The thesis is currently available in pdf only, in three large sections, but see Abstract and Paragraphs from the Conclusion below for a quick summary.
A 'journal' based on this model was built and subjected to extensive evaluation by a targetted specialist group. In a series of tests these users were invited to offer their views on this and more general journal models; because this was not a hypothetical model their responses could be cross-checked against the results of their experiences of using the new journal model. Thus, users could not hide their true preferences. Although small in number, the target evaluation group provided strong reactions both for and against the model, revealing some of the challenges that radical new electronic journal models face if the structure of the traditional printed journal is to be displaced in favour of journals that can properly exploit the prospect of fully interconnected and accessible networked distribution.
Electronic journals exist in a post-Gutenberg, post-Google information environment. The ramifications of the former have been widely discussed (e.g. Harnad 1991, Guédon 2001), the latter relatively little in terms of its impact on the journal literature. Google may be a search engine rather than a journal, but the ability to locate a specified item of information precisely and instantly among the mass of information available on the Web -- by March 2001 the Internet Archive had stored 10 billion Web pages (100 terabytes of data) -- has profound implications. In the electronic environment the search engine has become the de facto interface to information, rather than the fragmented packages that have migrated from the print world.
The task for journals is to reinvent themselves as expert filters, labellers and commentators on the literature, because when anything can be found instantly this is the service that scholarly users in particular will need most. Even in this respect Google is not content neutral, analysing authored Web links to rank results using techniques that have become familiar in scholarly citation analysis.
That few journals, while completing the massive task of creating electronic versions and services since 1996, have yet to reinvent themselves in this way is hardly surprising. The findings of this study suggest that many users are wary and unprepared for what must inevitably follow.
The volume of networked information available to users today may be large and intimidating, but in scale it is embryonic. Science is moving towards data networks on a very large scale -- 'e-science' will increasingly be carried out through distributed global collaborations enabled by the Internet, requiring access to very large data collections, and very large-scale computing resources. It is inconceivable that the published peer-reviewed literature will be unaffected by such developments, in form as well as format (Berners-Lee and Hendler 2001, Rzepa and Murray-Rust 2001). Digital information demands to be managed as a continuum: on this scale there is no other option.
... Closing paragraphs
Despite efforts to improve the supporting technology and user interface, there is a more fundamental feature that will ultimately determine the success or otherwise of the model on which PeP is based, or indeed any electronic publishing model: selection is key. The influence of Google on the digital information environment again becomes evident. When any scholarly work can be exposed publicly and located instantly, what should be the basis for selection? This transcends peer review. A peer-reviewed journal can deny recognition but can no longer deny publication. Works are not uniformly good or bad, which peer review suggests. Authors and readers deserve a more open dialogue that enables works to be viewed in context. In an open information environment, is fully automated citation analysis, such as demonstrated by ResearchIndex, to be preferred to editorial selection?
PeP seeks to inform research by anticipating improved access to papers in many new contexts. Users anticipate the empowerment of electronic access, but not the obligations that will accompany it, the need to interpret and assess works rapidly, identifying new insights and making connections, and responding within a framework that builds on that connectivity.
Scholarly research is difficult. Pursuing scholarly research requires skills in managing information, and maintaining a comprehensive and up-to-date knowledge of all information pertaining to the specified research. In the emerging scholarly electronic information environment, in which it can be predicted that access to research papers will eventually become easier, these skills will not become redundant. Neither will services such as PeP make these skills redundant. Rather, easier access will place higher demands on these skills, as researchers will need to mine vast data sources more extensively, more forensically, seeking previously unidentified connections. PeP may not be a ‘journal’, or the ideal implementation or the perfect model for a service designed to assist the researcher in coping with these new demands. It is certain that services like it will not just be desirable, but necessary.
Email contact: firstname.lastname@example.org
More about Steve Hitchcock
This page was created Thursday, 12-Sep-2002. Last
The URL of this page is