ChapterPDF Available

openAdap.net: A Collaborative Sharing Environment

Authors:

Abstract and Figures

openAdap.net is an Open Source project aimed at breaking the barriers existing in the flow of data access and data processing. The infrastructure will make the dissemination of resources like knowledge, tools or data possible, their exposure to evaluation in ways that might be unanticipated and hence support the emergence of communities of users around a specific domain. The architecture is designed by analogy with a virtual distributed operating system in which the dynamic resources are presented as files in a structured virtual file system featuring ownership and access permissions. openAdap.net will be open to exploitation during Q3 2006 by networked organisations and alliances taking into account the vital issue of Internet security and privacy.
Content may be subject to copyright.
openAdap.net: A Collaborative
Sharing Environment
Alessandro E.P. VILLA1, Javier IGLESIAS 2, Solange GHERNAOUTI-HELIE 3
1INSERM U318 Université Joseph Fourier, Grenoble Cedex 8, 38043, France
Tel: +33 4 7676 5625, Fax: + 33 4 7676 5619, Email: avilla@neuroheuristic.org
INFORGE Université de Lausanne, HEC, Lausanne, 1015, Switzerland
Tel: 2+41 21 692 3587, 3+41 21 692 3421, Fax: + 41 21 692 3585,
Email: javier.iglesias@unil.ch, sgh@unil.ch
Abstract: openAdap.net is an Open Source project aimed at breaking the barriers
existing in the flow of data access and data processing. The infrastructure will make
the dissemination of resources like knowledge, tools or data possible, their exposure
to evaluation in ways that might be unanticipated and hence support the emergence
of communities of users around a specific domain. The architecture is designed by
analogy with a virtual distributed operating system in which the dynamic resources
are presented as files in a structured virtual file system featuring ownership and
access permissions. openAdap.net will be open to exploitation during Q3 2006 by
networked organisations and alliances taking into account the vital issue of Internet
security and privacy.
1. Introduction
The information circulating in the Cyberspace generates a historically unprecedented
richness for sharing knowledge and methods of data processing. The software published on
the Internet in one instance becomes available to the whole society worldwide. However,
most of the latest processing and analysis tools developed and provided by research
laboratories and businesses are based on specific software and computer platforms. The
current situation generates a lag until these contributions become known within the same
field of competence and severely restrict their availability, in particular for cross-
fertilization application in other fields. This delay may provoke the “re-invention” of
methods for data processing and, more generally, the “re-discovery” of the same
knowledge. At a societal scale this delay inhibits the development of added-value activities
originating from sharing the knowledge and generates repeated unnecessary expenses and
even, erroneous applications. The societal benefits in terms of business developments,
market diversification and creation of employment generated by the fast circulation and
ease of access to the results of the Human Genome Project illustrate the interest in
developing an open and dynamic adaptive network for resource sharing with emergent
properties within the Information Society.
2. Objectives
Our project is aimed at breaking the barriers existing in the flow of data access and data
processing increasing the overall cost of information processing and restricting its
availability to developing countries, by avoiding re-invention of existing software and save
time and expenses by the whole society and even prevent incorrect applications. The ability
to evaluate and to find the best available solution to a given problem may have significant
impact on areas such as economics, physics, environmental sciences, meteorology, and
753
Exploiting the Knowledge Economy: Issues, Applications, Case Studies
Paul Cunningham and Miriam Cunningham (Eds.)
IOS Press, 2006 Amsterdam ISBN 978–1–58603–682-9
© 2006 The authors. All rights reserved
health. By adopting the best available method of analysing a set of data, it is possible to test
alternative solutions and choose the best from its overall performance.
In the construction industry, which is playing an important role measured by
investments in all economical systems, even small enhancements will make resources
available to other purposes. Reduction of the energy consumption during the whole
lifecycle of a building will decrease pollution and save money. By setting up an
environment where people are used to working with state of the art technical solutions,
companies throughout the World will be forced to increase their competitiveness work in a
completeness way.
We envision the emergence of communities of users sharing resources like knowledge,
tools, data, etc. for their specific domain. The key to this vision is the development of a
sharing architecture that is independent of a specific type of information: openAdap.net.
Through programs like web portals and standalone applications interacting with the system,
the community will share the set of tools that each of these domains typically used through
members contributions.
