Han Woo Park 1
Copyright © 2010 International Communication Association
e-science • e-research • South Korea • webometrics • link analysis • cyberinfrastructure • hyperlink network analysis
DIGITAL OBJECT IDENTIFIER (DOI)10.1111/j.1083-6101.2010.01517.x About DOI
In his book Cyberscience,Nentwich (2003, p. 28) notes that e-science studies are interdisciplinary in nature. A number of analytical traditions exist; for example, social construction of technology, actor network theory, socio-technical interaction network theory, and actor-centered institutionalism (for a concise discussion about similarities and differences between approaches, see Meyer, 2006, p. 38–39). The common argument of all these approaches is that e-science technologies are shaped by socio-institutional factors. In research on the adoption of digital photography in scholarship, Meyer (2008) applies the socio-technical interaction approach to examine societal developments and scholarly changes that take place when individual researchers and research and development (R&D) organizations use e-science technology and techniques to replace traditional scientific investigation. His research shows that e-science-related phenomena can be examined particularly in the broader context of social shaping of technology. This is because e-science is a moving target (Nentwich, 2003, p.4).
Further, the e-science community demands descriptive studies that investigate the specific process of e-science innovation at both the global and national levels (Jankowski, 2009). This study investigates the presence of e-science-related terms and the use of hyperlinks by e-science websites without preconceived specific theories about how the terms are communicated online and how the links are used. As Park and Thelwall (2008, p. 688) states, this approach is particularly useful because there has been little previous social-scientific Internet research of e-science. In this paper, we focus on a specific national e-science system: South Korea. More than other Asian countries, Korea can be expected to focus on e-science innovation (Soon & Park, 2009). Korea is well-known for having fast broadband connections and widespread Internet use. The government has developed a variety of e-science programs to link national research with international collaborators. However, previous reports (Byun, 2007, Cho, 2007; Jeong, 2007; NIA, 2007; Park et al., 2006) on Korean e-science do not adequately address the social forces shaping e-science technologies and have not focused on whether the general public is even aware of 'e-science' and its research practices. For the most part, e-science literature in Asia, including Korea, has been concerned with the hardware-oriented technical dimension of e-science tools, software, and advanced networks. While there is a generally positive view among policy-makers and early adopters of e-science, there has not been much serious examination of the direction of national/regional e-science policy and e-science discourse. Despite the tension in the UK research community over the term 'e-science' and the favorable view of the term 'cyberinfrastructure' in the U.S. community, previous studies have not examined what term is appropriate for R&D infrastructure and practices in Korea.
In order to solve new scientific problems that challenge academics around the world, a sizeable amount of financing to develop an e-science infrastructure and its related tools at the national and global levels is necessary. This demands the involvement of governmental organizations and cooperation with academic and industrial partners (for the global trends of e-science, see Dutton & Jeffreys, 2010 forthcoming). However, because of the lack of studies on the subject, little is internationally known about key actors in the Korean and broader Asian e-science communities, their institutional visibility, and their specific roles in the national and global contexts. Furthermore, the interaction of e-science organizations has not been investigated in a systematical way.
This article presents empirical findings from a webometrics1 analysis of the semantic variation, disciplinary scope, and institutional structure of e-science programs in South Korea. The problem with current e-science literature is that it hasn't been widely discussed, nor adequately examined in relation to Asia. Also, most of the studies that have looked at e-science so far have taken mainly a policy analytic approach.
Approaches to e-science
There have been studies related to this paper's topic—e-science. Two major strands exist and a third alternative strand is emerging in the e-science community (Bill, 2006; Gentzsch, 2006; Jankowski, 2007, 2009; Schroeder & Fry, 2007). The major strands are: 1) the computational perspective based on the use of high performance computing and 2) the networking perspective based on virtual collaboration through the Grid. A minor but growing approach to the study of e-science is the methodological perspective based on the use of new digital tools available online for conducting humanities and social science research. While the first two strands are closely associated with the natural and engineering science community, the third approach is less connected to that community and more associated with the broader interdisciplinary research community.
More specifically, the overall focus of most of the e-science projects in the natural and engineering sciences over the past decade has been facilitating high-speed processing of large volumes of digital data with the aid of grid computer networks. This is not surprising. Current e-science practices and collaborations, which rely on increasing computerization of the research process involving large capacity machines embedded in a networked environment, would not have been possible before the development of high-speed computing and networking technologies. At the same time, humanities and social science-based e-science projects often emphasize two major areas: 1) development of online tools to automate the research process, such as communication, research management, data collection and analysis, and publication software; and 2) experimentation with new types of data visualization, such as social network and hyperlink analysis and multimedia and dynamic representations (Jankowski, 2009).
