AUGMENT OPEN AND DISTRIBUTED SOCIAL NETWORKING
WITH CONTEXT AWARENESS
Claudio Venezia and Carlo Licciardi
Telecom Italia, Strategy and Innovation, Via Reiss Romoli 274, Turin, Italy
Keywords: Distributed and mobile social networking, Context awareness, WebID, Social multimedia sharing.
Abstract: Nowadays Social Networking phenomenon is gaining a significant momentum. Social Network’s users are
disclosing a lot of information about their relationships, profiles, preferences and interests but also an
increasing amount of multimedia contents. The Web is being populated by Social Networking Web sites
which are introducing fragmentation and mining future evolutions. In this position paper we propose a
concrete alternative to the current state of the art, to overcome vertical silos' aggregation and enable open,
distributed and Context Aware Social Networking focusing on users’ privacy and data ownership.
1 INTRODUCTION
Online Social Networking is determining a new
paradigm of experiencing the Web. The interaction
paradigm has definitely been evolving from Human
to Machine to Human to Human.
Online Social Networks have introduced and
fostered this dynamic but may not be future proof.
The main concerns are about data ownership,
Identity management and information overcrowd. In
prospective, mash up Web sites could be useful to
serve as directory to find friends but should not
mandatory mediate all interactions and store
whatever contents users would like to share.
The consumer electronic market is expanding
and highly sophisticated technology is rapidly
becoming affordable. New market segments are
created (tablet, e readers etc.) and costumers are
extremely responsive.
These devices are increasingly powerful and
capable of managing, creating and consuming
different kind of digital contents also in mobility.
Meanwhile the Web is targeting TV sets, Network
Access Storage and IP appliances are rapidly
spreading in domestic Upnp/Dlna environments,
storage’s cost is decreasing and bandwidth gets
cheaper.
In this context Telecom Italia is researching new
paradigms of peer to peer Social Networking to
seamlessly share multimedia contents, devices and
digital experiences without central aggregators’
mediation.
The Web platform (W3C Social Web IG final
report, 2010) already provides the means to create
such a kind of infrastructure but new standards and
practices should be pursued to maximize
interoperability and properly address specific
security requirements.
Telecom Italia is investing on this research area
within EU FP7 project SOCIETIES.
The SOCIETIES project (http://www.ict-
societies.eu/), aims to investigate and address the
gap between pervasive and social computing by
designing, implementing and evaluating an open
scalable service architecture and platform for
Community Smart Spaces (CSSs).
2 OPEN AND DISTRIBUTED
SHARING WEB BASED
PLATFORM
With respect to a sustainable and distributed Social
Networking platform we foresee and detail in the
following paragraphs two main challenges:
 Managing a Web Identity
 Managing content Read/Write operations on the
Social Web
2.1 Web Identity
Defining decentralized Identity management
infrastructures would be the basis of an open Social
298
Venezia C. and Licciardi C..
AUGMENT OPEN AND DISTRIBUTED SOCIAL NETWORKING WITH CONTEXT AWARENESS.
DOI: 10.5220/0003399302980301
In Proceedings of the 1st International Conference on Pervasive and Embedded Computing and Communication Systems (PECCS-2011), pages
298-301
ISBN: 978-989-8425-48-5
Copyright
c
 2011 SCITEPRESS (Science and Technology Publications, Lda.)

Networking ecosystem. We need a framework to
univocally identify a person or other entities on the
Web such as companies, organizations, web sites,
contents or devices.
Whichever content these entities may publish on
the Web their unique ID should serve to identify
them as owners. Most importantly entities may
reference each others, declare social relationships,
let others access and consume contents also if their
profile is not hosted on the same Web server.
Whatever you publish on the Web should be
reconciled to your profile and accessible, without the
need of creating and replicating profiles into
different silos.
The game is quite simple, you identify yourself
with a URL then you provide a document describing
yourself, your social relationships and available
resources at that URL. A certificate based
architecture and procedure can then easily support
authentication and enable access control list in a
browser friendly, encrypted way.
Internet users should be able to store and publish
on the Web, on their mobile or home server and
grant access to a list of other entities still being in
full control of their data and tracing requests.
Telecom Italia is contributing for this to become
a standard as this would be the only way to collect
requirements from heterogeneous domains prevent
fragmentation and encouraging adoption.
WebID is already an unofficial submission to
W3C, it basically aims at using X.509 certificates to
associate a User Agent (Browser) to a Person
identified via a URL. Telecom Italia is kicking off
the W3C WebID incubator group (W3C Web
Identity IG Charter, 2010) together with Apache
Software foundation, Nokia and DERI. This
initiative aims to pave the road for standardizing the
WebID protocol by bringing together people
involved in authorisation and authentication
activities, building on the existing WebID initiative.
