Communications of the ACM
Sharing Research in the 21st Century: Borrowing a Page from Open Source Software
The Internet continues to evolve and digital devices continue to emerge, increasing the possibilities for reshaping how we work. The technology revolution has impacted many aspects of society including the research publishing industry. Increasingly, hard copy journals and other scholarly material are made available online. Although many traditional research publications are now online, little has changed regarding the process used to create and publish research. The notion of an open document has recently emerged with regard to books. Bruce Perens and Prentice-Hall have initiated an effort to publish open books with an accompanying open book license [3]. This license, like the one proposed in this article for open source research, is similar to the GPL used with open source software.
One of academia's chief responsibilities is research and publication. This article proposes an alternative approachan open source research approachto research and publication. Two distinguishing aspects of the open source research approach are the extensive use of Internet technologies and variations in the current copyright practice. The open source research approach borrows significantly from the process used by the open source software movement. As stated in the Techweb encyclopedia online, open source software is: "Free source code of a program, which is made available to the development community at large. The rationale is that a broader group of programmers will ultimately produce a more useful and more bug-free product for everyone, especially because more people will be reviewing the code. Peer review is considered one of the most important safeguards to prevent buggy code, but is often not given enough, if any, attention by software companies. Peer review is a natural by-product of open source projects. In addition to having better code, open source software allows an organization to modify the product for its own use rather than hope that the vendor of a proprietary product will implement its suggestions in a subsequent release. Examples of popular open source programs are the Apache Web server, sendmail mail server, and Linux operating system."
Adapting this description, the proposed open source research approach might involve:
"Written manuscripts and other forms of research work, which are made available to the community at large after reaching completion as determined by the authors. Peer review is considered one of the most important safeguards to higher quality research and thus open peer review would make available more reviewers thereby increasing the scrutiny and feedback for each research work. In addition to the potential for increased quantity and quality of research, open source research would allow anyone to modify and adapt existing research and create derivative research work without the limitations imposed by copyright law."
The proposed open source research approach aims to enhance the creation and dissemination of research in four ways:
- Facilitation of research collaboration,
- Expansion of the formats for documenting research,
- Opening of the peer review process, and
- Changing how we make research available.
Collaboration. The Internet offers easy access to collaborate with others and access files regardless of location. The open source research approach as described would allow others to easily communicate with the authors concerning their work during its development. Since the Internet provides a convenient environment for communicating and retrieving electronic documents, an open exchange and idea sharing within the community could be easily achieved. Further, the Internet improves the opportunity for academicians and practitioners to bridge the gap between their worlds and engage each other in common areas of interests.
Academic conferences have traditionally been a forum for researchers to exchange research findings and make new acquaintances but enjoy only a limited participation from the academic community and even less from practitioners. Moving research to an online environment would make it easier for both practitioners and academicians to collaborate, hence expanding the opportunities for IS research. Research agendas could be developed online, perhaps targeting a different set of issues than would otherwise occur due to easier accessibility, an open process, and broader awareness of others with similar interests. Benbasat and Zmud suggested that future research should have more relevance with regard to the use of technology in practice [1].
By using the proposed open research approach, academicians and practitioners with similar interests could more readily discover each other and have the opportunity to team up on projects. Practitioners could post important questions or challenges facing their industries and academicians interested in those issues could participate, sharing their extensive research skills to help provide a more scientifically derived solution or understanding. Similarly, academicians could post an idea for a research project and have interested practitioners join the effort with useful information from their experiences and their ready access to real environments and other resources that could help support the research initiative.
Clearly, an open and online approach to the current research process would result in additional opportunities for collaboration and agenda setting. The implementation of this approach would provide a means for research initiatives to be date- and time-stamped with the authors' names even if the work has not been completed and released for public consumption. Such an approach would provide evidence as to who started a particular topic of research and when the research initiative began so its authors receive proper acknowledgment. The principle of an open and voluntary effort where others with similar interests can contribute in an open fashion has resulted in excellent outcomes, as demonstrated in the software world (for example, with Linux and the open source software movement). Using an adapted process could produce positive research benefits.
Expanded formats. The present publishing guidelines for many academic journals make it difficult and time consuming to publish research. The current practice typically involves a researcher designing and conducting a research project followed by the time-consuming process of writing a formal research paper that can take months to complete. Glass notes that writing style can be a primary determinant of acceptance in academic journals even if less formal styles of documenting equal the accuracy of the formal styles demanded by academic journals [2]. The additional time required to refine a draft research document into a formal paper could be used to conduct additional research while leaving the opportunity for others with more of an interest in writing to craft refined versions of the draft document. This provides an opportunity for the community to collectively leverage its strengths. It allows those with a strong interest in research design and data capture to focus on those activities while allowing others who prefer writing to refine the documentation.
If alternative formats for documenting research work were allowed, there could be an increase in the quantity of work that is made available. Working papers, notes, illustrations with brief descriptions, and other information formats could be posted online. By posting research work in alternative formats, input and participation from the community would be easier since other writing styles would be acceptable. In some instances it might be helpful to modify an existing document to address a similar set of questions or issues. Online documentation and discussion could permit straightforward editing changes after the original posting, and with the use of the GPL or similar licensing, existing work could be modified and used to produce derivative research. Any limitations that may be realized by not requiring the document to be in a formal research paper form may be offset by the benefits of increased research and participation.
