ABOUT |
Computer Science Internal SeminarsThe Internal Seminar Series is a relaxed forum for members of the Department to present their current research and discuss ideas of interest. Invited speakers are also welcome, in particular for presentations that might be to specialised for a general computer science audience as on the Friday's seminar. Semester 2Seminar programme
Seminar detailsICT for Saudi Health System Development
Ahmad Khudair (Loughborough University) The health infrastructure of Saudi Arabia and
its diagnostic, therapeutic and health prevention services
have made important and increasing progress during the
past decades. The Kingdom's medical colleges and hospitals
provide specialist diagnostic and therapeutic services and
medical education and training programmes, while they also
conduct health research in collaboration with other
research centres. However, Saudi health care is facing
some challenges. A factor that affects the sustainability
of Saudi healthcare is that hospitals are established
individually and independently with no consideration for
co-operation and co-ordination. The existing cooperative
research is aimed to study and develop the role of
Information and Communica-tion Technologies (ICTs) in
Saudi health systems. A literature search and document
analysis related to the development of the Saudi
healthcare system is presented, in conjunction,
non-participant observation and interviews are used to
provided deep understanding. In practice, Saudi health is
the least fortunate sector in utilising ICTs and most
hospitals have a poor information infrastructure. In
particular, the rate of use and adoption of Electronic
Health System (EHR) is low and small health practice
organisations appear to lag behind larger ones in adopting
EHR systems. A model for health system sustainability in
Saudi Arabia is proposed that is a framework for
developing e-health services in Saudi Arabia. It has been
concluded that development and technological advancement
will not reach their full potential if there is continued
lack of investment in management. A holistic view of the
current health system may be adopted to utilise ICT as a
driver for improved health system sustainability.
The Protection of Software Development - some interesting issues arising from EU v Microsoft Ann Walsh
The Microsoft case provides a fascinating set of new circumstances
within the established tensions surrounding intellectual property and
competition law.
The main aim of this talk is to give a brief introduction to some of the
main issues arising in the Microsoft case. It will also attempt to
highlight some of the particular difficulties which software brings to
this area of law, and try to stimulate discussion on how the development
of software should be protected.
Hong Qing Yu
Ossama El-Hassam
Gavin Cox
Reversible computation has a growing number of potential application
areas such as debugging, testing and the modeling of biochemical
systems. To look in the possible use of reversibility in such areas
some form of computer aided simulation would be of great
advantage. With this in mind I have created a prototype simulator
SimCCSK for CCSK a reversible version of CCS with communication keys
by Phillips and Ulidowski.
Migrating Legacy Systems to Service-Oriented Architectures
Carlos Matos
This presentation focuses on the issues that involve the
transformation of legacy systems towards Service-Oriented
Architectures. First a motivation for the problem will be given,
followed by an explanation of the strategy that is being used in the
context of Leg2Net project. This consists of the separation of the
problem in two levels of complexity: technical and functional. An
approach to address the latter is then presented, as well as some
ideias for its implementation.
Semi-automatic Software Reengineering: Migrating Legacy Systems to Layered Architectures
Rui Correia
In this presentation we put forward a methodology for migrating legacy
systems towards new architectural styles. Based on a metamodel for both
source and target style, the approach consists in (1) categorizing the
source code according to the different elements of the target
architecture they shall be mapped to, (2) obtaining a metamodel-based
representation of the code, (3) transforming it into the target
architectural style, and (4) generating the target code.
The categorization is carried out through an iteration of manual code
annotation and automated deduction.
The result of the categorization determines the level of detail required
for the metamodel-based representation which enables the use of graph
transformation rules on metamodel instances to describe the
transformation. Our approach follows that by Mens et al in which
refactoring is formalized as graph transformation, except that:
* we aim at fully automated transformation, controlled by the code
categorization.
* our transformations are not always semantics-preserving
We will also discuss issues of the implementation of the approach based
on existing program and model transformation tools and report on a small
case study, the transformation of a two-tier application in Java into a
three-tier one.
Georgios Koutsoukos
The increasing adoption of agile software development methods is
amplifying the message that people are one of the most critical
success factors of any software project. As A. Cockburn, one of the
pioneers of agile methods, states:
"People's characteristics are a first-order success driver, not a
second-order one. [...] Most of my experiences can be accounted for from
just a few characteristics of people. Applying these on recent
projects, I have had much greater success at predicting results and
making successful recommendations. I believe the time has come to,
formally and officially, put a research emphasis on what are the
characteristics of people that affect software development, and what
are their implications on methodology design"
In this presentation we describe a means to make certain people
characteristics a first-class concern in software development
projects. In particular, we present a systematic way of capturing,
communicating, reasoning and making such human characteristics
explicit through an abstraction and representation primitive that we
call mentality pattern. In a second stage, we use this primitive to
define what we call the Mentality Innovation Sub-Process - an
organized way to supplement and enhance software development methods
and processes in order to "manage" explicitly such human-related
factors and improve the effectiveness of individual work and the way
teams blend together.