E-science developments in Korea from an international perspective
There are national variations in e-science projects (Meyer & Schroeder, 2008). For example, the UK's e-social science initiative is focused on a broad range of social science disciplines, while in Germany there has been a major focus on e-science business applications. In the U.S., the National Science Foundation funds many e-science projects in the natural sciences and recently has begun funding development of e-research platforms analyzing the social network structure of the Web and collecting real-time multimodal behavioral data. In their review of UK arts and humanities e-science projects, Blanke et al. (2009) identify four separate research agendas: 1) addressing concrete user needs in the humanities, 2) managing the data deluge, 3) building virtual workbenches for digital data, and 4) creating an e-research structure for the performing art. This is in contrast to Germany, where the D-Grid initiative focuses on textual data editing (http://www.textgrid.de) in order to build an e-science community in humanities and social sciences.
In their study, Soon and Park (2009) provide a general description of the ongoing trends related to e-science policies and practices in Korea and Singapore. According to them, Korea's e-science program has evolved in the natural, biomedical and engineering sciences with a strong emphasis on high-performance computing and advanced research networks for long-distance collaboration. Cho notes that the main objective of Korea's national Grid initiative, the K*Grid project, is to construct the next-generation Internet and business applications: "Currently advanced environments or problem solving environments are being constructed in five application areas (nanotechnology, biotechnology, aerospace, equipment control and meteorology) as part of the Korea e-Science project. Each application will be developed using common application support software and infrastructures which was [sic] built by the K*Grid." (Cho, 2007, p. 248).
The broader e-science community would include humanities and social science researchers. However, according to a policy-maker in one of Korea's national e-science institutions, these researchers appear reluctant to use e-science technologies, even though they may be able get funding for their e-science research projects through government bodies (Soon & Park, 2009). This may be because Korean scholarship in humanities and social sciences is not mature enough to accept the use of sophisticated digital technologies in its research. In the first international survey of e-humanities and e-social science practices and technologies, Dutton and Meyer (2008) report that some proponents of e-science research practices in the humanities and social sciences are actually reluctant to promote e-science more actively. These proponents are concerned that researchers in these 'soft' fields would resist technological change as threatening to existing research practices and technologies that form the basis of their socio-academic stature in their fields.
On the other hand, the imbalance in Korea's e-science project deployment could be related to the e-science policy community. In their empirical study of the UK's e-science community, Wouters and Beaulieu (2006) criticize e-science policy making because only a few academics with advanced computers and physics backgrounds dominate the national e-science agenda. The assumption of the UK e-science community is that the input of humanities and social science researchers is not important in formulating e-science-related policy programs. The advanced computer technologies and related networking and archiving facilities needed for e-science affect the way in which social scientists approach e-science research, raising ethical and methodological issues. While e-science approaches tend to reflect a bias towards the natural and engineering sciences around the world (Hey, 2006), humanities and social science researchers are left out of core e-science programs more in Korea than in Western countries.
Webometrics analysis of e-science programs
Our study focuses on understanding the dynamics of the e-science agenda through tracking how web objects, related terms, and hyperlinks within and across institutions are circulated in Korea's webosphere. For a comparable approach, Ackland et al. (2007) examine the online structure of UK e-science programs based on a longitudinal hyperlink pattern between e-science project groups and funding agencies, such as the National Center for e-Science. They investigate how the web is being used as a place for newly emerging organizations to interact with established e-science institutions. Even though Alexandre (2005) does not examine e-science projects in cyberspace, a web link analysis of research institutions with long-term offline collaboration enable him to trace the networked characteristics of research groups. Similarly, we are able to discern the networked characteristics of R&D groups with offline connections from their electronic linkage patterns.
