next up previous
Next: Terms and Term Databases Up: Term Rewriting for Sale Previous: Measurements


Research Issues

Before presenting a number of research issues that may interest people working in the field of term rewriting, we give first some further sobering observations on the use of formal techniques in industry,

As already indicated in Section 2, IT industry is wary for new concepts. Some researchers may think that, e.g., correctness proofs will be beneficial for IT industry. This is not true. The return on investment (ROI) is poor: $0.90 after one year $0.80 after two years, $0.70 after three years and $0.60 after four years [35, loc. cit. p. 108]. The ROI of the reusability of high-quality artefacts is excellent: $3.75 after one year $9.15 after two years, $21.75 after three years and $43.75 after four years [35, loc. cit. p. 105]. The list of ROI of selected software technologies contains 143 entries, reuse is number one and correctness proofs are number 136. So if one thinks that we might have an unusual viewpoint on the research issues, namely no termination proofs and no confluence proofs, note that this is confirmed by cost estimations from IT industry (this does not necessarily mean that we do not like proofs, see [26] for an example).

In the following subsections we propose therefore research issues that encourage the amalgamation of several theoretical directions. The goal is to achieve high-level, high-quality artefacts that can be reused in a wide range of applications.



Subsections
next up previous
Next: Terms and Term Databases Up: Term Rewriting for Sale Previous: Measurements
Paul Klint 2001-06-12