Best wishes for 2021 from the ScienceOpen team! To kick off the new year, ScienceOpen would like to invite publishers and journal editors to get a free metadata check from our technical experts to ensure that your research is found and read by the widest possible audience. With ScienceOpen’s Metadata Services, we can help you take the first steps towards more impact. In the midst of a global pandemic, the digital transformation in academic publishing has accelerated. With over 2.5 million articles published annually, it is essential that scholarly content can be discovered, understood and processed by computers. The need for rich, machine-readable metadata has become more critical than ever.
A rundown on metadata
Metadata includes the basic information about your article, book or chapter such as author, title, journal, publication date, but also license and copyright information, institutional affiliation, funder, abstract, author contact, references and more. In the ideal case, an article has a Digital Object Identifier (DOI) that links to the URL of the version of record. The DOI, registered as a rule with Crossref or Datacite, is combined with a set of basic metadata about the article or book. For general processability, this metadata should be described in a standardized XML format, usually using the standard JATS, BITS, or Dublin Core tag suites, but there can be a great deal of variability. Metadata may be customized by typesetters or generated from a manuscript submission system. Publishers often use different metadata sets for their web sites and other services. And different abstracting and indexing databases have varying requirements. Also, the set of optimal metadata continues to evolve and grow, which creates yet more challenges.
Best practice metadata enhances discoverability
Structured metadata is a significant driver of discoverability. Abstracts, for example, provide words and concepts to drive machine-learning recommendation engines on a range of platforms. On ScienceOpen we see 25x more views for articles with abstracts in our research collections, due to article recommendation and browsing behaviours. Links to and from citing articles also drive usage, and we have leveraged openly accessible citations to create our own citation network on ScienceOpen. We are therefore happy to support initiatives like the Initiative for Open Abstracts (I4OA), the Initiative for Open Citations (I4OC) and Metadata2020, which are actively trying to make metadata more uniform, shareable and open. As such, affiliation and funding data can be used in creative ways that may promote your digital content unexpectedly. For example, the UCL and the UN Sustainable Development Goals Collection gathers publications around the 17 SDGs with an affiliation to the University College London.
How ScienceOpen can help
On ScienceOpen, we can show you how computers see your article metadata and help you to develop strategies to increase its machine-readability and discoverability. We will start by analysing your DOIs. We can work with you to deposit DOIs with Crossref or enhance your metadata there. We can analyse your website and RSS feeds to give you feedback on the machine-readability of your site. Feedback on sample metadata XMLs can reveal important weaknesses in your workflows. We are also happy to talk about linking to preprints or DOIs for book chapters or peer review reports. Our technical team would be happy to meet you!