TCS2004 will be held as part of the
August 23-26, 2004, Toulouse

Challenges of the era

In recent years, IT application scenarios have evolved in very innovative ways. Highly distributed networks have now become a common platform for large-scale distributed programming, high bandwidth communications are cheap and widespread, and most of our work tools are equipped with processors enabling us to perform a multitude of tasks. In addition, mobile computing (referring specifically to wireless devices and, more broadly, to dynamically configured systems) has made it possible to exploit interaction in novel ways.

To harness the flexibility and power of these rapidly evolving, interactive systems, we need to come up with radically new foundational ideas and principles. Now is the time to develop the theoretical foundations required to design these systems and to cope with the many complex issues involved in their construction. Now is also the time to develop effective principles for building and analyzing such systems.

Our computational goal is to discover techniques, models and algorithms allowing us to construct systems that are flexible, dependable, secure, robust and efficient. Our prime focus is not how to represent and manipulate data efficiently, but rather how to control and coordinate the entities in the system. Cost and performance measures must account for the integration of communication and computing, as well as the natural interplay between structural information and complexity. Algorithms must be developed to prescribe rules for system entities that are liable to move, fail or malfunction. Security, reliability, robustness, failure modes and risk are all new parameters entering the field of analysis and design.

In terms of programming, it is interesting to note that Internet applications differ substantially from traditional applications in the following areas: scalability (huge number of users and nodes); connectivity (availability and bandwidth); heterogeneity (operating systems and application software); and autonomy (administration domains with strong control of their resources). Hence, new programming paradigms (thin client and application servers, collaborative "peer-to-peer," code-on-demand, mobile agents) have been proposed for Internet applications. These emerging programming paradigms require mechanisms to support mobility of code and computations, as well as effective infrastructures to support coordination and control of dynamically loaded software modules. Furthermore, a semantic and logic framework to formalize Internet computations is clearly required as well. Such a framework may provide a formal basis to discuss and motivate controversial design/implementation issues, and to state and certify properties in a rigorous way. Crucial issues in the development of Internet applications involve the control of component interactions, since some components can be dynamically downloaded from the network. Security architectures can monitor the execution of mobile code to protect a host from external attacks on private information.

This scenario is very appealing from a scientific point of view, since most of the open problems and several of the concepts requiring fine-tuning fall within the domain of computer science research. More specifically, there is a need for models, languages and logics that are distributed, interactive and concurrent; open and reconfigurable; higher order and typed; equipped with abstract compositional semantics; efficiently verifiable.
 

 Benefits

For those who are already involved in the research process, TCS 2004 is a meeting place to present new results, share insights and ask pressing questions. For those interested in the process, TCS 2004 is an opportunity to learn about current and future orientations from active researchers in the field.