Last week a few us attended ReCon 2015 here in Edinburgh which proved to be very interesting.

Session 1: Beyond the paper: publishing data, software and more
Attended by Scott Renton

Scott Edmunds from Gigascience works on a platform which attempts to keep the publication of research data up to speed with its mind-boggling rate of production. Now that we’re moving “beyond dead trees” into the digital realm, many problems are having to be tackled, such as meaningful peer review (which won’t result in retractions), skewed incentive systems (reward republication, not advertising; reward executable data) etc. He praised such tools as knitr, which effectively generate papers from data, and pointed out the importance of getting the research out there (citing ebola and climate change as examples where problems could be mitigated in this way).

Arfon Smith from GitHub linked the whole Research Data publication problem to his own speciality: the Git version control system. He is aware that Open is the new norm, that PDFs are unsatisfactory, and that we don’t share in a useful, creditable way. Communities form around a shared challenge/shared data, and this open source model can be utilised by academia, in a similar way to software development.

Stephanie Dawson from scienceopen spoke  about aggregators, stating that these are the drivers of impact. She is very skeptical about any kind of journal level impact rankings (“negotiated, irreproducible, and unsound”), but is far more optimistic about article level tools where usage stats, citations etc. are meaningful.

Peter Burnhill from Edina talked about “link rot” and “content drift”, and suggested that more thought in general needs to be given to the persistence of URLs and content in order that the data remains available. He offered up- with some interesting case studies- some quite depressing statistics about how much of the PMC and Elsevier corpora are “rotten”, but he did suggest that the rates are steadily improving, so there is some good news out there!

Session 2: Digital communities and social networks for researchers by Steve Wheeler (@timbuckteeth)
Attended by Stephanie (Charlie) Farley (@Sfarley_Charlie)

Steve had a lot to say about the importance of open access to education in the digital age. We have the technology but do we have the will to be open? When you publish research you’re educating your community, hoarding knowledge from those who desperately need it is almost perverse. Additionally, there is a payback to making your work open and available. Steve recommended using Google Scholar to compare your own closed vs. open journal Google citations list and see for yourself.

He spoke of how open access works to serve the community of users. Communities are no longer related to or defined by locations, today’s definition of community is about a common interest. Communities celebrate, connect, communicate (online en masse), and in this way everything can be collaborative if we want it to be. Steve encouraged attendees to look up Dave Cormier’s research on the rhizomatic model of learning. Rhizomatic curriculum is not driven by predefined inputs from experts; it is constructed and negotiated in real time by the contributions of those engaged in the learning process. More of Cormier’s research on this can be found on his blog davecormier.com/edblog.

Some teachers don’t trust what the students are doing, they don’t like students sitting behind laptops with hidden windows and screens, but he again encouraged educators to take up this technology in an open and useful way. Create unique hashtags for your course/lectures and have a Twitter back channel during each lecture session. In one of his lectures students were using the Twitter back channel to discuss a book that was part of that week’s readings, he ReTweeted their question, tagging the author, and the author joined the conversation!

In addition to being useful in the classroom, Twitter can be an incredibly useful tool for academics and the power of networking can greatly improve your likelihood of being cited. Note: I went and did a quick bit of research on this myself and found some interesting research showing that actually yes, ‘highly tweeted articles were 11 times more likely to be highly cited than less-tweeted articles’.(Eyesenbach, ‘Can tweets predict citations?’, J Med Internet Res 2011, 13(4), DOI: 10.2196/jmir.2012)

The session was incredibly interesting and also provided my new word for the week: Darwikianism

Session 3: Altmetrics, analytics and tracking engagement
Attended by Dominic Tate & Ianthe Sutherland

The first presentation after lunch from was Euan Adie who is the CEO of Altmetric. He spoke about the lessons that have been learned by trying to measure impact – overall lesson is that you can’t measure impact! He spoke about how it’s hard to engage academics with impact as many feel that the quality of their research should speak for itself. Citation counts just give academic interest rather than the impact an article has made – impact should be academic attention as well as broader attention. Social media attention and blog comments doesn’t really speak for the quality of the research but can show how much research is being talked about online. Academics don’t seem to be credited for policy decisions and documents and this could be a direct impact of their research. Euan also spoke about Altmetrics, a manifesto (altmetrics.org/manifesto) and a conference 2:AM (www.altmetricsconference.com) coming up in October.

Then Anna Clements – Assistant Director (Digital Research), St Andrews University Library – gave an excellent presentation highlighting the role of a University Library in supporting the research undertaken at an institution.  Anna made a series of good points about the importance of partnering with researchers during the research process rather than simply dealing with the outputs retrospectively.  Anna also highlighted the changing role of the library as seeks to support new areas such as research data management and the research impact agenda.

The final presentation was from Kaveh Bazargan, Director, River Valley Technologies. He spoke about how research is submitted in PDF form and all the metadata about the research, such as the title, author, email, abstract keywords (in XML format) is then reverse engineered from the PDF when this information should just be provided with the PDF in the first place. He gave a good analogy that the PDF is represented by a smoothie and we need to know what the fruit is (the XML). He then gave a live demo of River Valley’s product for writing which automatically produces the PDF in TeX format and the associated metadata XML without any reverse engineering required, as the XML is produced at the source. It is easy to make updates and re-produce the PDF and XML as needed.