2.2 Distributed Social Contents
and Services
In response to increasingly powerful handheld
devices, people are constantly populating the Web of
new contents. We define as content whatever can be
read/written on the Internet ranging from Web
pages, Live Audio/Video Streams profile
information and so on.
As said the current practice pretends to force
people to apply to different vertical silos to reach
different audiences and this is extremely unpractical
for managing one’s information on the Web and for-
ces to waive data ownership.
The content sharing framework we foresee
would not mandate to publish information into the
same Web site but would let everyone choose where
to store them just providing the mechanisms to
reconcile to the owner identity.
In Telecom Italia we prototyped a Social context
aware platform which can be instructed by users to
seamlessly collect and correlate information, e.g.
generating a geo referenced picture album, or an
automatic daily blog to share with friends or
acquaintances. This kind of information can be very
privacy sensitive and would be unfeasible to store it
constantly to an online social network, the average
user would probably like to collect it into an home
server for personal recordings and possibly share
with very close friends.
The content storing/publication phase has to be
structured in a way to easy the look up phase. Some
of the contents (e.g. pictures) may span over
different devices (e.g. Home Server, Mobile, third
party Web site) but we expect the retrieval phase to
be agnostic of the data store. With this specific
respect we foresee the need of (standardizing) a
minimal semantics (e.g. taxonomy) and a user
friendly mechanisms (e.g. tags) to properly
categorize contents in the act of publishing.
Contents (e.g. photo, video etc) will include
further information in form of attributes (e.g.
picture, date, caption, context etc).
W3C is already providing standard mechanisms
such as RDFa (W3C RDFa Primer) to nest this kind
of information within standard HTML page for use
of semantic crawlers.
Merging Web Identity and Semantic
categorization of contents, would consent to create
effective views of one’s multimedia contents stored
into different devices or web sites (accessible by
means of references to the actual hosting devices)
and to share them without the need of passing those
to third parties. Even if contents may reside on
different devices, if they are organized by some
semantic criterion, performing an integrated view
would be an easy task.
Nevertheless we believe that electing some of
them as the master storage would make sense and
that a home server (e.g. Network Accessible
Storage) would meet the requirements in terms of
storage capabilities, processing power, continuous
Internet connectivity and low power consumption.
Depending on the type of contents published
(e.g. video) and number of concurrent requests
(number of interacting friends) this may be the
preferred option.
AUGMENT OPEN AND DISTRIBUTED SOCIAL NETWORKING WITH CONTEXT AWARENESS
299

Pretending a home server to be the master node,
users should be able to define the conditions under
which a synchronization with any other device
(camera, tablet, laptop) should take place, e.g. just in
WiFi network mode.
3 SOCIAL CONTEXT
BROKERING ARCHITECTURE
Multimedia sharing is definitely a critical asset of
communication in online Social Networking. Often
the descriptions turn out to be the real
communication payloads instead of the content itself
(Adrien Joly et al, 2009).
Multimedia content descriptions add both
semantics and context information. They are crucial
for optimizing the communicational benefits of a
Social Networking application especially in a
situation in which contents tend to increasingly
overcrowd these sharing spaces.
Nevertheless adding this information maybe time
consuming, which usually lead to low quality
information which result useless or even misleading.
We believe that automating part of this task could
significantly improve the communicative factor of a
multimedia sharing space.
There is some useful information which may be
collected from a wide range of devices, sensors and
information provider and automatically associated to
the content which is being published.
In a multiple devices’ appertaining to the same
entity scenario, we foresee an architecture based on
a context bokering/sharing node (e.g. home server)
plus several providers and collector (running on
various devices or clouds) for each entity.
For instance a context brokering node could be
deployed in the said home server (one of each user)
and in turn collect information from various mobile
devices, assist the content publication phase and
additionally provide user’s friends and
acquaintances with dynamic status information
about a user (kind of next generation presence).
We’ve pursued a distributed context-aware
system (Claudio Venezia and Luca Lamorte, 2010)
capable of aggregating and processing a variety of
context information provided by physical sensors
embedded on mobile devices (Zhdanova et al, 2006).
There is already a significant body of research on
context modelling and brokering platforms (Claudio
Bettini, 2010). Most of them address pervasive and
ubiquitous computing requirements and do not
explicitly target the Social domain. Furthermore
being Social Networking tightly coupled to Web
technologies we resolved to target a lightweight
environment, loosely coupled and based on Web
protocols. Usually in context-aware systems, domain
knowledge is very much tied to applications but we
resolved to pursue a service agnostic definition of
context. Current state of the art of online Social
Networking mainly focuses on presence and
location. Our definition of context spans over a wide
collection of information available from user’s
surrounding environment, his/her terminal, network
connectivity, current activity. It is indeed is a large
amount of information which tends to grow
proportionally to the observing time. Users’ social
graph role is twofold; it is a context source but plays
also a critical role in the content sharing phase.
