OpenAIRE Position Paper on Open Research Europe

 

OpenAIRE Position Paper on Open Research Europe

The European Commission recently announced plans to create “Open Research Europe” (ORE), an online platform allowing rapid, Open Access (OA) publication of Horizon 2020 related peer reviewed articles and preprints. The platform aims to be a fast, cost-effective high-quality service, with mechanisms for open review and alternative metrics. It will be a free, complimentary (i.e., non-compulsory) service for H2020 beneficiaries. In developing such a service, the EC will join a growing list of funders (e.g., Wellcome Trust, Gates Foundation, Chan Zuckerberg Initiative) who offer their researchers a direct, low-cost route to OA publication. Given this is the first initiative of its kind, from a large public funder, OpenAIRE would like to take the chance to make public its point of view.

OpenAIRE receives the initiative with great interest, and appreciates that a major funder recognises that scholarly publishing is a vital and integral part of the research lifecycle. The EC has long recognised the need to complement policy actions with investments in infrastructure to support OA implementation, through initiatives like OpenAIRE. ORE has the potential  to extend further these activities by enabling  a well-recognised, user friendly platform to publish researchers’ output in a timely and cost-effective manner. At the same time, however, it should be trusted, community led, open and transparent.

OpenAIRE presents the following considerations as the European Commission plans for Open Research Europe:

  1. Any platform should be openly and independently governed

In order to ensure long-term commitment and trust, independence of governance is crucial for the success of ORE. Any platform should not be put into the hands of one party and should be clearly and transparently managed. The service must be sustainable for the long-term and integrate transparency of revenue-management and editorial decision-making This is especially so because of the potential for the appearance of conflicts of interest for a funding organisation to directly support a platform which disseminates its research. The EC has advised it does not wish to become a publisher – this is the right approach. But the platform provider should also not take sole ownership of creating governance structures (see point 3 below on the need to avoid lock-in). Hence, the architecture of this management structure must be carefully explored from the onset (e.g. membership organisation, not for  profit, centralised, federated). Broadly speaking, a wide community of experts should govern all the aspects of the platform, from editorial boards to technical roll-out. This managed consensual activity would have oversight of several important areas, including: the ownership of publishing process assets (databases, coding); overview of transparent workflows between authors and editors; trustworthy terms and conditions for sharing and access of articles; ownership of data; decisions on budget and management of funds.

  1. Community consensus is essential

A horizontal publishing platform to potentially support all of EC-funded research will be extensive, serving researchers working across all disciplines. Community buy-in, especially from Member States and major Research Infrastructures, will be essential but will also take time to fully achieve. In terms of the scale of funded-research across disciplines, its scope will hence be larger than any existing funder publishing enterprise. Therefore a process of requirements gathering and building should be conducted that includes the perspectives of a broad range of experts. Issues that will need close oversight at all disciplinary levels will include: choice of editorial boards, publishing workflows, designing a modern, open and shared reviewing process for publications. In order to achieve seamless or aligned processes in the European Research Area , any effort by the EC must have the support of the Member States and national initiatives. There is also a need to avoid a “one-size-fits-all” approach and to respect the needs of particular research communities while also encouraging maximum transparency within processes. Community consensus will be especially important in convening boards to oversee peer review. These decisions should ideally be taken by a set of stakeholders chosen by the community and reflecting the vast range of scientific community interests. The platform should also consider how it interacts with new reward systems for researchers, for example being recognised for promoting sharing research results openly, and providing open peer review to pre-prints. This in turn could incentivise their use of the platform,  using alternative impact factors to bring into play when being evaluated.

  1. Avoid vendor lock-in

ORE should make use of the best available technologies, whether in the private or public sector. However, it is crucial that the ORE platform be organised such that it does not become bound to one specific organisation for its technologies or workflows. Although ideally any platform should be open-source, at the very least publishing workflows should be transparent and re-implementable on another platform. One way to facilitate this would be to bestow custodianship of the technological platform upon a different body than that which governs the editorial bodies and publishing workflows. In any case, ORE must not become bound to any one platform or organisation such that the cost of transferring to another platform/organisation becomes prohibitive.

  1. Ensure FAIRness and reproducibility

The platform should ensure that all content is openly licensed, and that the platform embraces the totality of research objects (including data, software, research protocols), including links between them, to ensure maximal reproducibility and transparency. All data and publications should bespeak the FAIR principles (findable, accessible, interoperable and reusable), including open licenses and open standardised metadata to enable seamless access and re-use by humans and machines.  

