<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-0303">
  <Title>Referential Annotations</Title>
  <Section position="7" start_page="17" end_page="19" type="concl">
    <SectionTitle>
6 Future Directions
</SectionTitle>
    <Paragraph position="0"> In the previous section we have compared two perspective-dependent annotation schemes that use a particular level of linguistic annotation as their primary organizing principle and have contrasted them with the perspective-independent annotation-graph model. We believe that both types of representation models have their independent justification. Perspective-based representations, such as SYN-RA and pie-in-the-sky, are well-justified for particular application scenarios. For example, for text summarization and other semantic tasks, the pie-in-the-sky model seems particularly well-motivated since the pertinent semantic information can be easily extracted from its predicate-argument-structurerooted feature structures. For other tasks, such as anaphora resolution, for which syntactic information is more relevant, the syntax-based representation of SYN-RA allows for an easier extraction of the relevant information for rule-based, statistical,  and machine-learning approaches to computational anaphora resolution. More generally, perspective-based representations are highly task-dependent. It would be misguided to consider them as ideal, task-independent annotation standards. If one wants to establish a task-independent annotation standard, then a perspective-independent annotation scheme such as the annotation graph model looks like a promising direction for future research. In particular, such research should focus on techniques that allow for easy conversion of perspective-independent representations to task-dependent views of the relevant linguistic information.</Paragraph>
  </Section>
class="xml-element"></Paper>