A FING reflexion, ISOC and Silicon Sentier.
Work for the French government to present and what will be the internet.
Nicely documented , easy to read, easy to understand ...
Read here . Pdf version .
Or below version copy / paste ... ( loss illutrations and good layout ! too bad ! )
"It is essential that the network
'universal' is the internet
an architecture that maintains
promotes accessibility and
innovation "
Future Internet :
to a " specification"
Response to consultation with the secretariat
of State for Forward Planning and
development of the digital economy
July 2009
a contribution
Common :
and several personalities
the French internet
(see attached list)
editors:
Daniel Kaplan
Jean- Michel Cornu
Internet of the future
2
Future Internet :
to a " specification"
Introduction ................................................. ................................. 3
1 - What problems is it to answer .................................. ? 5
2 - unsaid that cloud the debate .................................... 8
3 - Efficiency and universality at the bottom , that application does is to
the internet? .................................................. ................................ 11
4 - objects , subjects , two new goals for
Future Internet ............................................. ........................... 15
Conclusion: The three priorities ............................................. ....... 20
Appendix : List of participants in the working group .................... 21
About Fing , Silicon Sentier and ISOC France .................... 23
Internet of the future
3
introduction
In response to the public consultation on the future internet organized by
the Secretary of State for Forward Planning and Development
numérique1 the economy, Fing , Silicon Sentier and ISOC France and a
number of prominent personalities of the French Internet , have
chosen to produce all public funding.
This group2 consists mainly of non- technicians, who
However, the Internet daily and professional use to publish ,
innovate and produce .
Therefore, our contribution proposes instead the elements of a
"specifications " of tomorrow's Internet , envisaged by the "big
users, " a series of specific technical recommendations.
The questions posed by the Internet and its uses are certainly not
all their answers in architecture and network technologies himself .
Many answers will fall applications and services ,
users and their organizations , regulatory or ecological
uses. As part of this consultation, we decided to
we focus on what we had to ask to the Internet - and
symmetry, on what would be better not to ask.
It is also better to explore this second term of the alternative - that
we should not ask the network - we decided
evacuate (for now at least) infrastructure applications also
important that the web. This will enabled us to start from
question rarely asked: what do we really
the internet? What are we willing to sacrifice to get it ?
We hope that this short text will provide valuable insight and
complementary to other contributions.
It remains open to criticism, comments and enhancements to the
Site http://internetdufutur.wordpress.com/
Daniel Kaplan, Jean- Michel Cornu and Hubert Guillaud, Fing
Marie- Vorgan The Barzic and Mary Noéline Viguié , Silicon Sentier
Gérard Dantec and Laurent Ferrali , ISOC France
1 Introduced by a document [ www.telecom.gouv.fr/internetdufutur/ ] which must acknowledge the
quality , this consultation must allow " the definition of an action plan for positioning the
France in the development of the Internet of the Future , and thus promote the economic benefits
industry and for our country. "
2 See attached list .
Internet of the future
4
Internet of the future
5
1 - What problems is it meet ?
Correspondence with the consultation questions " Future Internet "
1a) What are the main new services and applications that will use
redefine the operation and use of the Internet ?
3b) A number of considerations (eg in terms of safety,
quality of service and governance ) should be taken into account more
early as possible in the design work of the Future Internet : according to you
why and how?
If so many people in so many countries, reflecting on " Future Internet " , this
must be that it is having problems , or that it should
meet in the future. To consider the directions to take , it is
important to identify these problems and to measure the more or
less critical. It must also be part of the problem really related to
technique, those from the organization of the internet or
its actors.
And yet it works!
For the most amazing , basically, is that for most uses and
users , the Internet works well enough while its architecture
fundamental has changed very little in nearly 30 years.
This certainly comes at the cost of great complexity , distributed to all
levels multitude of actors at all levels of the network -
to the administrators of corporate networks and users
( address management , security , various updates , etc.).
multiplication adaptations intended to allow the network to accept
new uses and new users "patches" associated with
security, QoS , mobility, etc. . But the fact remains that
exceptional adaptability of the Internet, and the extraordinary
diversity of users and uses for which it is
today a satisfactory basis , should lead to approach any
significant transformation with great discernment .