In a similar vein, Fry and Thelwall (2008) identify the diffusion of one particular European e-science enterprise (the Common Language Resources and Technology Infrastructure (CLARIN) consortium) by measuring different patterns of hyperlinks according to disciplinary boundaries. However, they find that there are too few bi-directional links in the linguistics and language resources communities to measure the patterns of diffusion across disciplines. They call on the research community to take their preliminary work and provide additional analysis using the webometrics method to complement their findings. In research analyzing the e-science webosphere in the U.S. and the UK, Meyer and Schroeder (2009, forthcoming) provide an interesting approach in response to Fry and Thelwall's (2008) call. They examine the social shaping process of online e-science knowledge based on searching activities conducted using Google Insights, a site that provides basic statistics about individual keyword queries via Google. For example, they find that the search pattern for the term 'cyberinfrastructure' reveals which U.S. regions are deeply involved in national e-science projects, such as major supercomputing centers.
As this review of the related literature has shown, e-science approaches vary according to terminology, discipline, and country. These e-science issues have been relatively well examined in the Western context, but it is essential that these same issues be studied in the Asian context. To fill this research gap, we explore e-science practices and issues in the context of the online structures of e-science-related institutions and the broader webosphere landscape in Korea. Some previous research suggests that webometrics data have flaws; however, if the collection process uses various sources properly, webometrics data can give a reliable indication of the focus and direction of e-science efforts.
Hence our research takes a combined approach in analyzing Korea's web networks in terms of the prominence of e-science-related terms, the densities of external hyperlink connectivity, and the structure of inter-linking associations among e-science sites. More specifically, we pose the following research question:
RQ) What are the characteristics of e-science developments in South Korea, based on substantial hyperlink and other web data, network-analytical results and qualitative investigation?
The data for this study were collected in January 2009 from Yahoo.com using LexiURL, a webometric analysis tool developed for social scientists by Professor Mike Thelwall (see Zuccala & Thelwall, 2006; see also the site related to the software: http://lexiurl.wlv.ac.uk). The search queries for this study included Korean words comparable to e-research, e-science, e-humanities, e-social science, cyberscience, cyberinfrastructure, e-infrastructure, digital humanities, and cyberresearch. These search queries were previously used in Meyer and Schroder (2009) for scientometric mapping of e-science and turned out to be useful in retrieving a wide variety of documents on e-science.
As summarized in Table 1, these search strings displayed 1,055 webpages2. We extracted the unique site names from the webpage URLs for each query. Additionally, MSN Windows Live Search, another frequently used search engine for webometrics analysis, generated similar results to those summarized in Table 1. However, Yahoo.com is the only search engine that could run the link queries, so we used Yahoo.com throughout the research. One may question the selection of
Yahoo.com because the domestic company Naver.com is the most popular portal website in South Korea. But, simply speaking, this fact has nothing to do with the choice of our research tool, Yahoo.com. Naver.com is a portal where South Koreans access their emails, blogs, and online cafes. The Korean version of Yahoo.com (kr.yahoo.com) is a portal with fewer visitors than Naver.com. Unlike such portal sites, Yahoo.com indexes the Web across national borders. There may be some bias about the coverage of certain countries, which is beyond the scope of this study. Further, as mentioned previously, Yahoo.com is the only webometric tool that enables researchers to conduct the hyperlink network analysis employed in this research.
We conducted a refined webometric analysis of 1,055 webpages and 810 sites. Three types of webometric analyses were performed in this study. First, the most prominent words were extracted from the summary information about the returned webpages. Second, site sources were classified by authors into the following categories: mass media, technology-focused media, portals/blogs, public organizations/governmental sites, academic associations/universities, and private companies/industry sites. The sites were classified in terms of the organizations providing content and services, not in terms of technical maintenance or web hosting organizations. Third, both co-link and inter-link network analyses were employed to investigate the structure of the online e-science landscape in Korea. While co-link network analysis focuses on a hyperlink to two different sites from a third site, inter-link analysis deals with a hyperlink between a pair of sites.
Word frequency analysis of the returned webpages
Out of the 1,055 webpage URLs, only two URLs were each retrieved three times and 23 URLs were each returned twice. The remaining URLs were each retrieved only one time. This means that a collection of webpages retrieved using e-science-related key words does not produce many duplicate URLs. Contrary to some previous criticism about the lack of humanities and social science researcher engagement in e-science in Singapore and Korea (Soon & Park, 2009), it is striking that 164 webpages were returned using the term 'digital humanities'. On closer examination of these webpages and sites, we found that ewha.ac.kr (run by the Ewha Women's University) is the most often retrieved site, and there are lots of media reports about the university's new inter-disciplinary 'Center for Digital Humanities' (http://home.ewha.ac.kr/~ehc, its syllabus was returned; http://home.ewha.ac.kr/~scranton/html/data/DH.pdf) and its 'Humanities Computing Research Group (HCRG)' (http://www.hcrg.org/sub/about.asp). After Ewha University first made the jump to digital humanities in 2004, other major universities started integrating the use of digital technologies into the education of humanities, and researchers have received funding from either their own universities or external funding agencies. These Korean universities are: Sungkyunkwan University (http://home.skku.edu:8088/liberal/02_aboutBiz/index05.htm), Yonsei University (http://humanities.yonsei.ac.kr/menu02/ct02_02.htm), and Seoul National University (http://humanities.snu.ac.kr/hu_in_04/hu_in_04_a/1233435_7189.jsp). Because of its novelty, there is a high level of public interest in Korea in 'digital humanities'.
