My name is Dominic Ferrisi and I am director of GrayNet International, an organization now marking its 30th year engaged in the field of gray literature. This field of information ranges from research to publication onto open access,

education, and public awareness of its diverse resources. In 2011, GrayNet International, together with the Dons Easy Archive in the Netherlands, entered upon an enhanced publications project entitled,

Linking Full Text Gray Literature to Underlying Research and Post-Publication Data. The project that is still ongoing and which has been integrated in GrayNet's shared workflow, also sought to encompass retrospective research data from

prior conferences in the GL series. While the types of research data that Dons accepts is in its archive are quite numerous, GrayNet's collection of published research data focuses primarily on survey data, interviews, bibliographic data, and use statistics.

Currently, GrayNet has 52 published datasets in the Dons Easy Archive, of which 20 are accompanied by a data paper. In 2014, the FAIR data principles were formulated and first appeared published in 2016.

They refer to research data that is findable, accessible, interoperable, and reusable. These principles, as they apply to research data, have given further impulse to GrayNet's enhanced publications project,

which is now considered a sustained program embedded in GrayNet's workflow. The challenge then is to demonstrate how these four principles are implemented as they apply to GrayNet's published research data in the Dons Easy Archive.

In 2017, GrayNet introduced the data paper as a means of implementing FAIR principles, and as a tool in publishing GrayNet's research data. In so doing, a new grade literature document type came to expand GrayNet's publication trail,

both in textual and non-textual formats. While metadata is already assigned datasets that are archived and published in Dons Easy, the metadata included in the data paper is further detailed,

providing a technical aspects that deal with the methods applied in capturing the data, the further description of the data, a dataset, onto a statement as to its potential reuse for research, education, and training.

Alongside standard metadata, the data paper seeks to include actionable, persistent identifiers, the PIDs. Emphasis here lies in the fact that the PIDs are linked metadata providing immediate access to resources.

These are namely the DOI assigned the dataset, the ORSIT acquired by the various types of contributors, and the RORID that identifies the research organizations of the contributors. Owing to the inclusion of these various types of PIDs,

the data paper contributes to the further construction of the PID graph. The PID graph links persistent identifiers together via relations in their shared metadata, which enables the discovery of connections at least two nodes away.

Using the PID graph makes it easier to describe complex use cases and relationships. It opens to new and further research via digital interoperable hyperlinks.

The illustration on the left side of the slide indicates that the conference paper published in the GrayGuide repository carries a DOI assigned by GrayNet. The research data associated with the conference paper is published in the DUNS Easy Archive, to which a DOI is assigned.

And the video presentation of the conference paper, published in the TIB-AVE portal, carries its assigned DOI. These three DOIs allow for the cross-linking by Data Site Commons, which provides a single search interface.

If we look to the illustration on the right side of the slide, we encounter the working of the PID graph in which the circles are referred to as nodes, and the lines connecting them as edges.

Starting in the center of this graph, the node in green stands for the RORID of an organization. The surrounding concentric circle is formed by DOIs in blue assigned to conference papers

relating to that organization. And in the outer concentric circle is formed by ORSIDs in red, representing the authors and researchers responsible for the content in the conference papers.

If we look further at the PID graph, we see on the very periphery three connected RORIDs in green. These indicate that the authors to which they are linked are from organizations other than the one identified in the center of the PID graph.

GrayNet's data paper template was first drafted in 2017 and later implemented in the autumn of that year. The template was compiled from a number of existing templates

and has since undergone two versions, each including minor revisions following upon developments in digital publishing. The template is the tool used not only to facilitate in writing a data paper, but also to guarantee a level of standardization and implementation of FAIR data principles.

Its structure contains five main sections and serves researchers in their search and retrieval of relevant data sets. The first section of the template

consists of the most inmost part in providing descriptive metadata related to the authors and the data set. Sections 2 and 3 address the more technical aspects related to the data set, including the method of approach in its creation,

as well as details regarding file formats, licensing, etc. The fourth section describes the potential reuse of the data as well as its limitations. And the fifth and final section provides a listing of actionable linked references.

Note fields are also available in each of the five sections of the template, providing details and explanation. GrayNet's use case relating to data papers focuses equally well on its publication.

The completed data paper is first published as a preprint in the DUNS Easy Archive, accompanying the published research data. The preprint is also entered in the GrayGuide, GrayNet's web access portal and repository. The data paper is then further scheduled for

publication in the Gray Journal, TGG, which is licensed by EBSCO Publishing, and further abstracted and indexed by a number of services including Elsevier's Scopus and Clarifat.

The dual channel of publication via open access and via licensing agent is not seen as a duplication, but rather as a means of making research and resources in gray literature accessible beyond its own community of practice

to wider audiences. Now, five years on since the publication of GrayNet's first data paper, a commemorative issue of the Gray Journal, recently published in the autumn of this year,

contains GrayNet's complete collection of data papers. The commemorative issue seeks to publish or republish its current collection in line with the latest version of the data paper template. The issue further provides a

unique resource that serves in training and education in the field of gray literature. The collection of 20 data papers was submitted by authors from countries worldwide, covering a number of disciplines and diverse topics dealing with gray literature,

such as policy development, open access, leveraging grief resources, digital publishing, circular economy, etc. The collection of data papers offers a variety of research methods used in creating the data sets. It addresses technical details specific

to the data sets, and as mentioned earlier, limitations of the data are explicitly stated. This is characteristic of gray literature. As with any information product or service,

it is important to acquire feedback regarding their use and usage quantitatively or otherwise. Since GrayNet's data papers are published alongside their corresponding research data in DonsEasy, monthly statistics on the downloads of the data are available. Based on recent

statistics, GrayNet's data sets housed in DonsEasy are downloaded on an average of up to three times as often as those data sets that do not have an accompanying data paper.

Data papers that are subsequently published in the Gray Journal are included in EBSCO's quarterly statistics and are likewise based on the number of downloads in that given period.

It is important to mention that GrayNet's data papers are assigned the same DOI as the one DonsEasy assigns the data set. While this is not the case with other content providers, GrayNet considers the combined data set and data paper as an enhanced publication.

While use statistics are in some way an indication of the reuse of the research data spelled out in the FAIR principles, it is by way of their actual usage through citation,

referencing, and sharing of data that reuse is demonstrated. Moreover, it is via education and training in the publication of research data and data papers that the awareness of their value for Gray literature is raised. In an effort to communicate

the benefits that data papers have brought to GrayNet and the potential they hold for other communities of practice in Gray literature, a series of workshops were held both in the States and in Europe. The takeaway from these workshops and trainings are here briefly summarized.

Data papers ensure the implementation of the FAIR data principles. While they rely on rich metadata, it is the actionable persistent identifiers that is the DOI,

the ORSIT, and the ROAR ID together with the inclusion of other linked references that establishes their interoperability. This in turn contributes to the PID graph. While the data paper template provides a much needed tool in writing the data paper, of itself

it is not the final published document. It is through the content contribution of the authors, especially in regard to the potential reuse of the research data, that the strength of the data paper is gained. While the data paper of itself allows for the findability and

accessibility of the associated research data, it is by way of its publication that the promotion and subsequent feedback from references, citations, and downloads that demonstrate

how they come to further scientific communication and knowledge transfer.