The Open Journal Systems (OJS) is used by thousands worldwide to manage their academic journals through the web. For the past two months Takuto Kojima from the Fidus Writer team and Alec Smecher from the OJS have worked on integrating the two. In the interview below they tell us what this means, how far they came and where this could go.
Alec Smecher is the technical architect for the Public Knowledge Project, and lead developer of Open Journal Systems (OJS), Open Conference Systems (OCS), Open Harvester Systems (OHS), Open Monograph Press (OMP), and the PKP Web Application Library (WAL).
Alec, more than 11000 journals have installed your Open Journal Systems (OJS) for the website of their journal. What is so special about OJS that journals choose your software over other website administration systems such as WordPress or Joomla?
Alec: The number of installations we’ve been aware of is significantly more than 11,000, but a more relevant number is the 6,000 OJS journals that we’ve determined host active scholarly content. Given that the number of peer-reviewed, active journals worldwide is estimated in the low tens of thousands (http://pages.cmns.sfu.ca/heather-morrison/appendix-c-how-many-active-scholarly-peer-reviewed-journals/ ), this is a significant number. OJS has a lot in common with tools like WordPress in that it permits the publication of content online for readers on the Web. Where OJS differs is that it provides a complete workflow for academic publishing; this is the bulk of the software even though readers won’t be aware it’s there. It permits authors to submit articles for consideration, guides the editors through the process of selecting and assigning peer reviewers, collecting their feedback, and assisting the author in correcting any flaws. Then it shepherds the article through copyediting and proofreading processes, resulting in an academically rigorous journal article with complete records documenting its status. Once the article is published, OJS provides additional tools and services like OAI-PMH, data feeds for Google Scholar and other services, and archival options like LOCKSS. Finally, it assists in gathering statistics that are both web-oriented and academic-oriented. We have been very fortunate to receive grassroots uptake via word of mouth; we are a small team and have tended not to advertise. Users have largely come to us via recommendations from fellow scholars. Between that and our philosophical alignment with the Open Access movement, I hope people consider us to share the values behind scholarly activity, in addition to supporting the practical elements particular to academic publishing such as peer review. We have also had remarkable success in the developing world, where the impact of the Open Access movement has been most obvious. Often Journal Managers are learning the process at the same time as they are learning the tools, and having a system like OJS that guides the user into a standard workflow has been useful. Alec, the Public Knowledge Project (PKP) has been working on OJS for more than a decade. It is made to organize the creation of journals. However, you have not included a way to edit the journal submissions. Why is that? Alec: When OJS began there was no practical alternative to the desktop word processor. Web-based alternatives didn’t exist, and scholarly authors have typically been hesitant to change their working environment. Web-based and cloud-based tools are now much more familiar to users in general, and at PKP we’ve been watching their ascendancy with anticipation but also some concern over where the data lives, who owns it, and how accessible it will be in 10 or 20 years. Academic journal publishing is a tremendously varied endeavour, and while our normal focus is the 80% mark — that is, our software attempts to provide good functionality for 80% of users — the remaining 20% is something we watch closely as the future of publishing will emerge from experiments there. At any time we have half a dozen experimental or research-oriented side projects around subjects like statistics, XML, and annotation; until now editing journal submissions has fallen into the wish list within that 20%. I’ve been periodically surveying the options for web-based content editing, looking for options that are technically viable, offer sufficient benefits to users (e.g. in facilitating XML generation), and avoid some of the major red flags around content ownership, archival options, and software licensing. Fidus Writer is the only viable option I’ve encountered thus far.
Takuto Kojima is making his living in Berlin as a freelance web developer.
He is one of the co-founders and developers of Fidus Writer.
Takuto, you are one of the developers behind Fidus Writer, an online editor. What makes Fidus Writer different from other editors? Takuto: Fidus Writer is an open source word processor that runs on modern web browsers with special features focused on academic writing. Fidus Writer makes it much easier to create and edit academic documents written according to specific academic formats. The idea is that authors should be able to write academic documents without being IT specialists, and without knowledge of document formatting rules such as citation styles. Until now, you had to study those academic formatting rules in order to correctly format an academic document. This can be time consuming and error prone. Additionally, in order to be able to edit the document according to a given format, you would need to know the advanced options of at least one text editor. Fidus Writer will free researchers from the overhead of considering the rules and formatting the document manually. Researchers just have to write about their research using the intuitive Fidus Writer interface, and Fidus Writer takes care of the document formatting including citation styles. Fidus Writer also has a integrated bibliography manager which this works seamlessly with citing on the document editor. Fidus Writer is special not only because of those features for academic writing, but also due to its modern functionalities as a document editor. Researchers can edit a document collaboratively. They can even edit the same document at the same time. They can export the document into HTML, Epub, Latex or PDF.
I understand you have been working on a way of allowing Fidus Writer and OJS to integrate. Alec is based in Vancouver while Takuto lives is in Berlin. How did you meet and how did you get the idea to try to put the two projects together? Takuto: We spent several months last fall presenting Fidus Writer at various conferences and trade shows in Europe. Those most interested were university based publishers who either were using OJS already or were considering moving their journal administration to OJS. OJS just seemed to be a household name among anyone working with high quality open access journals in Europe, so we thought it would be a good idea to contact PKP to see if they were interested in working with us on some way to add Fidus Writer into their editing process. Alec was quick to write back and I was pleased to hear that the OJS team was willing to let him work with us to see whether this could be done almost immediately.
How will integration work? Where will the author start when submitting a document to such a unified system? Takuto: Currently OJS supports Microsoft Word (or Libre Office/Open Office) for the peer review process. The author uploads his document as MS Word document and the reviewers review it in Ms Word after downloading it to their local machine. Ultimately, we would like to replace Ms Word with Fidus Writer in this process. The author can write (or paste) the document into Fidus Writer and submit it for peer view in OJS; reviewers will review it in Fidus Writer, all online, without ever having to download the document or manage files locally. We believe that Fidus Writer brings a lot of benefits to the author with its special features for academic writing. And reviewers won’t be obliged to have Ms Word on their computers. They will be able to review the document from any computer with a modern web browser.
A Fidus Writer document will be stored in a completely semantic form and this offers further possibilities of advanced sorting and searching of documents later on the front end of the journal.
Alec: This vision of integration dovetails with some of PKP’s broader strategies. It’s hard for OJS to work directly with content written in a word processor because it’s very difficult to determine what part of the document is the abstract, what is the bibliography, and so on, down to a fine level of detail such as references, figures, etc. Working with a system like Fidus Writer that introduces semantic knowledge at the beginning of the process means we can ease the editor’s workload throughout their workflow, from ensuring that submissions are anonymous before they’re sent to reviewers, to quality checking citations, to indexing content in external services such as PubMed Central. These processes are typically done manually and editors don’t have enough time to see them all through. On the flip side, authors will not need to understand the difference between symantic formats and layout formats — they will simply work with a tool that looks like an enriched word processor and the benefits will be automatic. What is the current state of your efforts? What do you need in order to make it ready for actual usage?
Takuto: Just a few days ago, we’ve completed the first rough integration of OJS and Fidus Writer in the peer-viewing process. The integration is currently at the stage of a prototype and not ready for production use. Basically the author can submit a Fidus Writer document to OJS and revise it based on editor feedback. But with the current integration the author may submit the Fidus Writer document created only by the Fidus Writer installation registered with OJS. In this version of Fidus Writer some valuable features like collaborative editing are deactivated.
Next we would like to secure funding to be able to dedicate some time to turn the prototype into something that can be used in production. We are interested in working with institutions or organizations who could make use of this feature. Of course we are also always interested in development contributions.
Alec: This first stage is a technical proof of concept and a chance for our teams to become familiar with each others’ technologies and working habits. This has been roundly successful. Next we’ll need to start talking about workflow decisions, such as how to integrate disruptive opportunities like simultaneous collaboration while maintaining sufficient content control for an academic workflow. These decisions will need to involve actual journals, leading to a second stage likely involving live content. When you are done, is the idea that it’s obligatory to use Fidus Writer and OJS together?
Takuto: For me it would be a dream come true if OJS could always run with Fidus Writer. But to be realistic, I think support for Ms Word will remain common for now, unless, of course, we implement the import filter for Ms Word document into Fidus Writer. Fidus Writer will continue to exist as a stand-alone product.
Alec: We will probably always continue to support our traditional upload/download process — but adding support for a web-based tool like FidusWriter will enrich the journal publishing ecosystem, and for journals that are ready to use it, will offer tremendous opportunities for further automation. Imagining the web 5 or 10 years from now, I suspect the distinction between local applications and web-based ones will be less obvious than it is now. Users may not even be aware which they are using.
So the goal is to make more organic the submit-review-revise aspect of the process, which is absolutely needed. But what about the rest of the process: publishing, and post-publication comments, etc., which remains a major week point of OJS (or maybe OJS journals)?
E.g. I think the world needs something with the UI elegance and power of PeerJ, but open source. How do we get there?
The way the data is stored in Fidus Writer makes it possible to easily create different outputs. We have created an epub engine and a PDF-layout-engine, open source, which are part of Fidus writer for this purpose. As for post publication comments, I would think that in the case of many sciences what makes most sense is to add a “publish directly to WordPress” option. WordPress sites are easy to install and maintain for nontechs and even smaller institutions will be able to operate one. There are plenty of commenting options available through WordPress, also ona per paragraph basis the way PeerJ has it. For other purposes, we will need one or several XML-export filters. The German national library just published on XML-standard which they want used for longterm storage. And then there is also the JATS standard which people in some sciences in other countries will want. Those can then be read and displayed by any tools such as what PeerJ has running.