GISST
Global Information Systems
and Software Technology
Project
Submission
(i)
August 26, 1999: Project Summary
(ii) September
13, 1999: Full
Project Proposal
This
document is also available on the following Web site:
http://www.cs.ubc.ca/~gisstnce/home.html
The
GISST Steering Committee, led by Dr. Son T. Vuong and Dr. David Lorge Parnas, invites
project submissions for the GISST Network application to the Network of
Centres of Excellence (NCE) Program.
The
GISST network has been invited to submit a full proposal by October 4,
1999, after being selected on June 28, 1999 on the basis of our Letter
of Intent (12 out of 45 were accepted). As part of this process, the GISST
Steering Committee invites research groups to submit (theme) projects which
could be included in the final proposal. All projects will be reviewed
by the ad hoc Steering Committee – an invitation to submit does
not mean the project will be accepted for the final proposal. If the
GISST proposal is accepted by the Network of Centres of Excellence program,
project funding is expected to begin in April, 2000. The following pages
present an overview of the NCE program, a summary of the research areas
targeted for the GISST network as well as the instructions for project
submission.
The
Network of Centres of Excellence (NCE) Program
· Stimulate
internationally competitive, leading-edge fundamental and applied research
in areas critical to Canadian economic and social development;
· Develop
and retain world-class scientists and engineers in key research areas and
technologies essential to Canada's productivity and economic growth;
· Create
nation-wide multidisciplinary and multisectorial research partnerships
that integrate the research and development priorities of all participants;
· Accelerate
the exchange of research results within the network and the use of this
knowledge within Canada by organizations that can harness it for Canadian
economic and social development.
NCE
Selection and Evaluation Criteria: The
networks are assessed against the following six equally weighted criteria:
· Networking
and partnerships
· Knowledge
exchange and technology exploitation
· Management
of the network
There
are currently 15 NCEs doing research in health and biotechnology, information
technology, natural resources, infrastructure, and computer-based learning.
In April 1999 a call for applications to establish new Networks of Centres
of Excellence was announced. The GISST network could be one of the three
or four new networks to be funded in 2000! More information is available
on the NCE program Web site: http://www.nce.gc.ca/
The
GISST Network – Global
Information Systems and Software Technology
Information
Technology affects every aspect of people’s lives: how we deal with information,
communicate with each other, learn, work, conduct commerce, design and
build devices, products and services, sustain a livable environment, and
organize our societal institutions. Information
technology has become the key technology for the next millennium and we
are all becoming increasingly dependent on it. While
it has brought us capabilities and convenience that we could not have imagined
a generation ago, it has also brought us frustration because of increasing
problems with unreliability and unpredictable performance. GISST
intends to be a sharply focused network of researchers working on two key
problem areas: · Reliable,
Predicatable Networks · Software
Quality Assurance. We
consider these to be the two key issues for the future of information technology. The
last decade has seen the emergence of new dimensions in global information
infrastructure. Internet communications supplemented
by
wireless networks have expanded the access to a ubiquitous and pervasive
global information system. Although
the technological growth has been stupendous, there is still a long road
ahead for a comprehensive, robust and secure global system of information
and communication networks, which is functionally seamless, efficient and
effective. At
the heart of all attempts to harness global information systems we find
software, and software remains an unmastered technology. Errors
and glitches introduced by computer and software are so commonplace
that they are considered unavoidable. Software
errors have become every company's excuse for failure to provide a service
correctly. Improved methods
of getting high quality software are essential step for the growth of network
based information systems. The
impact of this new NCE, GISST,
that combines the two important areas, Global Information Systems and Software
Technology, is definitely strategic to Canada’s socio-economy. A
major focus of GISST will be on the next generation Internet technology
and applications. The Internet
has grown well beyond the original vision of the initiators and designers
and new versions are already in use. As
its size, functionality and complexity increases, enormous technical, social,
political and legal challenges have to be met. In
a recent report to the Office of the US President, several technical challenges
were identified, namely: understanding the behavior and performance of
the global-scale network by collecting, measuring and analyzing performance
data, and developing new models for network behavior and performance; physical
structure of the network including optical, wireless, satellite, cable
technologies and bandwidth management; anticipation and plan for network
scaling; middleware to enable large scale systems; large scale applications
and scalable services such applications require; and a set of test beds
and research infrastructure. By
developing a research network that focuses on these two complementary areas,
Trustworthy Reliable Networks and Software Quality Assurance,
Canada will be better able to ensure its competitiveness in the information
technology sector in the global economy context. The
GISST approach
is to
work closely with industry and apply research methods to industrial problems.
