The speakers at this week’s seminar will be Kishore Swaminathan, Director of Research, Accenture Technology Labs. While not strictly a UBICOMP topic, it seems to be relevant to the way future systems in general will be built and may be illuminating for how UBICOMP services will be structured.
“Traditional factors such as cost and quality as well as new factors such as offshore development, platform consolidation, massive data volume, hardware virtualization, increasing cost of managing legacy systems etc. are poised to change the field of Information Technology dramatically in the next decade. In particular, I believe that three major trends will dominate IT over the next several years:
Software will be architected differently: Currently, functional cohesion in complex software is provided by tightly-coupled, monolithic design typified by ERP systems. Service Oriented Architecture or SOA promises to achieve the same through loosely coupled architecture. While SOA can provide significant benefits such as encapsulation, inter-operability, dynamic binding and resistance to change, it also represents a considerably different computational paradigm that will create a number of new technical, research and operational challenges.
Software will be developed differently: Currently, software design and development is essentially a craft practised by craftsmen. But there are technical and economic indicators to suggest that software development will move more and more toward industrialization. On the one hand, mathematical models and metrics will influence the current software development processes; on the other hand, Model Driven Development through Domain-Specific Languages will challenge the current paradigm by separating functional models from computational models, potentially leading to a highly automated process for software development.
Software will be bought and sold differently: New technologies and new business models are beginning to threaten the current economics of software systems and information technology. Software-as-a-service or SaaS (as opposed to software as a product) is gaining traction. Technologies such as RIA (Rich Internet Application) will attempt to create a “network operating system” and make SaaS considerably more appealing to home and business users. Open source – as a social and economic phenomenon – will influence and disrupt the economics of the IT industry.
In this talk, I will touch upon some of these issues and discuss how these factors could affect the IT industry and computer science research.
The Informatics Seminar is held in ICS2 136 at 3pm, followed by a happy hour at 4pm. See you there!
This talk will not be videocaptured.