Carlo Montangero, Dipartemento di Informatica, University of Pisa
\lambda_{req} is a core language for service orchestration, developed
inside SENSORIA by Bartoletti, Degano and Ferrari to explore how the
call-by-contract mechanism may be exploited to define services that
abide by explicitely stated safety and liveness policies.
In the spirit of SENSORIA, we are defininig and implementing a UML
front-end, to facilitate and standardize the exploitation of the
language. We will discuss both the problems related to the proper
utilization/extension of UML to express the relevant concepts, and the
VIATRA2-based graph-transformational approach taken to implement the
front-end.
Towards a Web-Service based Implementation for Architectural Management Laws
Ahmed Al-Ghamdi
Web-Service applications require a clear approach to management in
order to be efficiently and optimally deployed. For that purpose,
Service Level Agreements (SLAs) have been proposed as management
contracts between customers and service providers, using different
implementation technologies (what are called IT-SLAs). However, the
modelling of SLAs at the business level remains largely unexplored
despite the many benefits that they may provide in addition to
IT-SLAs. In this work, we propose to leverage Web-Service management
to this business level by putting forward adequate conceptual
primitives based on software architectural techniques.The main
advantage of our approach is the ability to distribute rules across
web services, which enables powerful ways of executing and separating
business rules.
O.N. Delpratt
This is joint work with Naila Rahman and Rajeev Raman
We consider the PrefixSum problem: Given a sequence of positive
integers k=( n_1,...,n_k), the prefixsums for the sequence are
(d_1,...,d_k), where d_1=n_1, d_2=n_1+n_2,...,d_k=n_1+...+n_k, such
that m=d_1+d_2+..+d_k. We wish to support the operation sum(k,j),
which returns d_j=n_1+..+n_j. Our interests is in minimising the space
required for storing k, where minimum space is defined according to
some compressibility criteria, while supporting sum as rapidly as
possible.
We are motivated by several applications such as Inverted lists in the
information retrieval (IR) community, String collections: web
databases and XML documents where we often achieve good text
compression, but a large overhead for string offsets. The main
application is XML DOM.
We demonstrate a close connection between a succinct compressilibity
measure and (IR) data-aware measure that is the best in practice. We
give theoretical solutions that use space close to other data-aware
compressibility measures (often within o(n) bits), and support sum in
doubly-logarithmic (or better) time, and experimental evaluations of
practical variants thereof.
A bit-vector is a data structure that supports 'rank/select' on a
bit-string, and is fundamental to succinct and compressed data
structures. We describe a new bit-vector that is robust and
efficient.
Organization-Oriented Measurement and Evaluation Framework for Software and Web Engineering Projects
Hernan Molina (GIDIS Web, National University of La Pampa, Argentina)
Quality assurance is one of the most important activities in every
software and web organization. Through the use of measurement and
evaluation processes it is possible to keep an updated view of the
engineering activities providing useful mechanisms to improve the
quality of processes and products as well. In this sense, a framework
that defines sound specifications of the information used in
measurement and evaluation activities in the organization, can provide
more consistent and comparable solutions among the organization's
projects.
In this presentation we describe a measurement and evaluation
framework called INCAMI, which is based on an ontology describing the
main concepts involved in this activities, and also in the WebQEM (Web
Quality Evaluation Method) methodology. This framework is
goal-oriented, since all activities are performed to satisfy a stated
information need; and it is organization-focused, considering
consistency issues among all organization's projects. A current
research work is evolving this framework to make it context-aware, in
order to be able to use the information relative to the relevant
context to help in the design activities of measurement and
evaluation.
A Categorical Model of Computation for Graph Transformation: True Concurrency and Logic, Part II
Reiko Heckel (Leicester)
See Jan 25
A Categorical Model of Computation for Graph Transformation: True Concurrency and Logic
Reiko Heckel (Leicester)
Categories with algebraic structure are well-known as
models of concurrent computations for, e.g., formal
grammars, Petri nets, and term rewriting. The approach, of
obtaining concurrent computations by means of an algebraic
construction from a set of rules, can be generalised to
graph transformation systems. The resulting computational
model, a free double category with finite horizontal
colimits, captures the established notion of concurrency
of DPO graph transformation.
Such a construction exposes the intrinsic structure of
concurrency in graph transformations and provides the
basis for defining a temporal logic for concurrent
computations. The logic represents a specialisation of van
Benthem's arrow logic to categories, extended by spatial
connectives like a concurrent composition of computations.
The talk will present the basics of the categorical model
and discuss the definition of the logic, its semantics,
deduction rules, and potential applications.