Problem dimension
One of the main challenges facing the Internet today is
that follow the growth of the number of connected devices. This
reflects not only the growing number of users of the
number of "terminal" user (many electronic devices
equipping itself with a network connection ) , but also the proliferation of objects
and spaces " communicating " .
An insufficient number of IP addresses provided by the current IPv4
is obvious. Various measures have so far helped manage the shortage,
Internet of the future
6
the price of a certain fragmentation of the network and no doubt,
unable to think of some uses ( eg it is not
direct access to each connected object by its own address
independent applications , which makes it difficult to use as
"server" ) .
Inflation routing tables could also end up creating
clogging phenomena. Despite promises of generations
successive routers , technological change does not seem to
address itself to the problem of growth.
Deficiencies for uses "critical"
The Internet poses problems for organizations or types
exchange for which the network, or such of its
characteristics , critical nature - in the sense that if a function
data is not met, the exchange becomes impossible or seriously
problematic :
• The internet is not readily offers secure end to end
exchanges . Technically, it is assumed that all
intermediaries provide much the same service in a manner consistent
with each other . However, the Internet is a network of networks
independent of one another , a major part of the security
falls that users and applications. note
today, in most cases , network operators
as users start very inadequately implement
technologies and approaches available and basic security.
• Based on the principle of obligation of means ( " best effort " ) and non-
good end , the internet can hardly be used to exchange
in need of a guaranteed level of " quality of service " in terms of
reliability, speed , latency , etc. . Consider, for example, uses
for which the real time is essential , including the high-definition video
definition, but also applications to synchronize
precisely different devices ( industrial applications,
security , scientific , etc.). . Again , solutions exist, but
is difficult to ensure that they are implemented, and the same
way , from beginning to end and across all networks.
Difficulty in accommodating new demands
As it stands, despite some adjustments, internet mismanaging mobility. it
also lend itself to the multicast signal ( "multicast " ) .
Difficulties in economic and social
Finally, the Internet creates or makes obvious difficulties whose character
does not seem technical at first, but one senses well -
Internet of the future
7
for good or bad reasons - some solutions
could go through changes in the network itself :
• The internet is difficult to control, but it is not so
uncontrollable as they say . The Iranian government is not so bad reached
restrict communications between its citizens, and it has seriously
rarefied their interactions with the world . We can not all
control all the time, but it is clear that in
many countries, we manage well enough to censor the Internet , or
least trace of those who pass through the mesh of
net.
• The economic model of the Internet, as a network , appears to be
fragile. For example , the profitability of service providers
is obtained today, by financing large users by
small ( as discussed below) , or by extending
other services, which may pose competition problems or
fragile areas (such as content ) whose profitability
is already proving problematic.
• The identity management on the Internet is hardly satisfactory . while
the network has become , for many people , an action space and
daily transaction , digital identity is usually related
physical and contractual network access conditions or a
service. For most people , changing ISP,
mail service or social networking means losing
address, contacts, short, much of its identity.
Digital identity is in fact not really protected (there
many ways to find out who is behind a nickname, a
IP address) , or actually certified (think knowing who was
case , but to be sure ) .
• Our life , our economic life , aspects
basic functioning of our societies become very
dependent vis-à -vis the Internet. But it does not really accept
responsibility that should accompany this dependency :
preservation and protection information in the time or
otherwise , secure deletion of information , resilience to
problem , etc. .
So there is a lot of good reasons to be interested in the internet
future , if any hand flat on the intellectual level at least. but
a number of unspoken also singularly obscure the
debate.
Internet of the future
8
2 - unsaid that cloud the debate
Correspondence with the consultation questions " Future Internet "
1a) What are the main new services and applications that will use
redefine the operation and use of the Internet ? What could be the impact
these economic changes?
1b ) What factors have the greatest impact in redefining the Internet and its
architecture?
2a ) Do you think that the six themes covering a clear and comprehensive
all aspects of the concept of the Internet of the Future ? If not, what (s)
other ( s) categorization (s) or other research topics would you suggest?