Industrial vision will help to set the direction of research
in GISST and interaction with industry will provide guidance. Thus,
its research outcomes are expected to have direct influences on strategic
decisions of the enterprise partners - that will include several leading
network and software firms. GISST
also wants to look beyond the Internet at various types of more specialised
networks that can provide information services more efficiently and more
effectively than the general-purpose Internet. GISST Research Themes and Thrusts
We
propose the creation of a Network of Centres of Excellence under the theme
Global Information Systems and Software
Technology (GISST), focusing on two key research areas: Reliable,
Predicatable Networks and Software Quality Assurance. In this
section, we examined each area in turn. The
proposed themes are applicable to one or more of the following thrusts: 1.1.Next-Generation
Internet Technology Emphasis
will be placed on scalability: scaling
to provide robust, high-speed access, with assured QoS when required; scaling
to provide multi-faceted access to create new ways for people to connect;
scaling to provide ubiquitous access; scaling
of services to reliably handle many users and requests; and scaling of
security infrastructure. The
following are descriptions of some suitable themes for GISST. They
are not meant to be exhaustive. Proposals
on other important related themes are also welcome.
Emphasis will be placed on innovation of the approaches, methods, models
and paradigms. (a) Next-Generation
Internet Infrastructure An
important challenge for the future growth of the Internet is to design
the next generation Internet backbone with enhanced routers that can efficiently
forward the exponentially increasing volume of traffic, and at the same
time provide service differentiation for various types of traffic. This
project will address a range of design and implementation issues related
to Differentiated Services, Integrated Services and
Multi-Protocol Label Switching (MPLS), including QoS-based
routing, packet forwarding schemes, dynamic resource and traffic management,
multicast routing and protocol for QoS backbone (QBONE), network and service
measurement and management. (b) Photonic
Networking This
project will develop specifications for providing QoS (Quality of Service)
and other desirable traffic management and control features for both Integrated
Services and Differentiated
Services IP networks based on optical transport facilities. The focus will
be on the design and performance evaluation of WDM-based systems
with huge bandwidth (tens of Terabits/s). (c) Dynamic
and Active networking: This
innovative concept and paradigm has the following attributes: self-configuring
network and network components, service discovery, dynamic service creation
through negotiations and consensus, dynamic
enhancement of device capabilities via self-learning, dynamic enhancement
of networking capabilities via active networking. 1.2.Software
Environment for Next-Generation Internet-Based Applications By
software environment for the next-generation Internet (NGI), we refer to
a wide spectrum of the middle ware, that
spans the higher layers on top of the enabling infrastructure to provide
to the applications and users ease of access to the complex, advanced network
services. In a technical sense,
this middle ware can be viewed as Network
Operating System (NOS) that supports NGI applications. The
following themes focus on the middle ware for NGI applications. Other
related themes that focus NGI middleware and/or applications are also welcome. (a) Scalable
Multimedia Computing: The
main innovation in this area will be in developing multimedia processing,
coding, transcoding, and communication systems that adapt automatically
to variations of capabilities in the client and server systems. The proposed
research will address
a range of projects, including design of
infrastructural proxy servers, multimedia
transcoding, Picoserver technology, client servers, embedded Internet applications, and
Internet appliances. (b) Pervasive
Computing: This
project/theme aims at developing a software environment that builds upon
and expands the notion of Web-based services by providing composability
(the ability to automatically aggregate many services together into a single
entity), customizability (the ability for users to inject code into
the system to customize a service's behavior), and accessibility
(the ability to access the service from a wide range of devices, including
PCs, workstations, cellphones, and Personal Digital Assistants), security
adaptability (the ability to adjust various levels of security for
a service, as critical in E-Commerce applications), and
user-mobility (the ability to move with ease the service available
at home to anywhere the mobile user might be, away from home). The focus
will be on support of next-generation multimedia (new media) and
NGI multimedia applications. (c) High-Performance
Global-Scale Cooperative Distributed Computing: This
project will address the technical challenges of scaling distributed systems
to run on the Internet with many thousands of clients who are widely distributed