Opening of peer review process. Research published in a hard copy research journal requires a thorough review in an effort to enhance the quality of research. The process of publishing research typically involves submission followed by a review by an editor and/or associate editor and several reviewers (see Figure 1). Further, today's process calls for editors to locate reviewers with the appropriate background to review articles for publication. Based on the feedback, articles are eventually accepted or rejected.
The open source research approach would allow academicians and practitioners to individually decide the merits of the posted research (see Figure 2). In essence, the responsibility for assessment of the research would shift from a limited number of editors to a large number of qualified reviewers in the readership, providing the opportunity for a more extensive review process. Further, it would reduce the risk that research of interest to some is not overlooked due to its failure to be published in a formal journal. Expanding the assessment of research provides a broader base of evaluation and offers the opportunity for higher quality work to be achieved, as evidenced by the quality of open source software.
Simple rating scales can be appended to the bottom of each research work, as is commonly used now on many Web sites, to allow readers to rate each work. In addition, all raters would complete a personal profile, so that anyone reading reviews would not only be able to identify the reviewers but also access information about their qualifications. Collectively, this information could provide an easy way to tabulate the usefulness and quality of the research posted with full information concerning the reviews and reviewers.
A change in the process of assessment will address the question of whether a design that relies on a few acting on behalf of the entire community is more effective than a design where the entire research community (both academicians and practitioners who are members of the research network) has the opportunity to decide the merits of the research. A similar process is used with open source software development, as the entire community has the opportunity to review and test software for completeness and quality, but with the understanding that any changes to the work are decided by the original authors. Consequently, the proposed open source research approach leaves the task to other researchers and interested parties to determine the relative merits of the research work. Further, it allows a greater involvement from the community that includes both academia and industry.
Another advantage of the open source research approach involves the time to press, which could be improved significantly since working drafts could be posted anytime and revised as often as desired. This approach would provide the researcher with the opportunity to receive feedback during the development, similar to the beta testing process used in software development.
In essence, hard copy journal articles represent static fixtures in journal archives, ones that limit the ability to create derivative works due to copyright restrictions. In addition, research is spread across a large number of publications, making it difficult to pool all related research on a particular topic. Providing an online environment for posting research and related discussions could make research more organic in nature, while allowing others to create derivative research work (with all works organized in one portal). Additionally, having the ability to conduct threaded discussions concerning the work could generate additional ideas that lead to more useful and higher quality research. This online approach is characteristic of many other changes taking place in our world since the widespread availability of the Internet. The Internet has removed many barriers to knowledge production and acquisition as the world moves deeper into the information age.
Research portal. The Internet represents a vast repository of information and communication capabilities. It has revolutionized the acquisition and storage of information and other media such as music, video, news, and audio due to its global accessibility and functionality. The Internet provides the ability to construct portals where documents and discussions can be housed with availability on a global basis. The design of the open source research approach suggested in this article advocates organizing research via a research portal that uses a taxonomy for classifying the research (by creating forums for different categories of research, for example), and providing the ability for threaded discussions and other useful Web capabilities (see Figure 2).
The open source research approach includes features that do not have an equivalent in open source programming. For example, using a discussion board program, each research working draft could be posted in a forum that represents a particular category of research. The authors or principal investigators could post an abstract with a link to the draft materials. For each posting, threaded discussions could be initiated, offering suggestions and comments concerning the work or the posting of a derivative work. The postings would serve as evidence of contributions being made, similar to a portfolio, as well as providing a vehicle for researchers with similar interests to share their expertise and ideas. The proposed approach opens the boundaries of the academic community to industry and others who may be interested in a particular research effort. Consequently, this approach allows a greater variety of research with differing formats to be available while expanding the opportunities for contribution and authorship.
In addition to the technologies needed for creating an open research portal, there would be a need to administer the portal. Today, each publisher has the responsibility for maintaining a site for its publications. The open source research portal would consolidate and standardize the organization of research, making it more efficient and convenient for the researchers, readers, and administrators. A reduction in maintenance could be realized since the technology necessary for creating such a portal is readily available and could be implemented to minimize redundancy. Candidates for the management of the portal could either be an existing association or society in the field or a volunteer university.
Conclusion
The proposed open source research approach makes use of software that is common on the Internet and can serve the purpose of creating a more collaborative working relationship between academicians and practitioners. Further, it expands the playing field for individuals to express their talents and gain feedback from a larger peer review. Finally, it has the potential to take research contributions beyond the formal research paper with alternative forms for representing research work.
References
1. Benbasat, I. and Zmud, R. Empirical research in information systems: The practice of relevance. MIS Quarterly 23, 1 (Mar. 1999), 316.
2. Glass, R.L. Stodgy by design, and the notion of `Dumbing Up.' Commun. ACM 45, 2 (Feb. 2002) 25.
3. Shankland, S. Book publisher tries out open source. CNET News.com, Jan. 17, 2003; zdnet.com.com/2100-1106-981018.html.
©2005 ACM 0001-0782/05/0800 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2005 ACM, Inc.