<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-1515">
  <Title>Experiences with the GTU grammar development environment</Title>
  <Section position="7" start_page="112" end_page="112" type="concl">
    <SectionTitle>
4 Conclusions
</SectionTitle>
    <Paragraph position="0"> Tools like GTU are well suited for learning to develop grammars, for experimenting with grammar formalisms, and for demonstrating the work of computational linguistics. The use of GTU as an educational tool in computational linguistics courses has been very successful. In a recent project GTU's flexibility is being challenged in a joint project with the Institute of Germanic Language at the University of Koblenz. In this project we examine the use of GTU for the benefit of courses in German as a foreign language.</Paragraph>
    <Paragraph position="1"> For the development of large grammars in combination with large linguistic resources and for processing them efficiently, GTU is less suited. We are now convinced that we need an open platform that provides a framework for combining modules for such a task. For this it is necessary to develop interface standards for different types of modules (taggets, grammars, lexicons, test suites etc.).</Paragraph>
    <Paragraph position="2"> Finally, we should keep in mind that a computational environment for grammar development offers help in engineering NLP modules for well-understood phenomena. The real hard problems in NLP (most importantly the resolution of ambiguity) need to be solved by bringing to bear the right information at the right time. But this is of yet a complex area with many questions that have not received a theoretical answer let alone an engineering solution.</Paragraph>
  </Section>
class="xml-element"></Paper>