The Initial Impact of the Open Scholarship Policy Observatory

In September 2016, members of the Implementing New Knowledge Environments (INKE) Partnership—a broad, diverse group working to advance understanding of, and resolve critical issues in, the production, distribution and widespread engagement of digital scholarship in Canada and beyond—met to discuss future directions and focus areas. One of the resulting initiatives is the Open Scholarship Policy Observatory. The Open Scholarship Policy Observatory tracks national and international policies and policy changes in order to assist INKE partners with developing timely and responsive policies. This paper describes the development of the initiative, and reports on the initial impacts the project has had to date. In September 2016, members of the Implementing New Knowledge Environments (INKE) Partnership—a broad, diverse group working to advance understanding of, and resolve critical issues in, the production, distribution and widespread engagement of digital scholarship in Canada and beyond—met to discuss future directions and focus areas. One of the resulting initiatives is the Open Scholarship Policy Observatory. The Open Scholarship Policy Observatory tracks national and international policies and policy changes in order to assist INKE partners with developing timely and responsive policies. This paper describes the development of the initiative, and reports on the initial impacts the project has had to date.


Introduction
The Implementing New Knowledge Environments (INKE) Partnership for Networked Open Social Scholarship brings together a broad, diverse group to advance understanding of, and resolve crucial issues in, the production, distribution, and widespread engagement of digital scholarship in Canada and beyond. The Open Scholarship Policy Observatory (C-SKI 2017) is an INKE Partnership initiative that tracks national and international policies and policy changes in order to facilitate a stronger understanding of the current state of open scholarship.
The Policy Observatory reflects these findings back to Partnership members, along with local institutions, associations, consortia, and government bodies, in order to assist these groups with developing timely and responsive policies. The initial conception of such an observatory emerged at an INKE Partnership partner/researcher meeting held at the University of Toronto Scarborough in September 2016. Originally suggested by representatives from the Canadian Association of Research Libraries (CARL), the observatory was envisioned as a means of concentrating efforts amongst the group on a shared area of interest: open scholarship policy. The Policy Observatory is coordinated by the Canadian Social Knowledge Institute (C-SKI), based at the Electronic Textual Cultures Lab (ETCL) at the University of Victoria.
The Policy Observatory examines open scholarship policies in the contexts in which they were created, and aims to provide a firm foundation for the development of relevant policy recommendations. This involves considering a variety of related policies, some of which are directly concerned with open scholarship. Other policies that are covered in the Policy Observatory may not mention open scholarship, but nevertheless have a profound impact on the wider ecosystem within which open scholarship takes place. Although the primary focus of the Policy Observatory is the Canadian context, international policies often provide useful counterpoints to their Canadian counterparts and therefore may also be considered. By building a framework around policies, the Policy Observatory hopes to disentangle what Stephen Pinfield (2015) calls 'mandate messiness'-a general confusion over the strategies and meanings of various open access mandates and policies.