Word frequency analysis was done using information extracted from the webpages. Yahoo.com returned introductory text about the webpages it displayed. A total of 2,079 unique Korean words were discovered through this text research. The distribution of frequent key words is visualized in Figure 1. In the diagram, words that appeared in English in the text search are presented in upper case letters, while words that appeared in Korean and were translated into English appear in parentheses. The word "science" appears twice in parentheses because the Korean word for science appeared in two forms: one form using the Korean alphabet (larger size) and the other form transliterated into Roman characters (smaller size). The frequency of occurrence of each word is indicative of its relative level of public awareness. In Figure 1, media-related terms are prevalent: News (98 times), WWW (60 times), Internet (41 times), Blog (39 times), Newspaper (37 times), Empas (29 times), and Information (28 times). Also, there appears to be a focus on the novelty of e-science as an emerging technology: IT (32 times), Digital (30 times), NEW (27 times), New (27 times), First (22 times), and Cyber (18 times). While e-science-enabling research terms (for example, Grid, Collaboration, GLORIAD, KREONET) appeared relatively infrequently, the key e-science institution KISTI (Korean Institute of Science and Technology Information) occurred in English 24 times and in Korean 10 times, and the term SCIENCE occurred 25 times. Note that common words and terms were removed in the list of frequently occurring key words.
Out of 810 sites surveyed, 104 sites were each retrieved more than two times in the Yahoo.com search results. In our research, we focused on these sites (mean: 2.91; std. deviation: 1.175). The most frequently retrieved sites were daum.net and empas.com (7 times each), which are major portals/search engines. The next most frequently retrieved site was etnews.co.kr, which is a popular daily newspaper in the field of electronics and information technology. Ten sites were each retrieved more than 5 times in the categories of mass media, technology-focused media, universities, governmental research, and web-based blogging sites.
The distribution of 104 websites under investigation can be better displayed analytically and aesthetically through tagging visualization as seen in Figure 2. Overall, portals and media sites—including major portals daum.net and empas.com and media site etnews.co.kr—stand out in Figure 2. This suggests that there are many journalistic resources conveying e-science information to the general public. Upon closer inspection of these media sites, we found that their dominant perspective is superficial; investigative reporting is scarce and press releases are often cited as sources.
Table 2 summarizes 104 Korean e-science websites in terms of the institutional identities of their authors. Media sites were the most frequently retrieved, with slightly less than half of the sites for this study (44 out of 104 sites). Mass media sites, such as newspapers and broadcasting stations, occupied the top position. Technology media sites were third in terms of frequency (17 times), behind public/governmental sites (18 times). On the other hand, academic organizations, including universities, were lowest in terms of frequency (13 times). It should be mentioned that, apart from mass media, there were only small frequency differences among the categories.
Based on the results obtained from Yahoo, we constructed a co-link network diagram of the frequently retrieved e-science sites in Korea (Figure 3). The size of the 'plus' representing individual sites corresponds to the frequency of their retrieval, as summarized in Table 2. The size of the lines between sites is proportionally thicker according to the number of external websites co-linking to the sites. Sites tend to be closely clustered when they are often co-linked, but the location of each group on the diagram is chosen.
