Improving access to scholarly resources, from anywhere, on any device: three considerations
You may remember that the drinks brand Martini ran an advertising campaign with the strapline "anytime, any place, anywhere". Back in the 1990's early adopting digital businesses and the businesses that served them used this famous line to describe their ambitions for the delivery of their media. Back in those days technology had not quite caught up with the aspiration - as anyone who remembers wrestling with 2G data connections will testify. The 2018 landscape has evolved significantly:
- In the UK in 2017, 78% of adults had used the internet 'on the go' (that is, away from home or work) using a mobile or smartphone, portable computer or handheld device. If we narrow the demographic to only look at Generation Z and younger Millennials (i.e. ages 16 to 24) the figure rises to 98%
- In the US in 2017, 73% of the population 'actively browsed' the internet via mobile
Two tipping points should be noted. The first, in April 2015 was when Google's search results for mobile users started to prioritize sites optimized for mobile devices, impacting significantly on search results. The second, (mentioned in our previous post) is the 2017 data that shows that Google Web Search traffic is 50% mobile yet Google Scholar is only 10% mobile - and Google's finding is that mobile traffic is additive to traditional web search traffic.
The statistics quoted above (and numerous others) tend toward the same global story: the rise and rise of mobile internet access. Faced with such data, what are the considerations when improving access to scholarly content on any device at anytime?
Here are three considerations.
To improve access on any device, anywhere, the first thing to be achieved is to optimize content for the device in question. The original solution to this was to run browser detection scripts that worked out the browser type, device and screen size and displayed content and functionality accordingly. The problem with this was that each screen format needed to be hard coded, making it an expensive, time-consuming approach that did not scale well. Responsive Design changed all that in that "Responsive Web Design allows you to use a single URL structure for a site, thereby removing the need for separate mobile, tablet, desktop, etc. sites". Embracing responsive, enabled the very cost-effective provision of a device-specific experience. Because responsive is based on breakpoints, solutions developed in this way were forward-compatible. Assuming the content is in HTML, responsive solves the delivery problem. But we also need to be aware of and plan for different format solutions as new technologies start to exert influence in the market. Potential future use cases of content access via watches, glasses and voice-operated input/output devices such as Alexa and Home will present us with new and exciting challenges.
Finally, we must be mindful that in our sector content (especially back catalogue) may remain in PDF format. PDFs tend towards a poor experience on a small screen because their format is 'locked' for large screens. There are workarounds such as Calibre and the insertion of reflow tags into PDFs, but these are not supported widely. Prof. Andrew Prescott (University of Glasgow) summed up the issue in saying that "My dystopian view of the future is that journals will consist of pdfs of journal articles".
2. Access and Authentication
The still numerous issues concerning access and authentication mostly stem from the legacy model of IP-based authentication within institutions and the various patches and workarounds applied to this model over time. There is a consensus that, within academic libraries at least, some principal actors seem to continue to underestimate (or are unaware of) issues of access caused by the distribution of services and the delivery of services onto multiple devices. The sometimes, labyrinthine user journey that ensues has been described in parody and is the subject of a plea to the actors within academic libraries to try out the poor user experience for themselves. Of course, the underlying issue is serious as those seeking frictionless access may be driven to platforms such as Sci-Hub to avoid the need for payment and/or authentication altogether.
The fact that traditional measure of 'abandonment' (the turnaway) does not measure those who did not attempt to access content via 'traditional models' in the first place creates uncertainty as to the scale of the issue. This lends credence to the view that many institutions are simply failing to deliver when faced with trends towards mobile and/or off-campus access. However, numerous approaches and interventions now seek to reduce the friction of off-campus access:
- Our significant intervention in this area is Campus-Activated Subscriber Access (CASA). Working in cooperation with Google Scholar, we developed CASA to enable seamless off-campus and mobile access to subscribed scholarly content. With CASA, off-campus and mobile users may access their library's subscriptions just as easily as when they are on campus.
- RA21 (a joint initiative of STM and NISO) also seeks to "move beyond IP-recognition as the primary authentication system". This initiative will not build a platform or solution but defines and supports best practice and provides guiding principles for frictionless access not based on IP recognition.
- Unpaywall claims to be the most extensive database of legal Gold, Green and Bronze open access research. Unpaywall is a significant intervention and harvests papers from open repositories, OA journals, hybrid journals, DOAJ and Crossref. Unpaywall claims that an OA version of a paper is found for about 30% of queries.
- Preprints. As preprints are by definition 'open' the year on year increase in their publication and usage is a useful barometer not only of changing definitions in the workflow but also of changing models of access and authentication. Data from arXiv and bioRxiv is therefore revealing.
Driven by increasing support from funding bodies and institutions this rapid growth in preprints will continue, enabling more frictionless access to the research.
- SAMS Sigma, another of our interventions in this part of the ecosystem, provides users with a single sign-on to access scholarly content across multiple websites. It enables identification of each user to provide a flexible way to license content and control access. Users only have to sign on once to access content, on any device, anywhere. SAMS Sigma, therefore, provides solutions to many of the complexities of institutional access.
3. Needs and Behaviors
Data that shows year on year rises in mobile access does not tell a nuanced story. In the face of such data should we assume that any and all digital content be equally accessible anywhere and at anytime on any device?
To answer this more research is required to understand what, when, and why content is accessed. More work also needs to be done to understand and meet user needs with regard, particularly regarding mobile search and usage. As we have described, our 'Quick Abstracts' project in association with Google seeks to enhance the mobile search experience on Scholar.
Of interest is data from eMarketer which shows a considerable skew of time to native (as opposed to mobile) applications in the US. It seems to us that the long-running debate over the merits of native vs mobile applications can now be closed. The sheer volume and continued growth of mobile internet access demonstrates that publishers and other content providers have at the very least to enable optimised access to content on mobile devices (i.e. mobile applications) and then consider if a native application is an adjunct relevant to their content and market. This issue is an open question for publishers and aggregators of the research.
Finally, it seems clear that strategies to improve access should consider the full range and use of channels and devices available to the user in the 21st century. To this end, it may be of interest to borrow the concept of 'dayparting' from the broadcast media. 'Dayparting' divides the day into time slices and considers platform access in those slices. Interesting dayparting data from Comscore shows how useful a concept it is in providing a nuanced view of device use. 'Dayparting' may be a useful tool with which we can start to better identify specific mobile use cases within our workflows. An example of this may be the identification of mobile use to 'mark and gather' content. For instance, the parts of the day (e.g. commuting to work or over lunch) when actors in our ecosystem may be identifying and gathering content (via eTOCS, search results, abstracts, titles etc.) to consume it later that day, on a different device.