Individual users of openAdap.net can be classified as either contributors of shared
resources, or end-users of such. People who develop and provide new methods of data
analyses are able to share their contribution and people interested in processing their own
data or access data stored elsewhere (e.g. in a community database) can extract the results
of their analysis. In addition to individual use, openAdap.net is open to exploitation by
networked organisations and alliances, by providing an infrastructure with which they can
share and integrate resources and develop new business solutions and opportunities.
The philosophy behind openAdap.net is that user’s privacy is as important as
contributor’s traceability. We believe that information sent by an end-user for manipulation
by a contributor‘s resource should be anonymized, despite the fact that all the transactions
are identified and that the activity is tracked like in any computer system (not to mention
the web). In contrast, meta-information concerning the contributed resource like program
authorship and version number should be made available to the end-user, as a mark of
diligence to the contributor, but also to point the responsibilities in the information
processing chain and enhance quality assessment and reproducibility. Each task submitted
to the system will be returned to the submitter with a record attesting where the information
travelled, which processes were applied and who was responsible for the computers and the
programs.
By sharing a program and the computer running it, for example, contributors will keep
complete control over their authorship as well as the source and binary codes for the
software. At the same time, they will be responsible for maintaining and checking the
quality of the results that the community will use and validate. From the viewpoint of
contributors, openAdap.net makes the dissemination of resources possible, and their
exposure to application and evaluation by a broader user community. The support for
broader evaluation of programs and data sets is particularly important in the research arena,
as is the ability of other researchers to reproduce computational results.
In a second step, openAdap.net will become an environment where new techniques and
methodologies will gain access to a wide range of users, possibly beyond the direct
community boundaries to the adjacent domains. The system allows the dissemination of
resources across domains in ways that might be unanticipated by the original contributor.
For example a solution developed by experts in the field of medical statistics could
permeate to the field of botany, and become a key tool in the analysis of tree distributions.
In this way, openAdap.net supports transdisciplinarity by breaking current boundaries to
resource sharing.
The
openAdap.net project is not directly aimed at the production of new methods of
analysis, but the platform helps the community to aggregate their already existing tools by
A
.E.P. Villa, J. Iglesias and S. Ghernaouti-Hélie / OpenAdap.ne
t
754
dynamically composing new information processing chains using the output of existing
programs as the input to others, thus opening the way to cross-fertilization and serendipity.
3. Methodology
openAdap.net stands in the area of complexity and aims at providing a distributed
environment where tools of all kinds (applications, data, knowledge, …) can be accessed
transparently via the Internet. At this time, three architectures are used in this field: Grid,
Web-services (WS), and Peer-to-Peer (P2P). The peculiar strongholds of these architectures
are briefly described here and synthesized in Table 1.
Grid: “Each user has a large dataset to manipulate with one application distributed on
a set of computers.” The problem addressed by the Grid is to distribute the processing of
large data sets. The required application is copied to distinct computers over a network with
each treating a portion of the data. The results are recomposed from all the partial results at
the end of the treatment. End-users have control on both data and applications, but little
information on remote execution hosts.
Web Services: “Many users exploit the same services permanently provided through a
centralized authority.” Web Services provide a secured and controlled access to
applications, usually to collaborators in the same company or institution. The goal is to
provide distributed access to the same references, databases or applications. The system is
articulated around a repository where the service interfaces are published in a self-contained
form. The architecture is rather static, as services seldom change and are expected to be
permanently available. End-users have no control over the applications and little
information on remote execution hosts.
P2P: “Many users exchanging pieces of data in an unsupervised way.” P2P (peer-to-
peer) systems address the data-sharing problem. Copies of the applications installed on end-
users’ computers keep open connections from one computer to peers’, forwarding queries
and results back and forth until a match is found somewhere on the network. The
architecture is open and competing implementations coexist on a self-organized network.
End-users have control over their data and information on the peer hosts. It is interesting to
note that end-users tolerate incredibly poor service quality and that this architecture raises
social as well as technical issues.
Table 1: Comparison of openAdap.net with the other approaches over Internet
Data
treatment
distribution
Hardware
resource
allocation
Hidden
execution
hosts
Application
sharing
Published
application
interface
Data
sharing
Highly
dynamic
system
Transparent
user /
resource
connection
Grid X X X
WS X X X X
P2P X X X
OAN X X X X X X X X
openAdap.net (OAN) falls somewhere between these three architectures exploiting
several of their interesting aspects, but with the intention to address a two-way problem: To
provide a community of users in the same domain with a means to interchange their
resources in an open, dynamic and secured way and to provide to a community of users the
access to the exploitation of information processing solutions contributed by users
belonging to other communities. End-users have control over their data, but do not need to
manage the resources, nor do they have complete information on remote execution hosts.