The Importance of the Open Scholarship Policy Observatory
As open access (OA; that is, free, unrestricted access to peer-reviewed journal literature) and open scholarship become increasingly prevalent, the need to better understand the tensions between various mandates and policies becomes all the more imperative. The Registry of Open Access Repository Mandates and Milligan, Sarah, et al. 2019. The Initial Impact of the Open Scholarship Policy Observatory. KULA: knowledge creation, dissemination, and preservation studies 3(1): 16. DOI: https://doi.org/10.5334/kula.43 Policies (ROARMAP), initially launched in 2003, contains hundreds of OA policies from all over the world. However, as policies are changed, updated or moved, the registry inevitably falls out of date. Worse still, once the links are broken, it can become very difficult to find a policy document again. Indeed, often the Internet Archive's Wayback Machine is the only solution for finding a lost policy. The ROARMAP policy entries themselves usually include the policy's terms, but are divorced from the details of the policy's context and reception. Institutional policies on OA vary greatly in what they actually require. In his definitions of OA policies, Peter Suber (2012) distinguishes between 'request or encouragement' policies and three different types of mandates: loophole mandates, which 'require Green OA except when the author's publisher doesn't allow it'; deposit mandates, which require immediate deposit in an OA repository upon acceptance for publication, but allow a delay in making the publication OA, or even allow the publication to remain dark if the publisher does not allow OA; and rights-retention mandates, which require an immediate deposit, but also secure permission to make the deposit OA. Yet even this final and strongest type of mandate allows faculty to opt out via waiver. Suber (2012) argues that the term 'mandate' itself is misleading since 'no policy anywhere pretends to impose an unconditional OA requirement.' Despite Suber's helpful definitions, the variety of policies and mandates stipulating Green or Gold, libre or gratis, immediate or embargoed access can be bewildering, with each policy seeming to advocate for a different strategy to openness. 1 Some OA advocates, including Suber, have endorsed this multiplicity of approaches, while others, notably Stevan Harnad, have made a case for directing all energies toward the path of least resistance which is, according to Harnad (2015), mandating Green OA. Regardless of whether this multiplicity is productive or not, it can certainly be confusing. As a result, it is often challenging to know what OA policies mean, and even more so to understand how they are enforced (Kingsley 2012). The Policy Observatory hopes to help rectify these issues.
Open scholarship extends beyond OA to focus ' on the wide and broad dissemination of scholarship by a variety of interconnected means (e.g., technology, licensing) aiming to broaden knowledge and reduce barriers to access to knowledge and information' (Veletsianos 2016, 16). In 2012, Danny Kingsley referred to the management of research data as the '"elephant in the room" worldwide' (5). Data management has received increasing attention since then, particularly in connection with open data policies. The 2015 Tri-Agency Open Access Policy on Publications includes a special clause for recipients of Canadian Institutes of Health Research (CIHR) funding, which articulates the types of data that must be made open (Government of Canada 2015); a year later the Tri-Agency Statement of Principles on Digital Data Management was released, which addresses ' digital data management practices and data stewardship in agency-funded research' (Government of Canada 2016). Recently, the G7 Science Ministers organized an Open Science Working Group that produced recommendations that targeted access to, and management of, research data. The UK funder Wellcome Trust's most recent open data policy requires funding recipients to make their data, original software and research materials openly available in a timely manner to facilitate validating or replicating results (Wellcome Trust 2017). Governments are also opening up their data: the Government of British Columbia was the first provincial government in Canada to publish its data under an open license as part of its 2011 Open Information and Open Data Policy. Other provincial governments have since followed suit. The term open scholarship, then, is meant to encompass all of these movements and more, and to represent the holistic shift in approaches to scholarship. Given this, the Open Scholarship Policy Observatory must consider policies advocating openness alongside those which do not address openness at all. Policies, like scholarship itself, cannot be created in a vacuum. The Policy Observatory seeks to contextualize the movement towards openness within broader conversations occurring at governmental, legislative, organizational and institutional levels.

