Staying open to open science needs

Share this on social media:

Hannah Heckner

When it comes to platforms and the shifting OA landscape there's no one-size-fits-all solution, writes Hannah Heckner

One of the challenges inherent in developing a content hosting platform is also one of the challenges that open science frequently finds itself faced with: there is no one-size-fits-all.

As a platform provider, we strive to design and deliver technology that meets and exceeds industry standards, offering publishers the benefits of scale and shared development while simultaneously building in enough flexibility to accommodate the unique needs of individual publishers. Likewise, open access models are designed to be widely applicable and to address the needs of researchers, funders, publishers, and readers – and yet new models continue to proliferate, each one with an approach that yields approximately the same result (namely, open access to research), while using occasionally different means to arrive there. 

As a technology vendor, this means we have to approach all publisher solutions in a business-model-agnostic manner, ensuring that the platform is able to satisfy those customers who are on the leading edge of open access innovation, those who rely on traditional subscription revenue, and those who manage publications with access models spanning the spectrum. 

By building infrastructure that works with diverse models, technology vendors afford publishers the flexibility to adapt and change with the market. In the last three years, the publishing market has changed as various industry initiatives have gained purchase, OA mandates have proliferated, library budgets have declined further, and the public health crisis pushed publishers to not only think twice about their paywalls, but to also ask more from their hosting platform.  

At Silverchair, our clients are innovators, experimenting with emerging content formats, OA models, and more, all of which rely on technology to be implemented in a way that is seamless for the end-users. This means maintaining (and scaling) technology that is flexible enough to accommodate whatever new models our publishers bring to us, whether it's IWA Publishing with Subscribe to Open, MIT Press with Direct To Open, our society partners piloting smaller scale transformative agreements, or whatever comes next. 

Some of our publishers are aggregators or distributors for a group of societies and publishing partners. These clients are in turn serving a range of societies, from very small to larger independents. And they all have different outlooks and different needs when it comes to open access. 

Some clients have even found that different sorts of open access models are needed depending on the type of content, whether that be arts versus hard sciences, book versus journal, or experimental formats. So, it really does take a variety of models to make this work, which also means that the technical demands vary and are perhaps larger than we'd hope.

Among the most basic platform-level functionality required to support OA are:

  • Rich article metadata;  

  • Connections to upstream/downstream and secondary/tertiary publication objects; 

  • Compliance with FAIR [findable, accessible, interoperable, (machine) readable] principles;  

  • Information about the users of content; 

  • Improving the discovery of OA content; and  

  • Shoring up other revenue streams with the dwindling of subscription revenue 

However, fully supporting open access practices means the development of more nuanced features to support publisher needs and to deliver the best possible user experience, such as:

  • Auto-filtered search, recommendation widgets, and topic collections to ensure that users accessing OA content are only presented with other OA content (eliminating the disruption of paywalls to a seamless OA user experience);

  • Accommodation of various and custom OA license types in the metadata;

  • Robust deposit services to meet OA indexing needs; 

  • Personalised content widgets presenting users with exclusively OA content from across a publisher’s catalogue; 

  • Powerful and flexible advertising options to support revenue needs on OA content; 

  • Additional features and tools for engaging OA users and directing traffic toward other revenue-generating avenues;  

  • More granular author and funding data provided upstream following emerging best practices/standards. This includes multiple layers of institutional affiliation, Ringgold or GRID identifiers, FundRef, ALIs, data availability statements, etc.; and   

  • Supporting the publication of peer review reports following JATS4R recommendations. 

Many of these features have become industry standards over recent years, while many solutions yet remain amorphous. The primary technology needs our publishers are concerned with to address the move to OA center on reporting, persistent identifiers, and access/funding indicators. 

Reporting, analytics, and data have shifted from useful to critical as publishers strive to adequately capture usage, particularly as relates to some of the new library and funder agreements we’re seeing. Communicating the value of OA models, especially those that rely on a collective community action model, is imperative to their success. Reporting not only demonstrates to stakeholders that these models stand on their own, but also prove pivotal in navigating internal bureaucracy around getting funding for open access projects.

However, the nature of 'open' makes it more difficult to capture and report on your stakeholders’ and your library partners' use of content. A prime example was early in the pandemic when publishers globally opened access to content, which presented both opportunities and challenges. The opportunities included increased traffic, while a difficulty was not being able to identify where that traffic was coming from. Publishers addressed this through a variety of methods, ranging from IP identification partnerships to registration walls. 

Addressing OA reporting needs at scale in a way that works for all publishers and models relies on the next technology requirement: persistent identifiers. Persistent identifiers are critical to the open access infrastructure, whether it's the ORCIDs, the DOIs, the fund registry data, or the raw data. All those fields being checked, validated, and required in the process of ingesting content to a hosting platform allows platforms and publishers to then have a reporting workflow that doesn't require significant disambiguation or other remediation. 

Our client community uses a variety of editorial, submission, peer review, and production systems. Ensuring that they have clean data with persistent identifiers coming into the process allows us to then report out on the other side with confidence. Our publishers are approaching these needs in different ways and with different partners, and all the bits of data that flow from those workflows / sources are becoming increasingly important in OA models and mandates. That data then appears in a publisher’s XML and that drives platform functionality, such as the ways we signal to users what content is open or free, what they're entitled to access, and what license that content is published under. 

This brings us to access, license, and funding indicators on the front end of the platform. Though there are certain aspects we’ve become used to (i.e., the locked / unlocked padlock), there’s a divergence of approaches, which means that even the lock icon fails to capture the distinctions between free content, OA content, content you’re entitled to through your institution, temporarily free content, etc. 

The pandemic certainly exacerbated these access and funding indicator needs. Suddenly a lot of our publishers were thinking more about whether users can make sense of little unlocked locks, little locked locks, locks that are red, locks that are green, or even other indicators that say you have access but may need to go somewhere else and enter some other bits of information to get into that content. Carrying this further into platform functionality, there’s a need for users to search and filter by the access types. And publishers are increasingly fielding questions from libraries about discovery and questions from funders about the display of funding information and license type on the front end. 

As a platform, our goal is to work with partners and industry organization on developing and delivering evolving industry standards for these and other open science requirements all while building in flexibility to accommodate new models and needs. It’s unlikely that we’ll arrive at a point in the open movement where there’s a single approach and a single way for platforms to best deliver open science for all its stakeholders, so the best path forward for technology vendors working in this space is ultimately, to remain open to change.

Hannah Heckner is director for product strategy at Silverchair