To enhance government public service efficiency, an information seamless flow across government agencies is required. The standardization and simplification of data set to be used in electronic data exchange is significant. The agreed guideline and framework is required in a real life application. Thai government created an e-government interoperability framework (e-GIF) to be used as a tool for data exchange implementation. It includes five main parts; business process modelling, data set standardization technique, XML schema building technique, a collection of agreed interoperable technical standards and change management. For the business process modelling, a technique based on UN/CEFACT Modelling Methodology has been used. For the data set standardization, a technique based on CCTS has been used. For XML schema building, a technique based on UN/CEFACT XML naming and design rules has been used. It is found that the reusability of the agreed data sets across various agencies is one of the significant success factors. In 2007 a project called electronic correspondent letter management system (e-CMS) began to standardize a data set in the domain of electronic official letters. In the project 30 ministerial departments involved to use and re-use the data set to achieve the seamless flow of electronic letter across those agencies. However in the project the data set is only in one domain of interest which is a letter. A new project had begun in 2008 to motivate more agencies to build further more standardized data sets from seven different domains. The finding is that the reusability of the data sets across various domain of interest is another significant success factor. In this project some data sets in the domains of education, labor, health, agriculture (livestock), research, governmental human resource and registrations had been built. Among the data sets from each domain, some data sets are very similar across different domain. To harmonize the similar data sets and reuse them as common data sets in different domain becomes significant. The common data sets are for examples; 'Person', 'Address', 'Education' and etc. The data set 'Person' appeared as 'Student' in education domain. It appeared as 'Labour' in labour domain and again appeared as 'Patient' in health domain. In this paper we illustrate that the reusability in those various domains increases the reusability across agencies which in turn enhances the seamless flow across government agencies.