around the world. Our goal is to develop system support for the many distributed
applications and services, taking into consideration shared data among
clients, the high degree of interaction among clients, and the high expectations
for reliability, availability and predictable network performance. This
problem can be combined (and further complicated) with that of Cluster
Computing, involving multiple servers. (d) E-Commerce: The
focus is on dependable infrastructure and environment for
e-commerce applications. Subjects
of interest in this theme may include: network security, testing of security
systems, monitoring and testing of e-commerce global distributed servers,
and data mining for e-commerce applications. An interesting research problem
is data mining to connect
between customers and suppliers. With
e-commerce the customer is faced with a bewilderingly large number of possible
suppliers and the supplier does not know where the potential customers
are. (e) Human-Computer
Interactions (HCI): There
are many facets to human-computer interfaces and interactions. Ease
of use would be important for many applications, especially next-generation
Internet applications. The
acceptance and popularity of Web browsers demonstrate the importance of
user models, human factors and other areas. To
achieve an information infrastructure in the fullest sense, we must address
intuitive models of use and user interface technologies to enable a class
of information appliances that will become a part of everyday life. (f) Socio-Economic
Impacts: As
more and more advances in high technology are introduced into our society,
it has become increasingly important for our
social and economic well-being that we understand the transformations and
potential dislocations affected by technology adoption and diffusion. 2. Software
Technology Software,
which is at the heart of all attempts to harness global information systems,
remains an unmastered technology. The
focus of research on software technology will be on Software Quality Assurance. Other
important related themes can also be proposed. 2.1
Software Quality Assurance Errors
in software products are so common that all software products carry a disclaimer
where other products would carry a warranty. While
we are beginning to understand software design methods, better quality
assurance methods are badly needed. The
GISST approach to Software Quality Assurance is to try to bring
together researchers who have worked with five complementary approaches
to the problem. Each of these
approaches has been pursued by separate research communities but they have
complementary strengths and weaknesses. We
believe that it is time to integrate these approaches to develop a set
of mutually supportive methods and tools. (a)
Testing Testing
is part of every software developer's "bag of tricks" but there is a fundamental
problem - you can never test enough. As
Dijkstra has commented, "Testing can show the presence of bugs, but not
their absence." GISST is interested
in proposals on tools and
methods to help developers to generate test cases, evaluate test sets,
reduce the number of test cases, estimate reliability using test results,
and combine information from other sources with information gained from
testing. (b)
Model Checking and Reachability Analysis In
the hardware design area, model checking and reachability analysis have
been found to be practical tools. In
software, we run into the "state explosion problem", because the number
of states is too high. GISST
is interested in tools that make model checking and reachability analysis
more practical for software. This can include, analysis of specifications,
combining model checking with symbolic mathematics systems, combining model
checking and testing, and using model checking as part of an inspection
procedure. (c)
Relational Methods Relational
methods allow unusually compact descriptions of software semantics and
quality standards. However
relational models have been pursued by an active and energetic research
group that is outside of the mainstream in Computer Science. As
a result there are only a few situations in which relational methods have
seen practical use. GISST is
interested in proposals to develop tools and methods that use the elegance
of relational notation to make formal analysis of programs more practical. In
particular, we are interested in proposals to apply relational methods
in such tasks as reducing the number of tests required, and inspecting
programs. (d)
Systematic Inspection of Programs Starting
with the pioneering work of Fagin, experience has shown that formal, disciplined
inspection processes can be far more effective than traditional reviews
or walkthroughs. Work at the
Naval Research Laboratory and the inspection of the Darlington Ontario
Nuclear Safety Software have shown that mathematical descriptions can be
practical ways to boost the effectiveness of inspections. GISST
is interested in methods of improving the effectiveness of software inspections
perhaps by use of relational models or symbolic mathematics systems. Combinations
of testing with inspection are also of interest. (e)
Use of Symbolic Mathematics Systems for Software Quality Assurance Theorem
