Скачать 81.38 Kb.
Scholarly journals originally aimed to actively develop, select and diffuse knowledge, but judging publications also judges their authors. Hence publication data has come to be used in hiring, promotion, tenure and merit pay decisions. Journals have become not only the cultivators of academic knowledge, but also the gatekeepers of academic power. They affect individual advancement, academic department rankings, research fund targeting and library fund allocation (Rainer & Miller, 2005). Administrators find the publication selection process a convenient way to allocate limited resources, but is this use biasing academic publishing?
Have journals become promotion arenas, with theories the battle weapons, rather than knowledge fields with theories the plows cultivating scientific truth? When authors say: “I have three MISQ papers in, and if one hits my career is made” rather than “I had a great idea on the future of IS…” then academia loses. To succeed a profession must look outward, beyond itself, not inward to its advancement options. If publishing becomes a promotion game, our discipline has forgotten its original business.
How can we refocus academia on the business of creating, selecting and disseminating knowledge? Some of the options we now present may make the job of promotion and tenure committees harder, but let us as academics leave such chips to fall where they may. Our job is to grow, select and disseminate knowledge, not serve administrative needs like selection and promotion. If we do our job and let others worry about theirs, the overall system will work.
To change knowledge castles into knowledge gardens one can change the people, the processes or the knowledge exchange architecture itself.
Changing the people includes ethical calls for people to do what is right over what is convenient, as when editors ask authors to improve submission quality (Paul, 2005). We could summarize the last section with the following ethics:
The general advice is that not making any mistakes is to miss all opportunities, so be willing to make a few.
Timeliness is part of relevance, and publication processes affect timeliness. It is the nature of opportunities to fly by, so when opportunity knocks the answer must be timely. Yet while IS practice changes in months, journal cycle times are typically in years. MISQ recently noted it had about a year’s backlog of accepted papers that could not be published for print cost reasons (Saunders, 2005). Adding a year backlog to 1-2+ years review and 1-2+ years paper development makes academic papers 3-5 years old when first published. IS changes a lot in 3-5 years. How relevant is high quality if it is too late to have impact? In today’s climate, timeliness is not an option, but a requirement.
Journal rating systems could add timeliness criteria to measures of perceived rank (Rainer & Miller, 2005) or citation rate analysis (Katerattanakul, Han, & Hong, 2003). Journals could report turn-around times (from submission to editor decision), end-to-end times (from submission to publication) (Snodgrass, 2003), and other value measures like readership size and composition, reader rated usefulness and knowledge source influence (Nerur, Sikora, Magalaraj, & Balijepally, 2005). This would help focus journals on relevance as well as rigor.
Another approach is to reduce review times. In 1999 the Association for Information Systems introduced two online journals, the traditional double-blind peer review Journal of the AIS, and the Communications of the AIS. The latter offered authors the choice of a light (one person) or a full three person review. In 2001 CAIS was rated significantly higher (18th) than JAIS (30th) in journal impact rankings (Barnes, 2005; Mylonopoulos & Theoharakis, 2001), and in 2003 while JAIS published 16 articles, CAIS published 95, as about 80% of authors chose a light review. This “experiment” in publishing suggests journals pursuing rigor without timeliness will feel increasingly “under siege” (Grudin, 2004, p20).
Journals could make their content more relevant by adopting an “affirmative action” innovation policy, aiming to publish a first time author each issue (or state none was found).
If journals are castles in a feudal selection system, computer-mediated interaction (CMI) could tear down their walls by changing the architecture of academic knowledge exchange. Just as physical walls limit what can be done physically, so cyberspace architecture affects what occurs there (Lessig, 1999). Changing a knowledge exchange architecture is more than just adding email or web access to print-publishing. It affects all primary goals, of education, of quality control, and of research growth.
