Kcite allows enables referencing of scientific articles by their primary identifier . As we currently support DOIs, arXiv, and PubMed identifiers this works well for most scientific articles, as well as some forms of data . It is also possible to reference arbitrary URLs which cover all web documents.
Books, however, are problematic. While they generally do have an primary identifier, that is an ISBN code, it is not easy to translate these into the relevant bibliographic metadata, as there is no free web-hosted translation service available.
Kcite and Greycite allow us to take an alternative approach however. We now support two library catalogues which provide distinct URLs for books: WorldCat and Open Library. These can be used both on some fairly old books, such as Principia , or Origin of the Species , or more modern literature in English , or in translation .
Currently, both WorldCat and OpenLibrary are somewhat inconsistent in the metadata that they present, which means that this form of referencing will not work for all books on their websites. For instance, and surprisingly, while the Spanish Harry Potter reference above works, the English equivalent seems not to have any metadata. While this situation is less than ideal, we believe that tools such as Greycite which attempt to make bibliographic metadata useful to authors and readers will help to improve this situation .
Kcite preserves and displays the URL, so readers can get full details of books if they wish to obtain them, or get full information regarding editions. While, ideally, it would make more sense to use permalinks, in practice, any link on WorldCat or OpenLibrary could be used. Greycite stores metadata about links, so this will remain even if the links change.
School of Computing Science