The Open Scholarship Policy Observatory in Practice
At the core of the Open Scholarship Policy Observatory is its website at: https://ospolicyobservatory.uvic.ca. Content is sorted into three categories: observations, responses and policies.
The key type of output produced to date are descriptive overviews, or observations, of important policies. These observations are shaped around answering several questions: (1) what is the policy, (2) how has it been received by both the communities it impacts and those on the periphery, and (3) how does the policy relate to open scholarship as a whole? As their name might indicate, the observations are meant to be neutral and descriptive. They provide a foundation for understanding a particular policy, its context and its impact. In developing these observations we are conscious of our responsibility to present the policies and the reception they received as objectively and fairly as possible. We have covered the following topics as of These topics are varied: some represent Canadian policies or recommendations; others are international in their scope; and still others represent INKE Partnership members' strategic directions. In selecting these topics, we wanted to ensure coverage of fundamental policies, such as the Tri-Agency Open Access Policy on Publications, while also responding to policy changes as they happened, such as the G7 Expert Group on Open Science, which met in September 2017. Despite the variety, we are already identifying important areas of overlap between the policies, which is reflected in the website's tagging system. As new content is added, we expect to see common threads continue to be woven together throughout the topics covered. Indeed, one topic will often inspire another as we piece together a fuller picture of open scholarship policies.
We also made the decision early on to present content in both French and English, when possible. Although there are limitations to what we can do as a primarily Anglophone team, we felt that translating the observations, in particular, would not only better represent our Francophone Partnership members and their constituents, it would also leave the door open for French content, which could be translated into English.
While the observations are intended to cover a topic on behalf of all Partnership members, we are also conscious that individual members have particular and valuable perspectives on certain key policies. As soon as we circulated drafts of the initial observations to INKE Partnership members, they responded with not only feedback on our coverage, but with commentaries on the policies themselves, anticipating our request for exactly these types of responses. The first partner response to be added to the website is Brian Owen's response to Canada's Fundamental Science Review. Writing from his perspective as Associate Dean of Libraries at Simon Fraser University Library and as Managing Director of the Public Knowledge Project, Owen identifies both the Review's strengths and the areas he believes could have been more fully covered (2017). Responses such as these enhance our observations by bringing a critical, on-the-ground eye to new policy-related documents.
The third section, 'Policies,' is further divided into a large policy collection and smaller policy clusters, representing two distinct approaches to tracking policies. The policy collection is a growing list of open scholarship policies from around the world. The collection relies on the ROARMAP collection as a starting point, but updates broken links, where possible, and removes policies that can no longer be recovered. Furthermore, we create an archived PDF version of each policy on the day we add it to our collection in an effort to ensure that policy updates do not overwrite policy history, and to allow us to better track developments in open scholarship policies as they evolve. Alongside the hundreds of policies included in the policy collection, we have also created 'policy clusters.' These are smaller groupings of policies that relate to one another. By bringing these clusters of policies together, we hope to provide helpful context to the open scholarship movement as a whole, and to the policies featured in our observations in particular. Our first policy cluster includes the OA statements from Budapest (February 2002), Bethesda (June 2003 and Berlin (October 2003)-three foundational statements that have defined the Open Access movement for the last decade and a half, and that are collectively known as the 'BBB' definition of OA. Our second cluster is of Canadian university statements in support of OA. While almost every Canadian university provides resources to support OA, a handful have also published their own declarations in support of the movement. Policy clusters are intended to provide a simple entry point into a particular set of policies. They address fundamental questions such as: What is the BBB definition of OA? Or, Which Canadian universities are leading the way in the OA movement? They also provide a jumping-off point for further research.
Our Twitter presence and forthcoming Twitter archive form the final piece of the Policy Observatory. We have been tweeting and retweeting relevant content using the hashtag #OSPolicyObs from our INKE Twitter account (@INKEProject). Once the website formally launches, this hashtag will feature new site content as it is added. However, both now and moving forward, the hashtag is also used to capture policy changes and responses to them as they occur in real time. By archiving this hashtag on the website in the forthcoming Twitter archive, we will produce a permanent record of these changes. The hashtag also serves an internal purpose of signaling potential future topics.

Conclusion and Next Steps
To date, all Open Scholarship Policy Observatory efforts have been about laying the groundwork. The Policy Observatory is intended to extend beyond its small team, and, indeed, beyond the INKE Partnership. To this end, we are already in discussion with other international groups pursuing similar initiatives in order to broaden our efforts and deepen our engagement worldwide. We also invite INKE Partnership members and members of the broader community to participate in the Policy Observatory. The more policies, observations and responses that are added, the stronger our grasp of open scholarship polices and the ecosystem in which they exist becomes. In his chapter on policies in Open Access, Peter Suber concludes: Every institution adopting a new policy brings about OA for the research it controls and makes the way easier for other institutions behind it. Like many other policy issues, this is one on which it is easier to follow than to lead, and we already have a growing number of leaders. A critical mass is growing and every policy is an implicit invitation to other institutions to gain strength through common purpose and help accelerate publisher adaptation (2012, 95).
Suber, along with many other advocates for OA, has also made the point that the greatest barrier to achieving it is misinformation, or misunderstanding (2012). Indeed, it is to this challenge that the Policy Observatory seeks to respond. As this critical mass of open scholarship policies grows, understanding becomes all the more important.