Provers have already been proven to be of great value in software inspections. There
are other symbolic mathematics systems that used extensively in engineering
but are not used in software development. These
are symbolic algebra systems such as Maple or Mathematica. It
has been shown that the correctness of a program can be reduced to a question
of whether or not two mathematical expressions describe the same function
or relation. That question can, in turn, be decided with the assistance
of symbolic algebra systems. GISST
is interested in proposals to provide practical methods or tools for verifying
software - particularly methods that combine symbolic manipulation with
model checking or testing. We
believe that practical approaches to software quality assurance cannot
neglect any of these areas. Testing,
for example, is not complete without the use of precise specifications
and requirements documents to allow the testing results to be evaluated. Complete
testing is usually impossible but model checking and symbolic mathematics
systems can be used to reduce the amount of testing needed and to help
in the selection of test cases. Issues of component-based software design
and acceptability test for software components also pose interesting research
problems. The
GISST approach
to research on Software Quality Assurance will be to work together on a
series of case studies. We
intend to obtain programs from our industrial partners (first
small simple ones with larger ones coming later) and approach each one
with a combination of methods. 2.2 Protocol
and Communications Software Engineering The
project aims at developing practical telecommunications software design
strategies that will be both user-friendly and formal, leading to fewer
design errors, better documentation, and greater conformance to requirements. The
methodology is to be applied to practical industrial projects. The emphasis
will be on advanced NGI protocols
and applications, e.g. design, verification and testing of multimedia,
mobile, multi-party, multicasting and security protocols General
conditions for GISST project submission Researchers
and organisations that receive GISST NCE funding must meet the general
eligibility conditions of NSERC. Both short and long term projects may
be submitted. The maximum funding period is 4 years, and there will be
the possibility of renewal. Note that all projects are subject to yearly
reviews by the Network management in order to ensure continued funding.
All major projects must target at least one of the themes outlined above. Major
projects should be based on the concept of a mini-network, where
inter-provincial collaboration is
strongly encouraged. In
addition, priority will be given to
1. Global
Information Systems
A
typical major project
would have both fundamental and applied research components and involve
three or more independent researchers, whether co-located or located at
different universities or research institutions, plus industrial/government
collaboration. Funding for a theme with 4-5 projects would
be typically in the order of $250,000 per year. Additional funding of the
order of $50,000, in cash or in-kind, or more must come from industry and/or
government. The project duration should be two to four years. Deviation
from typical projects must be justified.
GISST
Selection Criteria
· Research
methodology
· Industry/government
contributions and socio-economic benefits
· Excellence
of the project team
· Collaboration
and linkage in the network
· Project
management
· Budget
and milestones
The
GISST Major Project Proposal
August
26, 1999
(ii) Full
Project Proposal (8 pages):
September
13, 1999.
Interested
researchers who wish to submit a project proposal should notify the co-leaders
of their intention ASAP by sending an email to gisstnce@cs.ubc.ca
and
submit a Project Summary and Project
Proposal as follows:
Project
Summary (3
pages plus CV):
One
electronic copy of the major project summary and informal CV of all participants
must be sent to the GISST NCE Coordination
Office via the following e-mail address: gisstnce@cs.ubc.ca
by August 26, 1999. The
Project Summary of 3 pages should include a description of the theme objectives
and an abstract of each proposed project in the theme, as well as the names
of the theme (co)leaders and project leaders. In
addition, it should include a list of potential industrial or government
sponsors with the name of contact persons and the expected amount of funding
(cash and in-kind) from every sponsor for each project. Preliminary
screening will be made based on the project summary. Notification
will be made by August 31, 1999 as to whether the theme (co)leader(s) should
go forward with the full project proposal.
Full
Project Proposal (8
pages plus CV):
One
electronic copy of the full major project proposal and CV (in NCE format)
of all participants
must be sent to the GISST NCE Coordination Office via the following e-mail
address: gisstnce@cs.ubc.ca. Also,
one paper copy of the same - full major project proposal and Curriculum
Vitae (in NCE format) of all the participants - must be sent to the GISST
NCE Coordination office. Notification of inclusion of the major project
proposal in the Full GISST NCE Application will be sent out by September
20, 1999.