The visual mapping shows that governmental agencies are not frequently connected to each other and isolated from other actors. Compared to the public/government domain, the academic domain sites—composed of universities and scholarly associations—are relatively well linked among themselves and to both governmental institutions and industrial partners. The dearth of links among public organizations reflects the structure of Korea's e-science projects, which are managed at many levels; for example, there are different organizations for the allocation of supercomputing resources, the maintenance of advanced research and the development of e-science software. In Korea, some Advanced Research Networks (ARNs) are available for those who conduct distance research collaborations across countries: GLORIAD (Global Ring Network for Advanced Application Development), APII (Asia Pacific Information Infrastructure Testbed), and TEIN (Trans-Eurasia Information Network). On the domestic front, the Korea Research Environment Open Network (KREONET) was the first national ARN, launched in 1988. Another non-profit research network, Korea Advanced Research Network (KOREN), was established in 1995 and was initially used for academic purposes. However, in line with the national Broadband Convergence Network plan implemented in February 2004, the Korean government shifted KOREN's application to more general use.
Furthermore, it seems that the term 'e-science' has not been well-defined theoretically and empirically among Korean scholars and public officials involved in the study of e-science and the funding of relevant programs. In other words, a common e-science concept is not shared widely in the scholarly community or the policy community in Korea. Another interesting finding from the co-link diagram is the weak salience of mass media websites. As described in the classification section of the site analysis above, mass media sites are the most frequent author type, followed by government/public and technology media sites. On the other hand, portals/search engines/blogs and private companies sites occur less frequently. The co-link map reveals a hidden structure. While media sites have a weak presence in the co-link diagram, portals and companies are noticeably visible.
While co-link analysis as described above presents an external perspective on Korean e-science websites, inter-link analysis reveals direct associations among a number of sites (the arrow indicates the direction of inter-organizational linkage). As seen in Figure 4, there is one large cluster composed of eight organizational sites and one small cluster of three sites. Seven out of the 18 sites are not connected with the other sites in the public domain and isolated in this particular online network. Overall, sparse inter-linking among governmental sites confirms the lack of a unified public policy approach for Korean e-science programs.
Figure 4 provides more concrete information about the relative positions of important e-science actors within the public domain. The dependent character of e-science programs on technical infrastructure was clearly shown by this inter-linkage analysis at the level of institutional websites. Three sites (yeskisti.net, kisti.re.kr, kosen21.org) belonging to the KISTI organizational framework are densely linked to each other and form a visible triangular network within the larger group. Other sites are more or less connected to KISTI in terms of institutional collaboration. For example, the offices of GLORIAD (gridforumkorea.org, gloriad-kr.org) and the Super Computing Center (ksc.re.kr) are located in KISTI's headquarters and radar.ndsl.kr is KISTI's S&T information portal. The remaining site in this group, stepi.re.kr, comes from a separate organizational body, the Science Technology Policy Institute of Korea (STEPI), which develops national S&T policies. In this regard, STEPI is linked with Korea's public e-science network through radar.ndsl.kr, but it is not linked with KISTI-affiliated sites. Two websites, ipc.go.kr (Informatization Promotion Committee) and itglobal.or.kr (Global IT Network), are linked with nia.or.kr. These three sites are tied together through institutional management; the linked organizations are sponsored by the National Information Society Agency (NIA). The underlying structure identified from the hyperlink network analysis supports the notion that Korean e-science administration is overwhelmingly influenced by the management of supercomputers and advanced research networks, and not by a common interest in e-science-enabling research cooperation. For example, nia.or.kr is in charge of the Korea Advanced Research Network (KOREN), and kisti.re.kr manages both the Korea Research Environment Open Network (KREONET) and supercomputers.
While social science investigation of e-science initiatives is active in several parts of the world (Dutton & Eynon, 2009; Gentzsch, 2006; Jankowski, 2007; Schroeder & Fry, 2007), there is not much social science research being done on e-science in Asian countries, including South Korea. The current literature on e-science in Korea consists mainly of reports on technological advances and breakthroughs in software, tools, and advanced research network capacities (Cho, 2007; Park et al., 2006). There are a few policy reports (mostly written in Korean) about the management of national supercomputing resources and the deployment of the Grid by R&D institutions (Byun, 2007; Jeong, 2007; NIA, 2007). However, these policy reports often examine the status of technical facilities and focus on how much of the national science and technology (S&T) budget should be invested in developing e-science technologies for natural science and engineering research.
To correct this imbalance, this study examines the societal dimension of e-science. The particular focus of this article is on South Korea, where individual researchers and academic organizations are linked to a high degree via broadband networks across multiple institutions. More specifically, this article examines the ways in which e-science is perceived and distributed in Korea, especially in the webosphere, and in which public e-science institutions are linked. The data for this article were collected from Yahoo.com through LexiURL. Yahoo.com was selected because it is the only search engine that can run both key word and hyperlink queries.