All the problems mentioned above are real and significant . But the internet
lived with them for years. Some of these problems have found
solutions more or less tampered with. In other cases , solutions
available techniques , albeit perhaps imperfect , slow to get
implementation - this is especially true of security, both at the heart of
networks in businesses and workplaces .
So, why the theme of the Future Internet , or - for some - of
the necessary overhaul takes it so important today ? and
Internet what is it ?
An economic issue poorly documented
The issue of " net neutrality " has arisen in the news through
business section : it was, for some providers
American , supplement the income they derived from their end customers
( selling subscriptions , bandwidth and interconnection ) in
charging the sites most used or the most demanding web, exchange
a guarantee of priority routing on their links.
This amounted to the bottom to remind you that the economic stability of most
actors in the network based on a complex and opaque web of
cross-subsidization :
• At the heart of the network, small networks and poor countries
subsidize other paying interconnection fees well
more high3 ;
• At the local level , entanglement between networks and point
public and private peering , reserved and shared, making it difficult
any precise calculation ;
3 " peering or how the South is financing the North" Vox Internet , November 2006 -
www.correspondants.org / news / the - de- peering agreements -or- how - the -south- north- financing
Internet of the future
9
• In terms of access , clients connected to a medium or low flow, or
those who consume little subsidize heavy users of
downtowns
• The complex also transfers occur between
business operators : access , content, transactions ...
We now understand that certain assumptions behind the evolution of
Internet , are expressed wishes - possibly conflicting - of
change the economic balance and games players.
For example, it is difficult to guarantee quality of service end-
end if each communication takes a different path from
"at" a large number of independent actors to each other , to
all sizes follow different rules - this is the situation
today . But to do otherwise would require integration
much stronger networks to access , and much less
operators , each linked to the other by rules ( and economic
techniques) interconnection strict . It is clear that such
integration change the competitive landscape of the Internet.
Similarly, greater integration between
access and content, or a
network architecture that would more
difference between fixed and mobile access ,
result in any other
market organizations.
It is legitimate interests are expressed.
But they rarely appear as
as in the debate on the future Internet .
Interest is usually mask
behind functional arguments
techniques.
A better understanding of the reality of
the Internet economy , and the roles , uses, and issues of interest
all its actors, seems essential to an informed debate and
open to the future internet. France should engage in work
research in this field and invite the European Union to do the
same .
A desire to control that is not subject to debate
American reflections on the future Internet became more and
more visible from the attacks of 11 September 2001. This means that
security concerns plays a big role. Although it concerns
States against terrorism and, more generally , new forms
war . But the fight against crime and crime is
Also invited to the debate. And for their part , companies require
also safer for the wanton attacks or interested
they are often victims. And more recently , they are
Internet of the future
10
economic sectors experiencing new forms of
competition more or less fair , or even with their customers (in the case
cultural industries , for example) that require the network to
Read more about its users , filter , block , draw ...
Some of these applications are found in fact at the heart of debates
public , as is the case around the " Hadopi law." But many other
express themselves in a more discreet manner. One can understand why .
But at the same time , with regard to the network architecture as
essential that the Internet , the degree of importance to be given to the insertion of
new security features in the network must be open to
discussions. Because we do not secure without consideration, whether
in terms of opening the network performance, games players , etc. .
The virtues of the current internet should be the basis for building
Future Internet . Individuals and organizations use
collaborative potential of the internet for an active and effective way . taking
Addressing these issues is an asset for emerging countries see
the Internet as an opportunity in terms of development and (most
often) freedom of expression. It is , for all, a factor
innovation and growth. Any proposed change must be judged
based on its capacity to sustain this momentum or , conversely,
weaken .
An issue of global governance
The working group has not produced specific contribution of the
Internet governance , which, however, remains a subject of future
major . This governance remains today largely dependent
U.S. government on the one hand , and on the other , large
companies ( especially American ) who have both intelligence and
resources to finance the massive presence of their engineers in
standardization groups .
