<?xml version="1.0" standalone="yes"?>
<Paper uid="P99-1058">
  <Title>A semantically-derived subset of English for hardware verification</Title>
  <Section position="6" start_page="454" end_page="455" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> Much work on controlled languages has been motivated by the ambition to &amp;quot;find the fight trade-off between expressiveness and processability&amp;quot; (Schwitter and Fuchs, 1996). An alternative, suggested by what we have proposed here, is to bring into play a hierarchy of controlled languages, ordered by the degree to which they semantically approximate the target formalism. Each point in the hierarchy brings different trade-offs between expressiveness and tractability, and evaluating their different merits will depend heavily on the particular task within a generic application domain, as well as on the class of users.</Paragraph>
    <Paragraph position="1"> As a final remark, we wish to point out that there may be advantages in identifying plausible restrictions on the target formalism. Dwyer et al. (1998a; 1998b) have convincingly argued that users of formal verification languages make use of recurring specification patterns. That is, rather than drawing on the full complexity of languages such as CTL, documented specifications tend to fall into much simpler formulations which express commonly desired properties. In future work, we plan to investigate specification patterns as a further source of constraints that propagate backwards into the controlled English, perhaps providing additional mechanisms for dealing with apparent ambiguity in user input.</Paragraph>
  </Section>
class="xml-element"></Paper>