Four key words (cyberinfrastructure, cyberresearch, digital humanities, and e-science) used in this study produced the most webpages. For the search of 'digital humanities', most of the webpages retrieved were various media reports about universities adopting digital humanities courseware, while the other three search terms generated relatively more relevant web documents on e-science technologies and related topics. Despite the claim that 'e-research' is the most comprehensive term for e-science-related endeavors and programs (Meyer & Schroder, 2008), we found that surprisingly few Korean webpages used that term, or the terms 'cyberscience' or 'e-infrastructure'. Webpages containing 'e-humanities' and 'e-social science' were virtually non-existent. The results of this study indicate that Korea places a high priority on e-science infrastructure and tool development in the natural and engineering sciences, as Soon and Park (2009) observed in their article about Asian e-science. Furthermore, the Korean use of the terms 'cyberinfrastructure' and 'e-science'—the first term prevalent in the U.S. and the second term prevalent in the UK—demonstrated that Korea's e-science program is a hybrid of these countries' efforts.
On the other hand, governmental agencies responsible for e-science facilities (e.g. supercomputing, Grid, advanced research networks) are underrepresented in cyberspace and sparsely connected with each other. This is surprising given the important role their activities play in national e-science developments. The hyperlink-mediated relationships between websites function as proxies for relationships between institutions and stakeholders in Korea's public e-science domain. Both the co-link and inter-link diagrams presented important information about the collaboration structures of Korea e-science institutions and agencies. Institutional cooperation has not developed enough to spread e-science beyond a limited sphere of activity. In addition, e-science programs are dominated by KISTI, which is in charge of supercomputing and Grid management. These results may not be surprising because e-science is normally driven by massive computing resources and distributed network platforms. Nonetheless, the findings provide insights into what kinds of e-science discourses are taking place and what kinds of institutional actors are active in Korea.
With regards to collaboration and communication between academic and public sectors in Korea, we found a very interesting result. On a closer examination of websites run by universities having 'digital humanities' courses and/or research groups, none of those sites had explicit hyperlinks to Korea's public e-science institutes or high-tech facilities, let alone hyperlinks to digital tools, data and materials in related fields. Instead, the focus of these digital humanities programs and research groups is creating a bridge between traditional pedagogy in humanities and web-based educational methods; they have little to do with e-science technology developments, such as employing high-speed computing power. Given that large-scale analyses of humanities texts and interdisciplinary collaboration across geographical boundaries require e-science technologies and networks, the absence of links between universities and governmental institutions indicates that they do not pay the necessary attention to each other's activities. The HCRG site did have links to three humanities' sites, but they were all international sites: Association for Computer and the Humanities (http://www.ach.org), Humanist Discussion Group (http://www.digitalhumanities.org/humanist), and the Virginia Center for Digital History (http://www.vcdh.virginia.edu).
In summary, this article presents a non-technical overview of Korea's e-science development and descriptions of online communication networks of academic institutions, businesses, and public e-science agencies. Given that e-science is the backbone of cutting-edge research regardless of disciplinary field, universities and public research institutions in Korea need to reach out to each other more and cooperate on research activities in order to seize the opportunities offered by e-science.
On the whole, we suspect that there has been goal drift among Korea's e-science community. This might have been caused by the lack of institutional governance as well as a mismatch between the top-down e-science vision and bottom-up local practices. As Fry, Schroeder, and Mattthijs (2009) find in their review of 12 UK e-science projects, the reasons for such a drift could be disconnection between governance and expectations at the institutional level and local practices at the project level. E-science necessitates multi-level collaboration across organizational boundaries, research practices, and scientific disciplines. In order to address current problems and future challenges in e-science implementation, we need to be clearer about the SWOT (strength, weakness, opportunity, and threat) of e-science -enabling scientific investigation, instead of being carried away by excitement over the significant computational power and advanced networking capacity of e-science.
The existing mechanisms of communication flows among Korean academic, industry, and public e-science sectors using conventional and informal networks do not seem developed enough to carry the meaning and operation of e-science projects. Humanities and social science researchers could make valuable contributions to the e-science domain at the national level and through development of e-research practices. But currently, these researchers are not the ones examining the intersection between e-science technology and scholarly practices in Korea. Further, there may be two possible interpretations about the lack of humanities in e-science. We suspect the reason why humanities are not so prominent and/or do not link to other areas in Korea is because humanities are bound up with traditional pedagogy, and/or because they are low-technology. This leaves policy makers and hard scientists with the responsibility to fill in this research gap; yet, these two groups often have little background in the social implications of new technology. This has resulted in a discrepancy between e-science definitions and rhetoric in Korea, and a disparity in understanding between policy makers, hard scientists and other groups regarding e-science. Unfortunately, this lack of understanding and cooperation in the Korean e-science community could frustrate further e-science development at the national level by generating uncertainty in the policy-making process.