The future development of the Internet face of governance
more multilateral and multi -stakeholder . The movement engaged in
the auspices of the United Nations since the World Summit Company
information (2003-2005) , which has emerged a form of expression of the
"civil society" , is a first step , obviously insufficient.
Europe must become an active participant of the debates to come. the
European companies must be aware of the need
to be present when and where to draw the
standards. Finally, the work must go beyond the framework of the forum
officials and discussion lists for technicians to allow
other groups to assess the economic, social and
environmental changes considered.
Internet of the future
11
3 - Efficiency and universality :
basically, that application does one internet ?
Correspondence with the consultation questions " Future Internet "
1b ) What factors have the greatest impact in redefining the Internet and its
architecture? Possible what new forms it can take? What
are locks up?
3b) A number of considerations (eg in terms of safety,
quality of service and governance ) should be taken into account as early as
possible in the design work of the Future Internet : do you think that
and how?
3c) New industries and new economic models could
emerge as a result of innovations from the users themselves (...)
How to ensure that they will be taken into account in the design of the Internet
Future for a better match between networks and applications supported by
these? Or rather does it take a minimalist design type "utilities" that
can adapt to future changes difficult to model ?
Reflections on the Future Internet seem to assume that
network must become more efficient , on the one hand, grow and secondly ,
better meet the needs "critical" - security , quality of service,
etc. .
This idea seems indisputable , however worth the look
depth and in particular that it raises the question of what
we lose when we gain in efficiency.
Because it also requires the internet to be a "universal" network , and even
doubly
• It must be accessible to as many ( if not all ) , so
available everywhere, easy to deploy and connect , inexpensive.
The internet has made connectivity a new aspiration, we see
express themselves excluded from the networks and more, excluded
of democracy and development in the world. The Internet of
future must obviously continue to bring this hope.
• It must also accommodate the uses, services , applications and
even the most eclectic and most unexpected diversions and
enable them if necessary to deploy on a large scale . in
functioning as a neutral and simple infrastructure , supported
open standards, making no distinction between a priori
points it connects , the internet has become in a few years
one of the most extraordinary platform for innovation in history.
Millions of applications, websites, human organizations,
process ... rely on him. Virtually none were even
imaginable at the beginning of the network. Very few have been invented by
major players in the network .
Internet of the future
12
To achieve this level of universality, the priority criterion is no longer
efficiency, but adaptability, the ability to accommodate the unexpected - the lawyer
Jonathan Zittrain talks about a " generative " internet 4.
How to balance flexibility and efficiency?
Improve efficiency means that we seek to optimize the management of
scarce resources of the network ( IP addresses, the computational power of
routers , bandwidth ... and sometimes financial operators ) and
maximize security, which requires providing the better it
happening on the network ( the paths for example).
Getting in a position to respond to the unexpected requires almost the opposite:
constantly offer an abundance of choices , or capacity ,
between which every innovator , every organization or every
individual chooses the appropriate time, but not before ...
So on one side , we try to optimize
( manage scarcity by predictability ) and
the other to adapt ( manage unpredictable by
abundance) . These two approaches
seem incompatible, but in reality
everything is a matter of balance.
The sciences of complexity , indeed show that any network that
or technological , biological or human with optimum durability
when it finds the right balance between
efficiency and resilience ( who gets
certain notions of adaptability ) 5.
This optimum is usually a
closer than the resilience
efficiency. It is also at this point
of balancing optimization and
adaptability , between order and disorder ,
emerge new opportunities ;
short, the potential for innovation is
maximum .
Technical arguments usually focus on efficiency, not
may therefore be sufficient to determine the direction of work on
" Future Internet " . Of course we must respond in one way or another
needs "critical" of certain applications fail if some
functions are not filled to a certain level of quality ( guarantee
good end ) . But the internet should remain as a network -oriented
Universal - accessible to all and for all purposes , without prejudice to their
relevance - and this is ill-suited devices locked .
4 The Future of the Internet and How to Stop It, Yale University Press, 2008
5 Ulanowicz , RE A Third Window -Natural Foundations for Life , Oxford University Press, 2008
Internet of the future
13
There is criticism and critical
What are the applications that really require a "critical" network ,
built primarily on criteria of efficiency and safety ?
