any device. Specifically, I'm interested in:

* More Secure DNS (maybe based on NameCoin)
* Standardised Protocols for Voice, IM and Email communication
* All communication is encrypted.
* Etc.

If you want to look at this another way, there have always been those who have stayed two steps ahead of the authorities. A few years ago, those users were Tor users. Now it is time to move on and pioneer a new network.

Comments, criticism, assistance and suggestions requested needed. I also need developers. The purpose of this post is to attract such help.

Interest piqued.

[bardi.harborow]

Interest piqued.

Good to hear. I'm trying to raise support in the hope that we can get a team working on this. I'm also going to cross-post this on Reddit.

[jasinlee]

This is in sync with some of the future plans we had in mind.

[sdp]

Every attempt to start with a "new slate" often creates a competing standard adding to what you described in the original post.  It is a bit like reimplementing bitcoin-qt.  You'll need a team as good as the core dev team to create the software of said quality.  You'll need to spend thousands of hours on that.  This Bundle Protocol, is not IPv7.  It's completely new.  Why would you go to all this trouble?  Are you being paid for this?  I wouldn't bother, if not.

sdp

[bardi.harborow]

Every attempt to start with a "new slate" often creates a competing standard adding to what you described in the original post.  It is a bit like reimplementing bitcoin-qt.  You'll need a team as good as the core dev team to create the software of said quality.  You'll need to spend thousands of hours on that.  This Bundle Protocol, is not IPv7.  It's completely new.  Why would you go to all this trouble?  Are you being paid for this?  I wouldn't bother, if not.

sdp

When I mentioned the Bundle Protocol, I was just referring to an example of the "clean slate" idea, not saying I was think of getting involved with it. Also, I'm happy to use pre existing protocols if they are good enough. I've also edited my OP with more explanations.

[Hideyoshi]

https://xkcd.com/927/

[sdp]

See SQRL at http://www.grc.com for a login system.