Education and dissemination
Cost economics seem behind journal acceptance rates that turn journal editors into the Scrooges of academic knowledge. For a print journal, the pages disseminated depend on page cost and subscriber fee economics. Journals cannot publish pages they cannot pay for. Some journals ask authors to contribute to page costs to afford more (or longer) articles, and if the author’s institute pays, it is not the author’s loss.
Electronic publishing changes the situation by minimizing printing, binding, shipping and storage costs. Suddenly it is possible to publish not 5% but 100% of submissions. If memory is cheap, and it is, one could publish electronically everything submitted. This would make available many papers that print publishing would reject. While the literature seems huge, a particular topic may still have only a handful of relevant print-published papers. A researcher on that topic might find even rejected and first submissions useful. Further, not all rejected papers are all bad, as they may have brilliant parts but a weak main idea, so researchers may gain value from rejected papers in their field if they browse them correctly. As nature can grow a lily on a compost heap, so a bad paper can spark a good idea. Much that is currently discarded by the print accept/reject dichotomy could, if published, add value to IS research. Given search engine power, why not let readers not gatekeepers decide what they can read? Of course an online journal need not publish at 100%, but can choose up to that value.
Selection and quality control
Given one can publish all, it is apparent that costs are not the sole reason for quality selection. Publishing costs automatically restrict who can publish, but when e-publishing lowers those costs the goal of discriminating quality remains. If everything were published, guides to selecting would appear, a phenomenon like Zagat’s restaurant listings or movie and theater reviews. Print journals confound dissemination and selection when in fact they are separate issues, as selectivity is a valid goal in its own right. Electronic journals need not select less because they publish more, though e-journals seem already seen as lower status than p-journals for this reason. Yet conferences were seen the same way, but today some IS conferences are highly selective (Grudin, 2005). The challenge of online publishing is to increase both dissemination and selectivity.
Once dissemination and selection are separated conceptually, we see that online publishing can allow more discrimination not less. While print journals are limited to an accept/reject dichotomy, electronic journals can rank on a multi-point scale (Figure 2), to create more selection information. New submissions could enter the system at the bottom as not yet rated, and work up the selective hierarchy based on their assessment. What today is simply “rejected” may in the future be the “Best second tier paper of the year”. Further, as well as traditional assessment by editors and reviewers, online publishing allows value assessment by readers via electronic voting, where each reader has only one vote, to avoid rigged voting. Readers and reviewers would rate from different perspectives, so a paper reviewers disliked could rise by popular acclaim, but equally rave reviews could direct readers to a hard to read good paper. Finally some papers could rise quickly, while others might rise only after years of work. In gardening terms, not everything “grows” at the same rate.
Research and knowledge growth
If advancing research is the primary goal of academic knowledge exchange, some wonder: “… to what extent introducing advanced technologies supports the ultimate objective of research – creating knowledge.” (Hovav & Gray, 2004). We now argue that e-journals could create more new knowledge than ever before. Creativity seems to occur at the intersection of different fields of knowledge, so letting knowledge flow across field boundaries increases new knowledge. Electronic interaction enables this by offering:
Figure 3 shows a typical print journal’s interactions: authors send papers to editors, who assess and pass them to reviewers who advise how to interact with the author(s), with accepted papers going to the readers. “Computerizing” the process with email lowers transmission times but leaves the basic structure unchanged. Figure 4 shows the new connections e-journals allow, many of which already occur:
igure 4 suggests electronic interaction could increase knowledge growth by increasing the quantity, speed and diversity of knowledge exchange. It offers new types of online interaction like reader comments, voting, author help boards, reviewer issue boards and review feedback. E-journals also have advantages like live electronic links – clicking a reference could bring up the referenced content directly. The options are many and complex. For example, fast communication could permit two-phase reviews, with reviewers able to ask clarification questions (only) before giving an assessment. Computers can manage work-flows to improve efficiency, as electronic submission enables checks for requirements like a biography, or abstract word lengths. Computer management could let authors track their paper’s review progress online, as they can now track package delivery. Reviewers could choose papers by due dates that suited them, as now reviewers may field many requests at a bad time and no requests when they have time. Electronic systems can automatically show data like reviews per year, to recognize reviewer effort. Paper section comments could form an entire sub-layer of “letters” linked to the main archive and each other. Already other fields like Operations Research and Physics have entire journals of letters where people publish new ideas quickly, and IS could introduce the same as an online product. Finally an online publication database could manage and present citation links and knowledge flows (Zhuge, 2006).
