aboutTheDataAccidents = (title, numberOfSubjects) => md`
- **Digitised newspaper pages**: automatically-transcribed text and images of historical newspaper articles come from newspapers digitised for the Living with Machines project by FindMyPast. These articles were selected by searching for keywords related to accidents and machinery in close proximity in an article. As such, they're a small sample of articles from public domain newspapers; many, many more relevant articles are available on the British Newspaper Archive, and many are still yet to be digitised.
- **Information about the newspapers** themselves - their publishers, locations, interests, circulation and cost - come from Mitchell's Press Directories, digitised and processed by the Living with Machines project.
- **Crowdsourced classifications**: volunteers on the Zooniverse platform were shown an article from a 19th century newspaper alongside options to label it - most importantly, did it mention an accident involving machinery? If an article did mention a specific accident involving machinery, then we asked additional questions it, such as the age and gender of any victims mentioned, the reported location of the accident, and the type of site where the accident occurred. Annotations were collected from 2020 to 2023. We exported the volunteers' annotations, comments and tags, and related information from the Zooniverse platform, and are working with that and data extracted from natural language processing/machine learning, for analysis and presentation.
- **Zooniverse tasks**: We went through several iterations of the 'workflows' about accidents, experimenting with the best combation of questions, task types and backend logic, alongside testing different ways of using the results.
- The primary source for the *${title}* data are from 23 newspapers titles over 74 years from 1846 to 1920, resulting in ${numberOfSubjects.toLocaleString()} images (**subjects**).
`