Smarter Open Access Workflows: Digital Science Webinar Summary

As part of a continuing series, we recently broadcast a Digital Science webinar on smarter Open Access workflows. The aim of these webinars is to provide the very latest perspectives on key topics in scholarly communication.

The webinar explored the benefits of an Open Access infrastructure, different types of policies around Open Access, and the challenges universities face when implementing workflows and reporting to funders.

Laura Wheeler (@laurawheelers), Community Manager at Digital Science, started the webinar by giving a brief overview of the esteemed panel and their backgrounds before handing over to Julia Hawks from Symplectic who moderated and questioned the panel.

Julia started by explaining that although the benefits of Open Access are increasingly understood and discussed, there is less attention given to the infrastructure needed to support the Open Access policies being put in place, both by institutions and by funders. Julia also framed one of the key challenges in this space: how to approach getting extremely busy researchers and academic to engage with institutional and funder policies, which they may not fully understand or even be aware of? In Julia’s opinion, it’s not merely about implementing the right infrastructure and workflows, alongside that there is also a need to educate and motivate researchers so they fully engage.

Implementing Open Access policies at the University of California

The first presenter to speak in the webinar was Catherine Mitchell (@catherinemitch) from the California Digital Library, who spoke about the implementation of Open Access policies at the University of California (UC). The first policy implemented only applied to the UCSF Faculty, it was later broadened to all UC Faculty, and then to all UC employees who author scholarly articles. The policy is such that authors are expected to deposit their articles in the university repository, or if appropriate, to clearly indicate that the article is openly available elsewhere.

The constraints on the policy implementation were to provide centralized resources for education and engagement with Open Access, to facilitate the deposit of articles into the repository, to support waiver & embargo requests, and to make the whole system as convenient for faculty as possible.

Catherine explained that their early efforts to implement the policies involved the use of web forms, however, they found that despite efforts to make the forms intuitive and easy to complete, the authors simply found it too time-consuming and inefficient to fill in metadata fields about their publications.

This new system has some unique and important features, such as the automated harvesting of publication records and a streamlined author notification and verification process.

Catherine then explained that the California Digital Library were able to refine their infrastructure using Symplectic Elements to implement a “publications management system”. In Catherine’s view, this new system has some unique and important features, such as the automated harvesting of publication records and a streamlined author notification and verification process. It has been  highly successful, leading to a dramatic growth in the number of deposits into the university repository, in the past year they’ve had over 20,000. Catherine also explained how this increased number of deposits has tangible benefits for the authors and the university. They have direct evidence that there is widespread global usage and readership of the research in their repository.

Outlining some next steps, Catherine talked about the need to engage faculty more directly, as email can be an unreliable method of communication. Catherine also described how they would like to find ways to automate depositing by interacting with publishers directly, not all faculty have a final copy or correct version, so they would like to be able to access those materials and deposit them into the repository on behalf of authors. They would also like to integrate Symplectic Elements with faculty profiles and campus tracking systems.

Automate it! Open Access compliance as a by-product of better workflows

Next to present was Torsten Reimer (@torstenreimer) from Imperial College London, who discussed the idea of achieving Open Access compliance as a by-product of better workflows. Torsten started by explaining the main difference between the situations in the US and the situation in the UK, namely that there is stronger pressure from funder mandates in the UK. Indeed, although Imperial has had an Open Access policy since 2012, in effect funder policies have overtaken Imperial’s own policy.

Torsten described his brief from the university as: “give us a one-click solution”, the point being that academics wanted any systems or workflows to be as automated and smooth as possible. The publications tracking workflow at Imperial uses Symplectic Elements as its central hub. It receives data feeds from Scopus, Web of Science, ArXiv, PubMed, as well as Imperial’s own HR system and grant system. The system then outputs notifications to academics and enables them to deposit into the university repository, as well as linking with staff web pages.

The publications tracking workflow at Imperial uses Symplectic Elements as its central hub. It receives data feeds from Scopus, Web of Science, ArXiv, PubMed, as well as Imperial’s own HR system and grant system.

This system does, however, have some flaws – authorship recognition is not completely reliable as human names are not the best identifiers. Publisher metadata, for example, the date of publication, is not always accurate or complete. The ability to track non-traditional outputs like data and software is limited. There is no ability to automate the gathering of manuscripts and metadata from publishers. There are also issues with the sharing of data and inter-operating with other universities’ repositories.

Imperial’s new more simplified workflow, combines green and gold workflows. Under this new system, academics are only required to fill out five fields upon acceptance of their manuscript . APC requests are integrated into the workflow via a cloud-based APC management system. Torsten explained that the new workflow has been extremely successful, Imperial have had an 18-fold increase in the number of deposits to their repository since 2012, with 1,300 deposits last month. They’re also processing more APCs and have seen increased engagement with ORCID.

In Torsten’s view, the workflow could still be improved, he would like to move towards an automated “on acceptance” workflow employing CrossRef, ORCID and the Jisc Publications Router. The aim would be a workflow in which the author interacts only once with each output, ideally on acceptance, where all the information required to understand an output is embedded in the metadata, from acceptance, and then further enhanced through the workflow via the automatic sharing of data. Torsten argued that this would make scholarly communication more efficient for academics, for the institution, for funders, and for publishers. It would also make Open Access compliance a by-product of the workflow.

Rights retention Open Access policies

Providing a different perspective, Peter Suber (@petersuber) spoke about rights retention policies. Instead of focussing on deposit workflows, Peter focussed on the policy aspects of infrastructure, arguing that the right policies can eliminate certain hurdles in the workflow.

He asked us to suppose we have an Open Access repository, a deposit policy, a stream of deposits, but no rights retention. What are the available choices in such a scenario? We could not deposit into the repository, we could deposit but “dark”, so not Open Access, or we could infringe copyright. These are not great choices! In Peter’s words, we have a permission problem, we lack the permission to make a work Open Access.

Seeking permission can be time-consuming, difficult, inconsistent from author to author, and typically unsuccessful. If you ask faculty to seek permission, many of them won’t bother, if you ask your dedicated staff, firstly you’ll need dedicated staff, and secondly they will often still fail as publishers aren’t always willing to grant permission.

Suppose we have an Open Access repository, a deposit policy, a stream of deposits, but no rights retention.

However, Peter argued that the problem can also be solved if key rights are never granted to the publisher, but instead retained by the author or the institution. This is the basis of a rights retention Open Access policy. A rights retention policy assures permission, without depending on author initiative or publisher decisions, without variability from author to author, without uncertainty or delay. In Peter’s view all Open Access policies, both institutional and funder, should function in this way.

Peter argued that there are many benefits for authors, they have permission for Open Access distribution without negotiation, they have the freedom to publish in the journals of their choice, they have the freedom to reuse their work, in summary, they have more rights than they’d ever have with a traditional contract.