 

OpenAIRE’s well-established technical infrastructure can support in the following ways:

The OpenAIRE service infrastructure has been devised to support, facilitate, and monitor scholarly communication workflows for Open Science. Its services collect metadata records from scholarly communication providers (e.g. literature and data repositories, CRIS systems, publishers, funder registries, author registries, data sources registries, etc.) and end-users (e.g. scientists, project officers, Principle Investigators) to populate a graph of interlinked scholarly objects, i.e. literature, datasets, funders, projects, organizations, authors, and data providers. By integrating all these resources, OpenAIRE builds a number of so-called products – i.e. services accessible with user interfaces from the web – aimed at specific scholarly communication stakeholders (e.g. researchers, funders, research infrastructures, project officers, managers of data providers, etc.) and exposes open APIs (http://api.openaire.eu) to serve third-party services (e.g. CORDIS portal, Elsevier’s Scopus, institutional repositories, etc.).

OpenAIRE services could support ORE as follows:

Interoperability and data exchange: OpenAIRE can support ORE in reaching compliance with OpenAIRE guidelines in order to facilitate distribution of metadata according to global standards. ORE could therefore establish bilateral data exchange with OpenAIRE. This would enable:

  • ORE to become a data provider of OpenAIRE: OpenAIRE to incrementally collect the graph of scholarly objects maintained by ORE;
  • ORE to collect all possible enrichments regarding its objects and gathered by OpenAIRE, e.g. links to grants, links to other objects, etc.
  • ORE to collect/fetch research grant information from all over Europe and beyond available in the OpenAIRE arena. Such interaction would enable ORE users to upload metadata tagged with proper funding information.  
  • Data exchange to be implemented in bulk, via standard protocols such as OAI-PMH and ResourceSync (ongoing implementation), or dynamically via the OpenAIRE broker service.

 

Software and Service support: OpenAIRE can provide out of the box solutions for ORE, reducing the overall development and maintenance costs. The following services may be used to store, fetch, or distribute objects (metadata and payloads) and can be provided “as a service” (OpenAIRE operates the service) or as Open Source software to be shared and co-developed:

  • Zenodo.org: Zenodo is the catch-all repository service provided by OpenAIRE; it is maintained and operated at CERN on top of the local storage infrastructure; the repository is designed for researchers to deposit publications, datasets, software, and research packages (ongoing development) and links to other objects; it supports DOI minting (from DataCite), offers preservation facilities, and provides community support. It is fully OpenSource (https://github.com/zenodo/zenodo), has an open roadmap, and is non-profit.
  • Literature Broker Service: the service realizes the distribution of relevant metadata information between different sources based on subscription and notification protocols, e.g. a publication in ORE may be sent to the repositories of reference of the relative authors, links to/from ORE objects identified in OpenAIRE (e.g. collected from publication/data repositories or inferred) can be sent to ORE to enrich its graph; the ORE could therefore be bridged to repositories world-wide and vice-versa via the service;
  • ResourceSync Server/Client Services (ongoing implementation): will be used in OpenAIRE to export the OpenAIRE scholarly information overview in an incremental and efficient way, but the software is general purpose as it follows the generic specifications of the protocol; the same holds for the client
  • Information Inference Services: such services are capable of running a set of TDM algorithms on top of a collection of PDFs or metadata records in parallel (based on  HDFS)
  • Data Literature Interlinking Service: the service enables an openly-accessible graph of links between literature and datasets as collected from data centres and publishers; consumers can resolve links to DOIs on the fly or collect sets of links for local processing
  • Usage Statistics: the service tracks usage events from OpenAIRE Content Providers and generates consolidated, standards based COUNTER compliant usage statistics
  • Community support at the local level: ORE will require a pan-European outreach operation to assist users in understanding and embracing this innovative platform. OpenAIRE’s National Open Access Desks, present in every EU country and beyond, can assist with this support.

References

Bilder G, Lin J, Neylon C (2015) Principles for Open Scholarly Infrastructure-v1, retrieved [date], http://dx.doi.org/10.6084/m9.figshare.1314859  

Shashok; K (2017) Can Scientists and their institutions become their own open access publishers? https://arxiv.org/abs/1701.02461

European Commission (2017) Evaluation of Research Careers fully acknowledging Open Science Practices.Rewards, incentives and/or recognition for researchers practicing Open Science https://cdn1.euraxess.org/sites/default/files/policy_library/os-rewards-wgreport-final_integrated_0.pdf

 

Najla Rettberg

OpenAIRE2020 scientific manager. University of Göttingen, Germany.

More Posts - Website - Twitter

Top