The scholarly plumbers
Sian Harris reports from STM’s e-Production seminar on the challenges of editorial systems
Anyone new to the industry and thinking of launching an electronic academic journal would be well-advised – although possibly somewhat alarmed - to take a look at some of the presentations from the STM Association’s e-Production seminar held in London in early December.
The discussions at the meeting revealed that the processes involved in running an online journal – and the technology behind them - are considerably more complex than they might appear to authors and readers of the journal.
Richard Wynne, VP sales and marketing at Aries Systems illustrated this point in his talk by likening editorial systems to plumbing. A bathroom may look simple and luxurious, he observed, but behind the walls and under the floor the pipework is much more complicated and functional. ‘Ideally, plumbing is reliable, invisible and provides some luxury,’ he commented but explained that ‘in reality it’s complicated, messy and needs continuous maintenance and upgrading. If it goes wrong it costs much more to fix than simply making the repair.’
The reason for the plumbing description was that many of the same points can be made about scholarly publishing workflow systems.
He went on to describe one component of the ‘plumbing’: the reviewer form. To the reviewer, it might look like a simple web page to enter some comments for the author and then submit them.
In reality, he pointed out, it’s much more complicated. He pointed out that, as well as comments to the author, there need to be separate confidential comments to the editor and special characters need to be supported so that reviewers can copy and paste from software like Word. The system also needs to enable reviewers to upload Word or PDF files and strip out properties from those files so that the reviewer is not accidentally identified to the author. The reviewer needs to be able to proof and print everything they put into the system and keep a copy. Because they are often interrupted, the reviewer needs to be able to save their work and carry on later.
In addition, reviewer comments need to be anonymised, concatenated, re-opened, merged into emails, edited before sending to authors and shared. There are also plenty of elements needed to simplify the job for the reviewer and standardise the responses. These include yes/no questions, follow-on questions and things like radio buttons, check buttons, drop-down lists, default choices and the option to allow author access to responses.
This then this all needs reporting on, and reports on review process to be distributed to staff on a regular basis. What’s more, every journal has different terms and instructions have to be varied depending on the article type and what type of reviewer it is.
And, there’s a further challenge that Wynne explained: ‘The reviewer forms of today aren’t the forms of tomorrow. Things that weren’t important are now becoming more important.’
And review systems are just one part of the ‘plumbing’ required for digital publishing. There was also much discussion about submission systems and the challenges posed by traditional ways of submitting papers.
Kaveh Bazargan, director of River Valley, told delegates at the meeting why he dislikes the use of word processors in article submission. 'A word processor is good if the document is the final product but the direct analogy of a word processor is the food processor. You can use it to make a fruit smoothie but the problem is that you can’t put the fruit back together again,’ he said.
His argument was that publishers require full XML and XML-first workflows, which, he said, is equivalent to wanting the ingredients back after they have gone through a food processor. Much of the role of typesetting firms like River Valley, he said, is writing filters and creating bolt-ons to add back in the structure that was word processed out of content.
‘Authors think they are helping publishers but they are hindering them,’ he noted. ‘It would be better to catch content before it is processed and avoid page-based writing systems.’
One solution that he sees as good is using LaTeX. This, he said is already structured and does not process word. However, not everyone will use LaTeX so he suggested creating forms for authors to fill in, more like writing a blog. It needs to be user friendly and export XML. One option that he recommended was to use WordPress. ‘The author feels happier because filling in forms is easier - and they are structuring the data,’ he noted. What’s more, it enables copy editing and tracking of that process, it is all in one place in the cloud, and WordPress already has Mendeley plug-ins.
‘Help put us out of business,’ he argued. ‘We make our living out of cleaning up Word files. We should be spending our time doing something more efficient than reverse engineering.’
Ian Potter of Thomson Reuters’ Scholar One also spoke about the benefits of using electronic forms in article submissions. 'Systems are being asked to do more and they more often touch other systems,' he noted. ‘Electronic forms sit in the workflow and form filling can be tracked.’
A common theme running through the day was a need for balance between the needs of authors and publishers. As Potter noted, ‘there needs to be a balance between benefits for publishers and overtaxing authors. Electronic forms benefit publishers but also make it easier for authors too.’
In addition, he said that reporting systems enable you to do more things with data – for example, you could gather customer metadata at any step in the submission process and this can all go with a manuscript if it is transferred to another journal. ‘Every question that gets asked goes into XML; nothing is lost,’ he said of Scholar One’s system.
Unbinding content and devices
Structure was a key issue for Bret Freeman, director – new technologies at Aptara too. ‘Books and publishing are really, really changing. Consumers today are totally different animals; they want information not just when and where you want it but also how you want it,’ he observed. ‘If you have structure behind content it doesn't matter what the next device will be; it’s about being agile.’
He noted some of the possibly-unexpected challenges brought in by new devices with much higher-resolution screens. These devices have brought with them the requirement for much bigger files for the same content – and corresponding challenges with bandwidth and storage limitations. The solution, he observed, is to rethink how content is packaged.
'Resolution problems? I see unbundled opportunities solutions,' he said, referring to the idea of selling content in smaller chunks. 'The sooner we stop thinking of content as words on page the better; digital can do so much more.'