Format
for the Full Project Proposal:
· No
more than 8 pages (excluding CV)
· Electronic
versions must be prepared with Microsoft Word 6 or 7 word processor.
· Page
format: 8½ x 11 inches
(21.5 cm x 28 cm) with margins of ¾ of an inch (1.7 cm) all around.
Enter the project leader on the top of every page.
· Type
size: no smaller than 12 pts.
General Presentation
The
following points must be addressed:
Clearly
identify the project leader(s) and participants. Names, PIN (Personal Information
Number of NSERC, if applicable), affiliations and time (% of total salaried
time) committed to the network activities are all required. A participant
can be from the Canadian academic, public or private sector and must be
responsible for certain aspects of a network-funded research project.
Include
a Curriculum Vitae for every participant in the NCE standard format
(no other format will be accepted). Instructions and form are available
on the NCE Web site: http://www.nce.gc.ca/publi.htm
· Research
Description (3
pages):
Give
a short description of the following:
- the
current state of knowledge in the field;
- the
scope of the research, its focus and specific objectives;
- the
proposed methodology;
- the
expected results;
- the
project milestones;
and
include a list of references.
· Benefits (1
page):
Briefly
review the following:
- the
social and economic impacts of the research (1/2 page);
- the
training of Highly Qualified Personnel;
- the
linkage to the network themes and thrusts.
· Networking
and Partnership (1/2
page):
Discuss
the following:
- why
networking (and possibly multidisciplinary , multisectorial) approach is
important in achieving the project’s objectives;
- the
project strategy to ensure open communication and maximize the use of available
resources;
- the
expected participation of each participant including the government and
industrial partners, and their ability to contribute to the research.
· Knowledge
Exchange and Technology Transfer (1/2
page):
Discuss
the project’s plans and strategy with respect to:
- accelerating
the exchange of research results within the project and to the network;
- the
use of this knowledge within Canada by organizations that can harness it
for Canadian economic growth and social development (i.e. technology transfer);
- contribution
to the development of a sustainable network in NGI and its applications;
- the
strategy for commercial exploitation.
· Project
Management (1/2
page):
- show
the project management structure, defining the roles and responsibilities
of the key personnel;
- address
the management roles played by industrial and government partners.
· Budget (1
page):
Before
completing this section, refer to the NCE Program Guide for the list of
eligible expenditures on the NCE Web site: http://www.nce.gc.ca/books/guide.htm
a) Provide details for every year of the project duration (max. 4 years), for the proposed expenditures from NCE (GISST Network) funding (in tabular format):
- Salaries
and stipend:
· Graduate
students
· Postdoctoral
fellows
· Technical
and professional assistants
· Other
salaries (specify)
- Operation
of core facilities
- Equipment:
· Purchase
or rental
· Maintenance
costs
· Operating
costs
· Other
equipment costs (specify)
- Materials
and supplies
- Computing
costs
- Travel
expenses:
· Field
trips
· Conferences
· Other
travel costs (specify)
- Management
and networking:
· Research
management
· Workshops
· Other
management and networking costs (specify)
- Other
expenditures (specify)
b) Provide details for every year of the project duration (max. 4 years) for the contribution from partner organization (complete one table for each contributing organization):
- Cash contributions
- In-kind contributions:
· Donation of equipment, software
· Donation of material
· Logistical support of field work
· Provision of services
· Use of company facilities
· Salaries of scientific staff
· Salaries of administrative staff
· Project-related travel
· Other (specify)
For
Additional Information
CICSR/CS 289
University of British Columbia
2366 Main Mall
Vancouver,
BC V6T 1Z4
Tel: (604)
822-6894
Fax: (604)
822-9013
Email: gisstnce@cs.ubc.ca
Web: http://www.cs.ubc.ca/~gisstnce/home.html
Co-Leaders:
Prof. Son
T. Vuong
Computer
Science Department
University
of British Columbia
Tel
: (604) 822-6366
Fax
: (604) 822-5485
Prof. David
Lorge Parnas
Department
of Computing and Software
McMaster
University
Tel: (905)
525-9140 Ext 27353
Fax:
(905) 525 6246