Semester 1Seminar programme
Seminar detailsBioinformatics: Simulation (Computational methods)
Krishna Bandaru (Leicester)
I am going to cover
below topics.
Importance of the role of computers and computer science in
investigations and applications of biological data.
Convince you that bioinformatics will have a positive effect on our
understanding of life and improving human health.
Computation of the Electrostatic Component of Solvation Energy using
Poisson-Boltzmann Equation: Comparison of Krylov Subspace
Methods.
Software process engineering in the filed: "Doing the thing right" and "Doing the right thing" James Byles (BSc,
MRPharmS), Michael Isherwood (BSc) from Accenture
(accenture.com)
A type-based approach to service interaction (and sessions) Emilio Tuosto Service Oriented Computing (SOC) promotes the
metaphor of computation as service composition where
(basic) services are (somehow) composed to form more
complex services. Services interact by invoking each
other.
In this talk I'll describe a very ongoing work where a
type-based approach to service coordination is
presented. The proposed type system has been designed on
the Signal Calculus (SC), a process calculus based on an
event notification paradigm where services are modeled as
components which expose reactions (i.e., interfaces)
triggered by signals.
An emergent issue in SOC is the formalisation of sessions
within computations. Sessions are indeed widely used in
practice but, at the best of our knowledge, they have not
been very much studied. Only recently there has been some
research on this topic. We will show how the proposed
type system can suitably be used to model general notion
of sessions.
This work has been jointly done with Roberto Guanciale and
Daniele Strollo during their visit to our department.
BIP Language Framework for Modeling Systems Integration from Heterogenous Real-Time Components
Simon Biludze (VERIMAG lab, Grenoble, France)
A central idea in systems engineering is that
complex systems are built by assembling components. System
designers deal with a large variety of components, each
having different characteristics, from a large variety of
viewpoints, each highlighting different dimensions of a
system. A central problem is the meaningful composition of
heterogeneous components to ensure their correct
inter-operation.
We present a methodology for modeling heterogeneous
real-time components. Components are obtained as the
superposition of three layers: Behaviour, specified as a
set of transitions; Interactions between transitions of
the behavior; and Priorities, used to choose amongst
possible interactions. A parameterised binary composition
operator is used to compose components layer by layer.
We present also the BIP language for the description and
composition of layered components as well as associated
tools for executing and analysing components on a dedicated
platform. The language provides a powerful mechanism for
structuring interactions involving rendezvous and
broadcast. We show that synchronous and timed systems are
particular classes of components. Finally, we provide
examples and compare the BIP framework to existing ones for
heterogeneous component-based modeling.
Abstract Model Theory meets Computational Logic
Balder ten Cate (Amsterdam)
JSCL: a middleware for service coordination
Daniele Strollo (IMT Lucca, Italy)
The main goal of the Service Oriented Architectures (SOAs)
is to define software components which are (i) developed
in a language and architecture-independent way and (ii)
distributed in the web. Most of the current methodologies
are focused on composition of such services and several
programming languages have been recently proposed for
describing aggregated services. However, it is not
infrequent that such languages have drawbacks. In fact,
constructs are often informally specified which usually
leads to ambiguities or redundancy. In this talk the
prototype implementation of Java Signal Core Layer (JSCL)
will be described. Basically, JSCL is a coordination
language of distributed services based on an event
notification paradigm and has been inspired (and its
implementation driven) by the Signal Calculus, a formally
defined process calculus.
Pawel Sobocinski
(Cambridge, UK)
Reversible CCS was introduced in order to
model concurrent computations where certain actions are
allowed to be reversed. In fact, the core of the
construction can be analysed at an abstract level, yielding
a theorem of pure category theory which underlies the
previous results. This opens the way to several new
examples; in particular we shall demonstrate an immediate
concrete application to Petri nets.
Signal Calculus: A formal approach to service coordination Roberto Guanciale
(IMT Lucca, Italy)
Signal Calculus (SC) is a process calculus specifically
designed to describe coordination policies of services
distributed over a network. The formal specification
driven and inspired the design and the implementation of a
middleware for service coordination (JSCL). We present the
formal specification remarking the analogies with the
programming middleware.
Policy Support for Business-oriented Web Service Management
Stephen Gorton
(The University of Leicester, UK)
Policies have been adopted for many reasons within Web
Services and Service-oriented Architecture in
general. However, while they are a favoured method of
management, this only occurs at the service level and in
the software domain. Policies already exist in a narrow
variety more focussed on service properties such as
authorisation. As a significant number of web services
become available, more emphasis needs to be placed on
management of services in the business domain. In this
presentation, we propose a policy framework that can be
used to express business requirements for web services, at
a business level that is more abstract than the current
high-level composition and orchestration technologies.
This is joint work with Stephan Reiff-Marganiec in the
Sensoria project.
|
|
Author: Alexander Kurz (kurz mcs le ac uk), T: 0116 252 5356. |