Towards the definition of a basic toolkit for HLT

This paper intends to be an initial proposal to promote research and development in language independent tools. The definition of a basic HLT toolkit is vital to allow the development of lesser-used languages. Which kind of public HLT products could be integrated, at the moment, in a basic toolkit portable for any language? We try to answer this question by examining the fifty items registered in the Natural Language Software Registry as language independent tools. We propose a toolkit having standard representation of data and develop a strategy for the integration, in a common framework, of the NLP tools.