The “Social” Context management lifecycle
consists of four phases: acquisition, representation,
brokering, reasoning and publishing/sharing.
The context aware platform is therefore
architecture capable of:
 collecting context information from different
kinds of sensors or devices (e.g. domestic Upnp
devices, mobile, tablet) with respect to a particular
user
 storing and aggregating contextual information
 Perform a Blog like representation of the
collected information so to be ready to share
 Show detailed current status/activity information
to the Social Graph
This information can inspire several use cases: very
close friends may be informed about the movie
currently playing on my TV, the music I am
listening to or the food I am actually preparing and
decide to do the same or interact.
Moreover users can be triggered in case of
similar context status thus creating dynamic sub-
communities of people performing same activity or
sharing the same mood.
As said different devices (e.g. mobile, tablet, hifi,
IP appliances) belonging to the same entity (e.g.
person) may concur to provide contextual
information. We foresee that the home server,
elected as master node, it could be the ideal host for
a context aggregation and brokering component. We
envision also a light client running on devices which
we call local context broker (LCB) to collect data
from sensors and provide it to the broker.
The various devices’ LCB will trigger context
updates to the broker which will store the overall
information.
A common language for context information
representation has been defined to enhance
PECCS 2011 - International Conference on Pervasive and Embedded Computing and Communication Systems
300

interoperability: “ContextML”, an XML-based
language, which states a meta-model for the
representation that all Providers need to comply
with.
For simplifying context management user’s
information have been subdivided by into scopes,
namely sets related to the same information
category. For example, the scope named “position”
groups latitude, longitude and range with respect a
certain entity’s (e.g. user) location.
In the act of publishing a content from any
device a user may be automatically be provided with
fine grained contextual information (e.g. his current
state, location, activity) which he may want to add to
the description.
4 CONCLUSIONS
It is timely to supersede Social Networking silos
approach and unblock novel Social Area Network
paradigms. The key Internet players’ strategy
consists in acting as brokers. They provide user
friendly tools to publish and consume information
while they keep on storing and mining it, as required
by the asynchronous nature of the Web platform.
The current Social Networking platforms have been
designed and engineered to conduct people to keep
on feeding this mechanism. The ultimate goal is
increasing providers’ profits rather then realizing the
full potential of this technology/paradigm. Services
are not for free, Internet users pay for those services
by means of the information they release to the
brokers. Being constrained to release your
communications’ payload to third parties in order to
reach your audience will soon become anachronistic.
Decentralizing Identity management and profile
information will avoid a centralized control and
ownership of data. This will discompose the current
Internet control’s chains and disrupt the current
monopolistic approach.
Each Internet user should gain a WebID, publish
a Web document describing him/herself and
exposing interactive interfaces to share digital life
with friends.
Privacy sensitive Context awareness will enrich
interactions and help surviving the noise produced
by an exponential content production. Standard
cryptography mechanisms and certificate based
authentication should be put in place to protect
interactions.
ACKNOWLEDGEMENTS
Part of the research leading to these results has
received funding from the European Community’s
Seventh Framework Programme (FP7/2007-2013)
under grant agreement no. 257493 of the Societies
(Self Orchestrating Community ambient
IntelligEnce Spaces) Collaborative Project.
REFERENCES
W3C Social Web IG final report. http://www.w3.org/2005
/Incubator/socialweb/XGR-socialweb-20101206/.
Societies consortium. http://www.ict-societies.eu/
W3C Web Identity IG Charter. http://lab.linkeddata.deri.ie
/2010/WebID-XG/webid-charter-draft.html
W3C RDFa Primer, Bridging the Human and Data Webs
http://www.w3.org/TR/xhtml-rdfa-primer/
Adrien Joly, Pierre Maret, Johann Daigremont context-
awareness: The missing block of social networking.
International Journal of Computer Science and
Applications,Vol. 6, No. 2, pp 50 -65, 2009.
Claudio Venezia, Luca Lamorte. Pervasive ICT Social
Aware Services enablers. Proceedings of IEEE ICIN
2010.
Zhdanova, A. V., Zoric, J., Marengo, M., van Kranenburg,
H., Snoeck, N., Sutterer, M., et al. (2006). Context
Acquisition, Representation and Employment in
Mobile Service Platforms. Proc. 15th IST Mobile &
Wireless Communications Summit.
Claudio Bettini, Oliver Brdiczka, Karen Henricksen,
Jadwiga Indulska, Daniela Nicklas, Anand
Ranganathan, Daniele Riboni: A survey of context
modelling and reasoning techniques. Pervasive and
Mobile Computing 6(2): 161-180 (2010).
AUGMENT OPEN AND DISTRIBUTED SOCIAL NETWORKING WITH CONTEXT AWARENESS
301