Software Component integration Markup Language (SCiML)

Integration is a painful process due to programming complexities of the existing technologies and methodologies. This thesis presents a new framework and an associated vocabulary implements this framework. This new technology aims to further interoperability and adaptability among systems and their constituent parts, especially which of distributed systems dispersed across a plethora of platforms and architectures.

The thesis first concentrates on the problems pertaining to current integrated systems. At present, system integration relies heavily on high-level language programming. These languages and their underlying technologies like COM, CORBA are inherently complex. Due to this complexity, there is also a likely hood to err. In addition, these integrated systems are less adapted to change and ‘plug-and-play’ style integration.

During the survey of existing technologies and methodologies it was observed that though may of the legacy systems were integrated using structure charts there has been no attempt to map such design directly into the functionality present in the finally integrated system. Therefore, the framework aims at map interaction depicted in structure charts directly into a working system. Furthermore, despite the advancement of Object Oriented concepts, which creates interaction in web like fashion, application integration in general is still done in a structured fashion; therefore, a case established that if the framework needs to deal with such interaction. In this chapter concepts of modularity and extensibility present in Tcl and XML upon which the new framework can be build is identified.

Then the framework and SCIML is then discussed in details. The framework is quite simple – and it was deliberately kept so. This framework and associated technologies were designed so that it would be easy to learn and use. In fact, it borrows some of the concepts of Tcl and tries to ‘marry’ it with that of XML to give a simpler outlook of components combine to form the integrated system. The XML vocabulary: SCIML – is a more concrete implementation of this new framework. It designed so that it inherits the positive features of Tcl and that of XML and is designed so that it can directly map a ‘structure chart’ into a working system. Furthermore, it is compact and easy to learn.

In conclusion, this framework and SCIML provided mechanism for easy system integration. This framework would bridge this gap between how integration is conceptualized and finally implemented. Using this new framework and SCIML systems integrates will only have to be concerned about the constituent components of a system and their interactions, and interdependence. Specifying them in SCIML or a like vocabulary would result in the integration broker automatically gluing the system to form the integrated system. Thus, the coding effort in integration would be eased. Since SCIML or like vocabulary is text based and succinct the structure of the integrated system can be easily manipulated programmatically or manually.