The open consultation for the Next Generation Initiative (NGI) was held between 10th November 2016 and 9th January 2017. 449 people took part and provided their views both on technologies and values that are relevant for the Internet of the future.
Please read more about the results in the Final report of the Next Generation Internet Consultation.
You can debate them and provide additional input via adding comments.
The report has lots of good wishes but is thin on the direction of technology development. The key area where the current Internet fails to act as a critical infrastructure for the society is CyberSecurity. From both the engineering and business point of view, it seems that we should first and foremost improve the Internet in this respect. It should become impossible to do malicious acts against other users and hide behind obscure identities. It should be impossible to flood entities with malicious packets (DDoS) - this is a prerequisite of providing reliable services over the network. It should be possible to trace back all the resources used in attacks automatically. It should be possible to offer end system security services in a rich environment using cloud assisted methods such that the end user can delegate responsibility for its security to a service provider that will then always use the latest and most up to date software to enforce security. It should be possible to effectively cooperate between benevolent entities for the sake of security against all the bad guys as agreed between the nice people.
All this should be embedded in the architecture itself and not provided as an after thought by all kinds of ad hoc methods.
The question of IPv4 vs IPv6 is irrelevant - these are just forwarding protocols - both of them and others should be supported. Architecture is about the functionality residing in different places - protocols just give the syntax's for the functions to talk to each other.
I agree that IPv4 vs IPv6 is irrelevant, but IP vs other ways of forwarding packets is definitely relevant. Anyone can inject packets with a given destination IP address into the system, and the system should then deliver them to the interface identified by the address. Devices such as firewalls are very definitely an "ad-hoc afterthought", but with IP's connectionless paradigm they are the only defences we have. Other technologies can do a better job; see http://www.ninetiles.com/Flexilink_details.html for instance.
It seems to be missing in the report the concept of incentive for the users to sustain the NGI networks
As a simple example most of P2P networks failed because most of the users are freeriding and have no interest to sustain them, except the bittorrent network where users, even if freeriding also for most of them, are "forced" to sustain the network when they use it (ie when the incentive is to download)
Another similar aspect concerns the blockchains repeatedly mentioned in the report but without addressing the recentralization issue that most might face due to the lack of incentive for peers to run full nodes
And this is the same for applications providers over the NGI networks, how to deploy quickly/widely the apps securely (email, chat, social network, file sharing, data storage, etc) and be rewarded for it
This is what this proposal http://www.peersm.com/Convergence.pdf was proposing to cover (this was submitted to the CSA ICT-12-2016 H-2020 call that we misread, therefore was not eligible, while unfortunately no other calls could match it and still can't today), among others, since many topics appearing in the report are covered by this proposal
This is a beautiful research programme!
I am an active member of the IFIP's Domain Committee of Cloud Computing, and we recently released a policy statement entitled "Towards a Human-Centred Cloud Computing: an International Perspective on the Public Interest”.
Read more at: http://www.ifipnews.org/ifip-cloud-experts-release-policy-paper/
I think this is fully in line with the ICT-41 initiative.
I didn't see too much discussion about what physical infrastructure the NGI would take i.e. broadcast stations using WiFi as opposed to cables in the ground, or other more innovative methods. When WiFi was first developed we had hotspots where we could connect and the vision was to transition to something less cumbersome and more ethereal. I imagine a worldwide 'net' of available WiFi covering the planet to ensure connectivity at all times. Or, something similar, perhaps managed by satellite from space.
The two areas found to be most important in your consultation both resemble our principle of data portability. MIDATA.coop promotes data portability, and our initiative is described here under "Switzerland", https://en.wikipedia.org/wiki/Data_portability
