<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-2315">
  <Title>Speech Graffiti habitability: What do users really say?</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1.2 Related work
</SectionTitle>
    <Paragraph position="0"> Although several studies have previously explored the usage of constrained or subset languages (for example, Hendler &amp; Michaelis, 1983; Guindon &amp; Shuldberg, 1987; Ringle &amp; Halstead-Nussloch, 1989; Sidner &amp; Forlines, 2002), they have generally been concerned with performance effects such as task completion rates.</Paragraph>
    <Paragraph position="1"> Sidner &amp; Forlines (2002) reported a &amp;quot;correct utterance&amp;quot; rate of approximately 60-80% for their user studies, although this was not a main focus of their work. While we understand the focus on such performance measures, we believe that it is also important to understand how habitable the constrained language is for users, in what ways users deviate from it, and what impact habitability has on user satisfaction.</Paragraph>
  </Section>
class="xml-element"></Paper>