Reuse of Wikimedia Commons Cultural Heritage Images on the Wider Web
Bibliographic record
Abstract
Abstract Objective – Cultural heritage institutions with digital images on Wikimedia Commons want to know if and how those images are being reused. This study attempts to gauge the impact of digital cultural heritage images from Wikimedia Commons by using Reverse Image Lookup (RIL) to determine the quantity and content of different types of reuse, barriers to using RIL to assess reuse, and whether reused digital cultural heritage images from Wikimedia Commons include licensing information. Methods – 171 digital cultural heritage Wikimedia Commons images from 51 cultural heritage institutions were searched using the Google images “Search by image” tool to find instances of reuse. Content analysis of the digital cultural heritage images and the context in which they were reused was conducted to apply broad content categories. Reuse within Wikimedia Foundation projects was also recorded. Results – A total of 1,533 reuse instances found via Google images and Wikimedia Commons’ file usage reports were analyzed. Over half of reuse occurred within Wikimedia projects or wiki aggregator and mirror sites. Notable People, people, historic events, and buildings and locations were the most widely reused topics of digital cultural heritage both within Wikimedia projects and beyond, while social, media gallery, news, and education websites were the most likely places to find reuse outside of wiki projects. However, the content of reused images varied slightly depending on the website type on which they were found. Very few instances of reuse included licensing information, and those that did often were incorrect. Reuse of cultural heritage images from Wikimedia Commons was either done without added context or content, as in the case of media galleries, or was done in ways that did not distort or mischaracterize the images being reused. Conclusion – Cultural heritage institutions can use this research to focus digitization and digital content marketing efforts in order to optimize reuse by the types of websites and users that best meet their institution’s mission. Institutions that fear reuse without attribution have reason for concern as the practice of reusing both Creative Commons and public domain media without rights statements is widespread. More research needs to be conducted to determine if notability of institution or collection affects likelihood of reuse, as preliminary results show a weak correlation between number of images searched and number of images reused per institution. RIL technology is a reliable method of finding image reuse but is a labour-intensive process that may best be conducted for selected images and specific assessment campaigns. Finally, the reused content and context categories developed here may contribute to a standardized set of codes for assessing digital cultural heritage reuse.
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 | 0.000 |
| Bibliometrics | 0.007 | 0.006 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".