Evaluation of Tools for Hairy Requirements and Software Engineering Tasks
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Context and Motivation A hairy requirements or software engineering task involving natural language (NL) documents is one that is not inherently difficult for NL-understanding humans on a small scale but becomes unmanageable in the large scale. A hairy task demands tool assistance. Because humans need help in carrying out a hairy task completely, a tool for a hairy task should have as close to 100% recall as possible. A hairy task tool that falls short of close to 100% recall that is applied to the development of a high-dependability system may even be useless, because to find the missing information, a human has to do the entire task manually anyway. For a such a tool to have recall acceptably close to 100%, a human working with the tool on the task must achieve better recall than a human working on the task entirely manually. Problem Traditionally, many hairy requirements and software engineering tools have been evaluated mainly by how high their precision is, possibly leading to incorrect conclusions about how effective they are. Principal Ideas This paper describes using recall, a properly weighted F-measure, and a new measure called summarization to evaluate tools for hairy requirements and software engineering tasks and applies some of these measures to several tools reported in the literature. Contribution The finding is that some of these tools are actually better than they were thought to be when they were evaluated using mainly precision or an unweighted F-measure.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.011 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it