Were often presented for example the need for a perfectly Speed
continuous ( quality of service ) as a critical element in enabling
moving voice over a network. But the widespread use of phone
mobile , IP telephony , video on the Internet or even games
line of " low resolution " shows that users arbitrate more
often we think for the selection, price, mobility,
their own ability to participate in the
expense of the highest quality.
By against military networks , research
or interbank transactions can be seen
as critical : security, time of arrival
information , speed of interaction , have
a vital character . This is also the case of any
networks that allow remote control
pacemakers or other devices
Medical : in this case the security and continuity
Service must be provided to the maximum,
leaves to reduce innovation on such networks .
To coexist two logical
Without neglecting the importance of own networks and efficiency criteria
the "critical" uses, thus can not reorganize the internet
around these criteria. It is " essentiel6 " , "universal " network that is
Internet keeps an architecture that promotes accessibility and
Innovation ( symmetry of trade, direct interconnection of different
points in the network ... ) . From this point of view, it is necessary in the short term
facilitate the deployment of IPv6 (which is immediately
available and keeps the balance between " best effort " and optimization)
while promoting the search for new solutions that respect the same
équilibre7 of guy.
On the other hand , it is economically reasonable for networks
critical , not interconnected with the same " universal " internet share
where possible , the same physical infrastructure.
Different technologies now make it possible to work on
the same physical network of "virtual" networks , and even different
watertight between them. This has the added advantage for the future ,
facilitate experimentation (preferably participatory , with all
6 The term "essential" is sometimes used in networks to distinguish from the
term "critical" .
7 For example, the new ideas proposed by the Pouzin Society, or other initiatives.
Internet of the future
14
stakeholders ) of innovative solutions for the internet that
not better balance performance, reliability, scalability , and economy
innovation.
Economic hardship
This vision of a "universal" network arbiter deliberately in favor of
potential for innovation at the expense of certain expectations in terms
efficiency and even security makes it difficult to set up points
of passage -for example, toll or control.
Note however that in other areas , these functions are
not managed by the network ( case of transport, including
modernization was accompanied by the disappearance of subsidies), or
they are essentially the ends of the network (case of
energy). The internet does not any exception .
If we can not have both maximum efficiency and
maximum adaptability must then find a way to allow
two forms of networks grow while interconnecting and
sharing as many resources.
Internet of the future
15
4 - objects , subjects : two new
goals for the future internet
Correspondence with the consultation questions " Future Internet "
3b) A number of considerations (eg in terms of safety,
quality of service and governance ) should be taken into account as early as
possible in the design work of the Future Internet : do you think that
and how?
5a) Do you agree with the analysis above about the importance of
disruptive research on the issue of the Future Internet ?
5b) Specify , in search of disruptive Future Internet , themes
you seem a priority.
A real " Internet of Things "
Today, what is often described as the " Internet of Objects"
deployed most often out of the internet ( in internal networks to
companies or sectors ) and on principles very
different from the internet: each team a chip that object or
place is generally communicates with his "master" , in devices
star around a centralized system , and for specific purposes .
Objects and spaces begin well networked , but in a way
selective and usually hierarchical , inside silos. We can not
generally not accessible to the sensors and actuators for them to do
something other than what has been planned their installers , is not accessed more
data output : the camera images to go PC
monitoring, traffic measurements to the traffic , and no
one else will ever do anything. In the short term , it will probably
identify " smart objects " in a manner almost universal ,
but there is a connection "star" to a server is not a
" Internet of Things ".
This has two consequences : first , most chips
installed in objects or spaces must be profitable on a
only use for one actor - who installed them . As formerly
specialized computers, mono -tasking and single users. And other
hand, the innovative imagination is little to apply, since access to
infrastructure facto that all these chips are still under
control , and the combinatorics of these chips , objects , spaces,
users remains constrained.
A market also condemns forced to stay small.
16 but they practices. 17 network. 18 to users.and19etc. .20**exist**21they22Fing23company
