I’m happy to be in the Editor in Chief of the Journal of Big Data, a new peer-reviewed publication offering an integrative approach to the subject of big data.
As a discipline, big data is broad. It brings in diverse topics from areas such as database design, machine learning, statistics, visualization, social media, policy, privacy and ethics. All these things come together to underpin the work being done by today’s data-centric organizations.
Hitherto, all these areas have their own journals and conferences, and different academic communities. By bringing the areas together, I hope to foster dialogue between the various technical and user communities. For example, issues such as fine-grained user control over privacy will require engineering at the database level to efficiently accomplish.
Secondly, in big data there is a short cycle from R&D to production. In fact, many of the influential papers in the field, such as Google’s MapReduce paper, were published after the work described had been running in production for some time. This underlines that a successful forum for the area must draw together industrial and research contributions, and take a pragmatic hands-on approach. Method, professional and industrial experience are as important as results.
A similar model in my mind as I edit the journal is the Communications of the ACM, which melds experience with research articles.
In addition to philosophical reasons, I believe open access publication is important for this journal in order to:
- ensuring widest possible distribution — to be integrative over industry, we must recognize that many readers won’t have subscriptions to journals as academic institutions do: advance in big data is coming from all quarters
- ensuring broad & timely publication — any journal in computing has to compete with the conference-paper route taken by many academics. We will publish and make papers available online as soon as they are accepted. For an integrative approach, this is superior to the siloed model that conferences facilitate.
Additionally, Big Data does not impose article processing fees on the author, though obviously it costs the journal to peer review, edit and prepare articles for publication.
Instead, the journal publishers intend to make money in two ways:
- in addition to peer-reviewed content, we are publishing additional content of high quality and interest to readers, which is only available via subscription. So subscribers help underwrite the open access publishing costs and are rewarded with exclusive content;
- secondly, as the journal, its contents, and much of the research work in big data is valuable to commerce, we are recruiting sponsorship from companies to support the journal, which will naturally garner them exposure to the readership. Many big data companies have benefited from the free publication of research and development of open source software, I hope they will take a lead in supporting its continuation and the existence of a viable integrated forum for big data.
How you can contribute
On behalf of myself and the Big Data editorial board, I invite you to join the wider dialogue on big data.
If you have an original article you wish to submit to the peer review process for publication, please refer to the submission guidelines.
If you wish to discuss content for the subscriber magazine side of the journal, feel free to email me to discuss.
Also, please get in touch with me or the publisher if you want to sponsor the journal or commission custom content.
The first issue of the Journal will be published mid-February, in time for Strata 2013. In the meantime, please enjoy the free preview issue, available for download from the journal’s home page.