An exploration of evolutionary change in an example of scientific software
Bibliographic record
Abstract
Scientific software is typically long lived and does not seem to decay, despite high complexity and a relative lack of textbook software engineering practices. The study of such software may provide useful guidance for the evolution of other kinds of software. There is a wealth of examples to study. Unfortunately, most lack historical documentation, so new techniques have to be found to do longitudinal evolutionary studies. This thesis develops a change identification technique based on examining changes in the central data structures of the software. This technique is used to examine change in a particular example of successful, long-lived scientific software. Change is classified and analysed using a new change model, which is based on the idea that drivers for change come from a set of knowledge domains and are filtered by factors present in the environment, including the software itself. The combination of change drivers and change filters successfully accounts for the distribution of changes as observed. The thesis then considers the response of the original software design to changes that happened over roughly a 20 year period. A set of metrics is used to quantify the impacts of the changes, and to search for patterns of stability among a base version and three later versions of the software. Surprisingly, given the very different forces at work on the four versions, some interesting patterns of stability emerge. Some of the results of the thesis challenge established practices, such as the relative importance of information hiding, or the need to do extra work to prevent decay.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".