Rethinking the journal article

In 2009 Elsevier’s Cell Press division began to push the boundaries of the journal concept when it announced its ‘article of the future’ approach on the journal Cell. This approach was based on the realisation that, although the way that journal articles are delivered has changed dramatically with the advent of the internet and e-journals, the actual format of the articles has remained the same for hundreds of years.

The Cell team set out to change this by introducing a range of different ways to navigate around and use articles. One of the key changes was to introduce a ‘graphical abstract’, where the traditional abstract also includes a diagram from the article, illustrating a key finding from the research. Articles also began to include ‘research highlights’, a summary of the key points of the research. Another development was to structure the article so that different sections of the paper – such as experimental details and references – could be reached by a series of tabs across the top of the paper. This structure also enables supplementary information and data to be included in a more meaningful way alongside the parts of the article it relates to, as well as the use of semantic tagging.

The experiment has been well-received and Elsevier is now beginning to roll out this approach across other titles. Graphical abstracts and research highlights can already be included in the ScienceDirect platform and there will be some further changes during the course of 2011, according to IJsbrand Jan Aalbersberg, Elsevier’s vice president of content innovation, who spoke about the initiative at Online Information. ‘Adding value to the online environment makes users want to read the article online,’ he observed.

However, he said that the extension of the concept is more complicated than simply replicating the life-sciences approach. ‘Articles are discipline specific,’ he explained. ‘Graphical abstracts are more appropriate for chemistry and life sciences. It doesn’t really make sense for maths.’

The company is now looking at other disciplines, what is applicable to them and how to cater for all this on the platform. ‘There are a lot of things that are cross-disciplinary and we will support functionality from each discipline. For example, we are working with maths but maths is also in lots of engineering.’

Understanding the different approaches and expectations of different disciplines makes a significant difference in how willing authors are to work to the new requirements. In life sciences and chemistry, for example, authors are already used to providing diagrams or chemical structures with their text so it is a simple matter to pick one of these to be in the abstract. ‘We get push back from authors when we ask for too much,’ commented Aalbersberg. ‘Much of this comes down to the publisher finding the spot where it’s the least effort for the author. Of course there are some spots where a lot of effort is required so we can’t make it mandatory but we do see a ripple effect as authors start to do these extra things.’

A role for data

Data also plays an important role in the company’s plans for rethinking journal articles. ‘Research is much more about research data now,’ he said. ‘Much of it is data that we don’t have ourselves so we go out to something like PANGAEA or the Protein Data Bank. We really want to collaborate more with data repositories. We tell them: “we have the platform, you want to be seen, use our platform”. We believe that we need a really tight connection between research papers and data. We are still thinking about how to make supplementary data available to applications.’

Data adds to the challenges that the article of the future approach is already facing in terms of preservation. ‘On a very basic level you could see video as data – and we see that video from 10 years ago might not be readable. The same is true with experimental data. Data should best be kept stored in discipline-specific repositories,’ observed Aalbersberg. ‘It is expensive as well. The quantities of astronomical data, for example, are astronomical. And there is so much data that could be collected. Scientists could miss something today that might be possible to monitor in the future.’

Preserving ‘articles of the future’ could be a challenge too, given their different structure and semantic links to things that might change. ‘If you have an article that is more built on interactivity with more functionality then preservation is going to be more challenge,’ Aalbersberg commented.

For this reason he anticipates that, for the foreseeable future, the article of the future would exist alongside a more traditional version for preservation. And this will be good news for the significant numbers of researchers who still prefer to read papers from printed versions of PDFs.