As web-data collection tools are used more widely by research communities, social science scholars are increasingly empowered by large-scale analytic tools to develop scientific explanations of online activities and behaviors. But the technological advances, relying on large-scale web analysis and ready-made data gathering tools, need some complementary research techniques (Ackland, 2009). For example, our webometrics analysis of Korea's e-science landscape would have been greatly enhanced if we had known the motivations behind the hyperlinking activities (for additional interview analysis on linking motivations, see Park & Kluver, 2009). Further, scientometric analysis of e-science-related articles published by Korean researchers might strengthen this paper if there were a significant correlation between webometric and bibliometric analysis.
Another technique would be to treat hyperlinks between institutional websites as an indication of their 'critical' inter-organizational networks offline and the frequency with which websites are connected as the value of the cooperation ties between them. We would need to investigate a number of other data sources in order to discover the social shaping process of e-science in Korea. A recent study by Fry, Schroeder, and Matthijs (2009) is an example of this approach. They interviewed all UK e-science project coordinators in order to determine the extent to which there are social and institutional barriers to sharing of research tools, data, and output generated by their e-research programs. The findings from this research are indicators of ongoing trends.
As new scholarly materials and empirical data from other sources gradually become available, researchers might be able to illustrate some of the emergent structures underlying e-science practices, particularly in Korea. Nonetheless, the method employed in this study provides researchers with the basis to construct a longitudinal analysis of e-science within Korea and to compare the Korean case with other nations, since the data can be reproduced for time-series and international comparisons.
1 Webometrics is broadly defined "as the study of web-based content with primarily quantitative methods for social science research goals and using techniques that are not specific to one field of study." (Thelwall, 2009, p. 6).
2 Yahoo.com reported that these search strings yielded a total of 103,709 webpages on the first line of the search results. Of these, Yahoo.com displayed only 1,055 webpages when the author tried to click through from the first page to the last page. Previous research (Park, 2004; Thelwall, 2008) has consistently documented that search engines have tended to display no more than 1,000 webpages per query. Regarding this problem, the following disclaimer from Yahoo.com is given for queries resulting in large numbers of results: "In order to show you the most relevant results, we have omitted some entries very similar to the ones already displayed. If you like, you can repeat the search with the omitted results included." Further, as Yahoo.com acknowledged, there were a number of duplicate and irrelevant webpages among the non-displayed results. Therefore, we decided to limit our study to these 1,055 webpages.
3 While the term webpage refers to a specific web document (e.g., html, xml, pdf, doc, ppt, pdf, etc.) with its URL (Uniform Resource Locator), the term website is one that encompasses any collection of webpages that form a coherent whole in terms of content or organization. In webometrics, the term website is often used for a single web entity that represents all collections of webpages sharing the same domain name (Thelwall, 2009).
Dr. Han Woo PARK is currently working as an Associate Professor in the Department of Media & Communication, YeungNam University, South Korea. Over past several years, he has contributed important works in the area of Link Analysis (also called, Webometrics) from the perspective of Social Network Analysis. He is a Principal Investigator of an international research project "Investigating Internet-based Politics using e-Research Tools" (Apr. 2008–Aug. 2012) funded by Korean government's WCU (World Class University) Program (http://english-webometrics.yu.ac.kr). Prior to this position, he had been a research associate at the Virtual Knowledge Studio in the Netherlands. Dr Park worked as a visiting research fellow at the Oxford Internet Institute while he was taking part in Oxford e-Social Science project. He served as a guest editor of Journal of Computer-Mediated Communication's Special Issue on Internet Networks. Since 2007, I have served as co-editor of Journal of Contemporary Eastern Asia.
Address: Department of Media & Communication, YeungNam University, 214-1 Dae-dong, Gyeongsan-si, Gyeongsangbuk-do, Zip code: 712-749, Republic of Korea. Email: email@example.com, firstname.lastname@example.org Homepage: http://www.hanpark.net