Collaboration within the openAdap.net network allows the dynamic integration of these
resources, possibly yielding new or previously unexpected composite resources. This can
A
.E.P. Villa, J. Iglesias and S. Ghernaouti-Hélie / OpenAdap.ne
t
755
be summarized as follows: “Many users interchanging resources (data, applications,
knowledge, …) dynamically provided by interconnected domain-oriented brokers.”
end-user computer
<<access>>
<<access>>
<<bind>>
<<bind>>
<<bind>>
<<call>> <<call>>
<<bind>>
OAN-aware
application
end-user PDA
OAN-aware
application
end-user computer
web browser
application server
web portal
contributor’s computer
OAN worker
program
Aprogram
B
OAN broker
RDBMS
broker server
Figure 1: Deployment diagram for the openAdap.net components. Boxes represent different computers,
rectangles represent processes, and arrows represent inter-process communications over Internet.
4. Technology Description
openAdap.net is a distributed system composed by three types of interconnected
components: brokers, workers and OAN-aware applications (see Figure 1). A broker is a
process permanently running on a server in charge of managing a community of users and
dispatching tasks and results on their behalf. Workers are processes shared by community
members in charge of giving secured distant access to contributed resources like programs
or data. OAN-aware applications are pieces of software (standalone applications, web
portals, command line tools, etc.) providing access for an end-user to the community shared
resources through identified connections to a broker.
The components are running on a network of computers, each of them defined by their
specific CPU architecture, operating system (OS), amount of memory and available
programs. Their resources are partially shared in a dynamic way. openAdap.net is designed
by analogy with a virtual distributed OS in which all the resources are presented in a
structured virtual file system. Using this high-level paradigm, resources are assigned to files
in the file system tree. Security is enforced through ownership and access permissions
defined on a per-file basis. openAdap.net goes beyond a traditional OS as the configuration
is highly volatile. The file system structure and contents are the result of the constant
runtime aggregation of several sources of information: the user database, the inter-process
message queues status, the worker status, etc. A dedicated URL name scheme (oan://)
will be proposed to identify each file in a transparent and interoperable way.
The Java 2 Platform was chosen for the implementation of the project, based on
portability and platform neutrality requirements. Brokers, workers and OAN-aware
applications are loosely coupled, distributed components that asynchronously communicate
through a message-oriented middleware (MOM) as defined by the Java Message Service
(JMS) API. The use of established industrial standards such as JMS allowed reusing
existing Open Source software as a starting point for the project. It can also be expected
that JMS implementations available for other platforms will allow applications written by
third parties to connect to openAdap.net brokers and interoperate seamlessly.
A
.E.P. Villa, J. Iglesias and S. Ghernaouti-Hélie / OpenAdap.ne
t
756
In the context of commercial exploitation, security is a vital issue and this is an intrinsic
part of the openAdap.net specification and design. Focussing on security within open
environments means to define the targets of threats. Without doing a risk analysis survey,
the main targets of security threats are: end-user, network access point, network and all the
infrastructures connected to the network as servers and information systems (Figure 2) [1].
Figure 2: Security within open environments.
For openAdap.net the security is addressed transversally by defining rules on how to
handle data at any point in the system, providing the required profiles for identification and
authentication of the entities, and offering strongly encrypted connections when
communicating through the Internet. Standards and models of the field were applied from
the start of the project. The use of unconditional secure encryption schemes (such as
quantum cryptography) is also possible [2].
As the users need to discern technological and informational elements to build the
confidence in online services [7], these points could furnish to Internet users informational
and technological elements necessary to having confidence on E-services and E-business.
Without confidence, the relationship between the E-services' provider and clients is not
possible [8]. The confidence is the base of the exchange over the Internet. openAdap.net
with its one-point web access gives more guaranty of security than other open systems.
Many stakes and challenges are related to users' privacy over the Internet. So, the basic
rights of privacy must be respected and guaranteed to all openAdap.net users wherever they
are located. By anonymizing the information sent by the users and emphasizing the
traceability of contributors, openAdap.net provides more than the minimum confidence
level mandatory for an effective digital exchange based on e-services.
5. Developments
Internally, brokers are responsible for decomposing and routing end-user tasks to
appropriate workers for execution. A key element in the next stage of development consists
in making brokers adaptive and dynamically interconnected into an openAdap.net network
(like a neuronal network). The requests for resources will be processed and dispatched
among the components of the system following a set of learning rules dynamically
modifying the routing according to, for example, the computing load generated by specific
tasks, availability of the resources, or the number of accesses. The rules themselves will
evolve and optimize in an unsupervised manner, thus allowing the emergence of
unexpected dynamics. In that sense, the required negotiation between brokers (and workers)
may be compared to agent interaction. The openAdap.net network will also be able to self-
adapt via learning processes derived from neural and artificial life learning. Such learning
might result in new broker-broker connections, reassessments of the value of such
connections, specialisation or generalisation or broker behaviour, etc.
The adaptive and behavioural models for the broker implementation represent major
innovations of the openAdap.net project. This is definitely a novelty and a plus to the
A
.E.P. Villa, J. Iglesias and S. Ghernaouti-Hélie / OpenAdap.ne
t
757
existing architectures for distributed environments like grids and web services that points
out the project expected income to the networked computing field. We believe that SMEs
and research institutions will be interested in developing novel interdisciplinary solutions
associated to the psychological and technological aspects of evolvable simulation tools, the
psychological environment of remote user support and the formal aspects of artificial
processing in resource-sharing.
We also expect to interface openAdap.net with established distributed systems like
grids and clusters. For that purpose, specific workers will be developed to provide
openAdap.net to Condor or Portable Batch System (PBS) interfaces.
Pushing existing paradigms like neuronal network inspired learning rules for the
adaptable information processing or the operating system paradigm for the overall
communication layout, and the lessons learned for 10 years on the self, dynamically, openly
organized content on the web are key aspects of the openAdap.net philosophy and
architecture for resource sharing. Besides, openAdap.net is an Open Source project
designed as an open architecture. Anyone is invited to contribute their own enhancements
to the system, via a set of libraries and tools provided by the consortium. Such an initiative
is aimed at increasing the impact of the project with all the contributions that competent
contributors will imagine and realize within their specific domains.
Expressions of interest have been committed by international organizations like CERN
(European Organization for Nuclear Research) and UIT/ITU (Union Internationale des
Télécommunications) and we are looking to broaden the Consortium towards the
submission of a FP7 proposal.
Figure 3: Screenshot of the openAdap.net web portal
using Mozilla Firefox on Apple MacOS X (as of July 2006 - alpha version).
A
.E.P. Villa, J. Iglesias and S. Ghernaouti-Hélie / OpenAdap.ne
t
758
6. Results
In the last few years, we have been developing, running and testing a prototype for
openAdap.net. The concept proof and feasibility have been checked. In the last few
months, we have been re-implementing the project from scratch (Figure 3) based on the
experience we have acquired with the prototype. openAdap.net being an Open Source
project released under the GPL and LGPL licences, contributions are welcomed from
developers and interested professionals. This is encouraged by the availability of publicly
accessible development tools like a version control system, mailing lists and a bug tracker.
There is enough experience on the impact of such coordinated and distributed development
scheme on the quality of the resulting software to embrace it.
During the prototyping phase, the need appeared to have a portable and user-friendly
tool that could provide a fast glimpse on the numerical output of unrelated programs. We
searched for a generic front-end for plotting a wide range of graphics obtained through the
web, and none could be found that was able to run on multiple platforms without requiring
complicated installation procedures, and capable of producing journal-quality graphics
output for free. XY-Viewer is a Java application that is being developed for that purpose as
a by-product of the main project, featuring a dedicated XML data format. This standalone
application can be launched directly from the openAdap.net portal to visualize files in the
appropriate format that can be produced with the help of libraries made available to
interested contributors in C and Java under the LGPL licence.
7. Business Benefits
A particular exploitation of the project is the development of web portals tailored to the
needs of end-user communities centred on specific domains, possibly by SMEs.
Communities will appear first in the domains of competence of the consortium partners.
These early adopters will constitute real-life case studies that will assess the validity of the
concepts and the usability of the implementation before a broader diffusion. The prototype
has already attracted a few individuals from different research groups to cooperate and
share resources in the domain of multivariate time series analysis [http://dan.unil.ch/].
In the future, we expect to give rise to synergies within and between both new and
existing communities. The identification of such communities will necessitate further
developments such as community specific ontologies and benchmark file repositories.
Users will be invited to contribute resources, articles, benchmark data, and to build the
domain ontology with the initial help of openAdap.net consortium.
Interactions with Publishers and Editors could have an incredible impact on the way
scientific dissemination, including visualization, is performed, mainly by facilitating peer
review in refereed publications and for results comparison and validation for peer scientific
readers. Such interactions could be envisioned once the openAdap.net system will be fully
available to the scientific community and should be encouraged. One major impact on the
scientific community could be obtained by attracting well known scientific editors to en-
courage the scientific authors to provide source data to the community repositories, and to
share their methods through an openAdap.net broker, hence promoting reproducibility of
results and direct peer review of published article methodologies and results by the readers.
Citizens of less favoured countries will have access to all shared openAdap.net
resources with a basic Internet connection, thus benefiting from the knowledge transfer and
available assets, and contributing back to the community with their own approaches and
resources. The outcome of the expected cross-fertilization is unpredictable. Side effects are
expected on the quality and harmonization of the resource documentation, as an effort is
dedicated to the elaboration of tools to enhance them.
A
.E.P. Villa, J. Iglesias and S. Ghernaouti-Hélie / OpenAdap.ne
t
759
8. Conclusions
This paper has presented the main features of openAdap.net, which is an intelligent
network infrastructure supporting the use of shared resources, such as data, knowledge,
tools, expertise, etc. aimed at providing the most advanced tools for data analysis and
manipulation to a broad audience over Internet. Individual users can be classified as either
contributors of shared resources, or end-users of such. openAdap.net will be open to
exploitation by networked organisations and alliances taking into account the vital issue of
Internet security and privacy.
The ability to tackle a scientific problem from a new perspective relies on both the past
experience and new skills adopted by an individual. The openAdap.net project is based on
the collaboration between information scientists, electronic engineers, computer scientists
and neuroscientists having diverse scientific interests and very specialized backgrounds.
We feel that such a transdisciplinary approach is a necessary way for the achievement of
real advances in producing impacts in the Information Society Technologies. A key point
for the success of openAdap.net dissemination is the build-up of Communities who share a
common data format tailored to their interest. In the prototype that we implemented the
multivariate time series format is well known and accepted by the major leaders in the field.
The
openAdap.net infrastructure makes the dissemination of resources possible, and
their exposure to application and evaluation across domains in ways that might be
unanticipated. Simulation processing tools issued from physical sciences could permeate to
study problems as different as the dynamics of societal interactions, linguistic analyses,
crops forecast, traffic congestions, and life sciences. openAdap.net is aimed at breaking
current boundaries to resource sharing and hence supports transdisciplinarity. End-users are
provided with the ability to browse and apply shared resources, and dynamically compose
and integrate existing resources to leverage new research insights.
openAdap.net brokers are responsible for dynamically decomposing and routing end-
user tasks to appropriate resource sharers for execution. The negotiation between brokers
(and workers) is inspired by the way how the brain processes information. When
completed, the openAdap.net project network will be able to self-adapt via learning
processes that could give rise to modifiable broker-broker connections, specialisation or
generalisation of broker behaviour, etc. The non-linear dynamics that will emerge from our
approach makes openAdap.net closer to the complexity of a living organism.
Acknowledgments
This work is co-funded by the European research project SECOQC (www.secoqc.net).
References
[1] S. Ghernaouti-Hélie. Guide de cybersecurité pour les pays en développement, ITU publication, 2006.
[2] S. Ghernaouti Hélie, M.A. Sfaxi, G. Ribordy, O. Gay. Using Quantum Key Distribution within IPSEC to
secure MAN communications. MAN 2005 conference, 2005.
[3] Konstantin Läufer. A hike through post-EJB J2EE web application architecture, Computing in Science
and Engineering, 7(5):80–8, 2005.
[4] G. Eisenhauser, F.E. Bustamante, K. Schwan. Publish-subscribe for high-performance computing, IEEE
Internet computing, 10(1):40–7, 2006.
[5] X. Zhang, S. Chen, R. Sandhu. Enhancing Data Authenticity and Integrity in P2P systems, IEEE Internet
computing, 9(6):42–9, 2005.
[6] A.E.P. Villa, I.V. Tetko, J. Iglesias. Computer Assisted Neurophysiological Analysis of Cell Assemblies,
Neurocomputing, 38-40:1025-30, 2001.
[7] S. Ghernaouti-Hélie. Challenges to develop and deploy a unified e-security framework. UNECE
Workshop on E-Security and Knowledge Economy, Geneva, Switzerland, 2003.
[8] E. Nardeli, S. Posadziejewski, M. Talamo. Certification and security in e-services, from e-government to
e-business. Kluwer Academic Publishers, 2003.
A
.E.P. Villa, J. Iglesias and S. Ghernaouti-Hélie / OpenAdap.ne
t
760
Conference Paper
Full-text available
OpenAdap.net is an Open Source project aimed at breaking the barriers existing in the ow of information access and information processing. The infrastructure makes it possible the dissemination of re- sources like knowledge, tools or data, their exposure to evaluation in ways that might be unanticipated and hence support the evolution of communities of users around a specic domain. The architecture is de- signed by analogy with a virtual distributed operating system in which the dynamic resources are presented as les in a structured virtual le system featuring ownership and access permissions.
Article
Full-text available
Quantum cryptography could be integrated in various existing concepts and protocols to secure Metropolitan Area Networks communications. One of the possible use of quantum cryptography is within IPSEC. The applications of quantum cryptography are linked to telecommunication services that require very high level of security in Metropolitan Area Networks. The aim of this paper is to analyse the use of quantum cryptography within IPSEC to secure MAN communications and to present the estimated performances of this solution. We analyse classical IPSEC advantage and limits to point out how quantum cryptography could enhance the security level of IPSEC. After having introduced basic concepts in quantum cryptography, we propose a solution that integrate quantum key distribution into IPSEC. A performance analysis is done to demonstrate the operational feasibility of this solution.
Article
Full-text available
High-performance computing could significantly benefit from publish-subscribe communication, but current systems don't deliver the kind of performance required by applications in that domain. In response, the authors developed Echo, a high-performance event-delivery middleware designed to scale to the data rates typically found in grid environments. This article provides an overview of Echo, the infrastructure on which it's built, and the techniques used to implement it.
In this article--the first in a planned series about Web application development--we take an exploratory hike through the architectural layers of a Web application built with state-of-the-art, widely used technologies. We focus on the upper layers that provide the application's user interface (we'll get into the lower tiers next time). By no means intended as a complete treatise on Web application development, this article is overview meant to spark your interest and provide a starting point for further exploration.
Article
Peer-to-peer systems let users share information in distributed environments because of their scalability and efficiency. However, existing P2P systems are vulnerable to numerous security attacks and lack a mechanism to ensure shared information's authenticity and integrity. A proposed general architecture enhances these aspects by leveraging trusted computing technology, which is built on a trusted platform module and provides a mechanism for building trust into the application layer. Preliminary experimental results show that the proposed scheme can ensure data authenticity and integrity in P2P systems with acceptable performance overhead.
Article
Since the mid-nineties, server-based Web applications have emerged as a convenient way to provide functionality to a user audience without any specific software or system requirements except the need for a reasonably up-to-date Web browser. The tricky integration, installation, and configuration tasks are under the control of an expert on the server side. These advantages apply to both the general public and small research teams. Typical Web applications have browser-based user interfaces for one or more user roles and might keep some information in persistent storage, such as a relational database. Since the early days of common gateway interface (CGI) scripting, technologies for developing Web applications have evolved by leaps and bounds to address growing expectations with respect to reliability, maintainability, extensibility, performance, scalability, and other goals. In this article - the first in a planned series about Web application development - we take an exploratory hike through the architectural layers of a Web application built with state-of-the-art, widely used technologies. The paper focus on the upper layers that provide the application's user interface. By no means intended as a complete treatise on Web application development, this article is an overview meant to spark your interest and provide a starting Doint for further exploration.
Challenges to develop and deploy a unified e-security framework
  • S Ghernaouti-Hélie
S. Ghernaouti-Hélie. Challenges to develop and deploy a unified e-security framework. UNECE Workshop on E-Security and Knowledge Economy, Geneva, Switzerland, 2003.
  • A E P Villa
  • I V Tetko
  • J Iglesias
A.E.P. Villa, I.V. Tetko, J. Iglesias. Computer Assisted Neurophysiological Analysis of Cell Assemblies, Neurocomputing, 38-40:1025-30, 2001.
Guide de cybersecurité pour les pays en développement, ITU publication
  • S Ghernaouti-Hélie
S. Ghernaouti-Hélie. Guide de cybersecurité pour les pays en développement, ITU publication, 2006.
Certification and security in e-services, from e-government to e-business
  • E Nardeli
  • S Posadziejewski
  • M Talamo
E. Nardeli, S. Posadziejewski, M. Talamo. Certification and security in e-services, from e-government to e-business. Kluwer Academic Publishers, 2003.