However not all electronic options are rosy, as more information exchange is not necessarily better information exchange. Online knowledge exchange systems are social-technical systems (STS) that must meet social as well as technical requirements (Whitworth, 2006). The mapping from technical to social is not simple, so apparently minor technical changes can have major social impacts. People have social reputations that seriously affect their lives, whether in the family, workplace, club or in public. How others see you affects how they treat you, so damage to one’s social reputation can have direct physical outcomes. For example, an electronic archive with early versions of a published paper could, perhaps for educational reasons, let readers click on a paper to see those versions, and how the paper developed. However some authors may not want early (and faulty) versions of their papers on display to reflect poorly on them. Increasingly complex STS systems face such issues, not of what can be done but of what should be done, a subtle issue we now briefly consider.
As software becomes more powerful, it can give a user all action choices, no choices or any choices the program decides. However what it cannot do, logically, is give all users all choices, as in social interactions one person’s choice can deny another’s. Society developed the concept of social “rights” to resolve disagreements without costly conflict (Freeden, 1991). While simple societies give all power to a few, modern societies have evolved to share rights in complex ways, based on legitimate ownership, where legitimacy is defined as fairness plus public good (Whitworth & deMoor, 2003). For example that people own themselves (freedom) is both fair and benefits society, and so is legitimate, The same is true for each person to own information about themselves (privacy) (Regan, 1995). Likewise copyright is legitimate, as it is fair that each owns what they create (Locke, 1963) and rewarding innovators benefits society, though some argue today’s copyright laws deny innovation, and so are no longer legitimate (Lessig, 1999). Legitimacy analysis transfers these concepts to information systems, proposing that like physical resources, information resources can be legitimately owned and rights allocated accordingly (Whitworth & deMoor, 2003).
E-publication issues like whether to show early paper versions now become issues of social rights. One solution is a simple central rule - never publish version archives - but legitimacy analysis allows another option: leave the decision to the information owner. If the author(s) who created the earlier versions permit their display, no privacy principles are contravened, as privacy is not about information secrecy but information control (Nissenbaum, 1997). In an online archive, allowing some authors to choose to release their paper’s earlier versions while others do not seems legitimate. In a similar way, applying legitimate distributed rights could usefully release much information currently bottled up by archaic and simplistic central control systems. Likewise changing the new paper’s view level between:
could be delegated to authors rather than editors, as central control can create bottlenecks. However every e-journal could set its rights and privileges differently. The above ideas to improve academic knowledge exchange are just examples not answers. The real answer is an attitude change in IS publishing, from seeing errors as failures to seeing them as the price of progress.
|Modeling & Tools: Information Systems Using the Knowledge Pyramid to Characterize Systems J. N. Martin, The Aerospace Corporation Modeling & Tools: Multiple Sectors||Main Categories of Information Technologies Systems Regarding Process Orientation and Knowledge Orientation|
|The move to "open systems" requires re-thinking interaction with knowledge representation systems||Knowledge is a power. That is truth. But the greatest power is the knowledge about knowledge, I e. how to learn|
|The 21st century belongs to the knowledge age, where acquisition, possession and application of knowledge are the most important resources||Computer Science, Knowledge & systems|
|Computer Science, Knowledge & Systems||Eglash, R. “Nanotechnology and Traditional Knowledge Systems.” In Donnie Maclurcan (ed)|
|Representing and using legal knowledge in integrated decision support systems: DataLex WorkStations||Avoiding Information Overload: Knowledge Management on the Internet|