I answered the survey, the site was ok this evening, but when I
tried to submit, the survey had been shut down... glup :-(
If anyone knows anyone at DG Connect, do not hesitate to send him my
answers bellow (a kind of decentralized answer to a centralized
survey :-)
HTTPS://EC.EUROPA.EU/EUSURVEY/RUNNER/NEXTGEN-INTERNET
1.1 Your occupation and expertise
CEO of Cozy Cloud, a 18 people startup founded in 2012.
I have a master of engineering in computer science in France
(ENSIMAG)
1.2 What is your assessment of the current status of the
Internet and its impact from a European perspective ? What are
the major trends ?
The internet is increasingly centralized onto few nodes of
control. This centralization of data and therefore of services, is
at the core of the power of Californian oligopolies. They can buy
new incomers to consolidate their position, whatever the price is.
This imbalance leads to more and more tensions between these
companies and governments (taxes, mass surveillance), competitors
(abuse of dominating position) and users (less choice because of a
less competitive market, information manipulation (one says B.
Obama has been elected thanks to big data...). Many other
stakeholders should be in this list (the taxi men, hotel trade,
news papers...).
1.3 View towards 2025 and beyond How do you think the
internet will look like in 2025 and beyond?
Key words : Decentralized, User centric, Personal Cloud, PIMS,
Cryptography, Blockchain like (decentralization of trusted third
parties), Big data in peer 2 peer
Vision : (in very few words, I have been notified of this
consultation the very last day of this survey...) is that
decentralized ecosystems will emerge. Exactly as in 70's the rise
of personal computers was impossible from the institutions point
of view. IBM sold the MSDOS license to Bill Gate and in 1977 the
CEO of DEC said "no one will ever want a computer at home"...
Today we are in the very same situation, the big guys seems more
powerful than ever, but the decentralization will naturally
emerge.
At Cozy Cloud we are developing a personal cloud, a personal
server that anyone can have, self-hosted or hosted by a hosting
company such as a bank, where he can aggregate all his data in
order to have more frictionless and relevant services. 8 years ago
nobody had a smartphone. In 8 years everybody will have his
personal server.
The convenience of having a lot of your data at Google is the
frictionless experience and the relevance of personalized
services. In your personal cloud you can go much further on those
two axes since on your personal cloud you can aggregate safely
much more data than Google will ever have : health, finance,
center of interests... The services can thus be more relevant,
more proactive, and more seamless. You will be able to choose the
AI you want to plug on your Personal Cloud in order to benefit
from a virtual assistant who is truly at your service.
In a word : as surprisingly as it may seems from the current
situation, I thing that the future of the internet will be
decentralized and user centric.
1.4 What will be the essential functional building blocks of
the Internet then?
Decentralization could be built around : 1/ PIMS (personal
information management systems) 2/ third parties who can hosts the
PIMS (banks, hosting companies, hardware vendors to self host…) 3/
vendors/administrations/associations providing data to individuals
from their services (data from our connected devices, consumption
statements, invoices, health data, …) 4/ P2P social networks for
decentralized social requests, such as social comparisons or
epidemiology without disseminating data… 5/ trust hardware
provider (cf bellow), 6/ Addressing systems (ideally a
decentralized DNS, others wise a fragmentation of the web could
happen)
1.5 Could you indicate where we should focus our activity
research in the next 5-10 years to achieve?
The evolution of an always more centralized internet is not
sustainable, especially for Europe since the centers of the
internet are out of its boundaries. Institutions remain focused on
building the "next Google", but when you look at the underground
digital scene, what is striking is the vitality of decentralized
technologies. The bitcoin being the most visible today. So there
is a lot of relevance for Europe to massively invest in the
decentralization of the internet. Here are few hints, they are
mainly for the next 5 to 10 years, not beyond.
1/ personal cloud, PIMS (personal information management systems),
self-data... whatever you call it : development of system that
empowers the individual to take back of their data, in order to be
really empowered by their data working for their interest, and to
benefit from an open and profuse market of services.
2/ Big Data and P2P : peer to peer is the core of a
decentralization of the internet. Being able to make some request
over a distributed individual databases (personal clouds) solves
the issues of big data (privacy, quality of data, centralization
of power over a few actors being able to collect data) in order to
go beyond what can be achieved with the classical centralized
approach.
We have started a R&D project with Orange and INRIA, but the
field is huge and many different approached and challenges must be
taken into account.
3/ Trust hardware: With INRIA we consider that in the digital
space is unprivate by design. That's why we tried to contain the
chain of trust as small as possible: the idea is to depend of the
smallest hardware and of the fewer lines of code : so we put all
the sensitive algo and data in a SIM card. But this card must be
trusted... there is a lot to invest to bring such a hardware to a
high level of integrity.
4/ personal data portability: things are going well in France and
in EU : regulations has begun to take this point as a fundamental
right. But 1/ the texts are not perfect and 2/ many initiatives
must be conducted so that an economy taking the portability
constraint as an opportunity can develop : for instance
standardization, research over interoperability, project of
collaboration involving some big companies, new laws with better
definitions (I can help :-)
5/ decentralization of the DNS and of certificates
I cannot develop here an argumentation for each of those points,
but just to show that we are convinced of what is stated above :
at Cozy we are financing two Phd's with INRIA on those subjects.
The means for this fields are not only monetary : incentives for
big, challenged, European companies to change their approaches and
to innovate thanks to decentralization for instance would be
great.
On 09/04/2016 19:38, Pierre Ozoux wrote: