The first day of the APE conference in Berlin, which, as mentioned before, focused on the impact of publishing in the digital age, started with a keynote by Georg Winkler from the European University Association (EUA), entitled Universities in the 21^st century. Winkler started off by asking the question of what makes an university unique, quoting Ralf Dahrendorf’s conception of a university as “a ragbag of institutions with no clearly defined boundaries and no substantive core”. After a historical overview of different types of universities and different kinds of university reform, Winkler asked what we want to achieve for the university of the 21^st century, which is now functioning more and more in a digital system. While sketching his vision for a future university, Winkler made a plea for the relevance of “open science” for knowledge societies. As Winkler states, in knowledge societies the bulk of new knowledge should be generated and disseminated via rapid publication by giving up the rights over it. This will not only facilitate the generation of further knowledge, but it will also help students to be equipped with the best and latest knowledge and it allows the latest results to feed into the innovation system. According to Winkler, open science is justified by the huge positive external effects it will have in knowledge societies. But as a counter note he states that this ideal can clash with the academia-business relations, and can create problems with the incentive to do research and to publish (the principal agent dilemma). Winkler however strives for coexistence of these matters. The question remains how to give incentives and set the mechanism in such a way without obstructing the scientific freedom of the researchers. Winkler concludes that the future definition of universities can be seen as the outcome of these kind of discussions.

A nice flashback to the preconference day occurred with yet another lecture on the Semantic web, as the second keynote lecture was delivered by Rudi Studer from the Semantic Web Science Association (SWSA) on Semantic web applications and tools. Studer starts by comparing the classical web, which focuses on humans as consumers of (semi-structured) content and in which the meaning of information is not accessible to computers and search is keyword-based, with the semantic web, which publishes data in structured and linked formats, specifies the meaning of data and their relationships with formal models (ontologies) (meaning that application systems of computers are able to understand, capture and process the meaning ) and focuses on technical standards as set out by the W3C.

Studer goes on to discuss RDF (Resource Description Framework) as a common data model of the Semantic web. RDF gives a natural representation of all kinds of data and it is graph based, which means it is made for integration. The Semantic web can thus be seen as a global database, as a web of data and data models and the interlinks between them. Data integration can than take place using triples. They allow information to be combined in a very flexible way. The question now is how to generate these triples? Studer explains that we can generate semantic web content by using semantic wikis. He states that Web 2.0 approaches can be used to generate ontologies in a Semantic MediaWiki. Through this Semantic MediaWiki you can specify the specific property in the RDF and than connect it in the ontology, which allows you to add semantic annotations to the wiki content. In this way mashups can be created in a Semantic MediaWiki. The Web 3.0 thus becomes a big knowledge mesh connecting people and information to create and share knowledge. Using semantic wikis enables the producing of semantic content collaboratively and more easily.

Creative Commons also uses powerful semantic models which allows search for works based on their licenses and allows for precise definitions of desired rights (e.g commercial use etc…). This can of course be very helpful to help solve the problems coined on integration and standardization of licenses at the preconference day.

In the second session, on usage and impact, Ian Rowlands (CIBER Group) talked about Electronic journals: modeling journal spend, use and research outcomes. CIBER is a research group at the University College London (UCL) that studies the digital transition to scholarly communication and more in specific scientist’s information behavior in this respect. What do scientists do online? How do they behave and how do they use the online content? They conducted a study framed around two very important research questions:

How have researchers responded to the unprecedented levels and convenience of access to scholarly journals?

How has enhanced access to the literature affected the research process and research outcomes?

According to the study, Google is hugely popular and influential, it is the first choice of preference for scholars and so it drives a lot of content to journals. Opening up journal content to Google is thus a way to get more traffic to your journal. Now does this continued or enhanced access lead to greater productivity, Rowlands asks? According to the study, the work activity of researchers increasingly goes on beyond the working week. Working days becomes elongated and one third of the searches on which the study was based were made outside the working week. What does this mean: does this mean scholars are much more pressurized? Does this mean that information consumption and production are in some way related to each other? Is there a link between efficient search and successful research? According to Roland we still need to explore these questions more thoroughly. What was clear however from the study, was that the most successful research institutions tend to use gateways more often and this is reflected in much shorter sessions on the publishers platform. Does this mean that super-users are also super-producers? Yes, says Rowlands, they found a tentative link between e-journal consumption and research outcomes. Rowlands concludes by mentioning that this is still a preliminary report, primarily used also to frame (upcoming) questions for the stakeholders in scholarly communication. The study will be published on the CIBER website in 4 weeks.

Highlights of the second day of APE will follow soon.