The Globalization of an Author
Science & Technology Dynamics
Amsterdam School of Communications Research ASCoR
Kloveniersburgwal 48, 1012 CX Amsterdam
Cybermetric methodologies can be expected to reveal dimensions of communication other than those shown by scientometric operationalizations. In a previous study entitled “The organization of the semantic space of an author,” I studied the use of words in titles of articles by Professor Tibor Braun as a scientific author. This was on the occasion of his 60th birthday. This year, on the occasion of his 70th birthday, the Internet has become available as another domain. Among other things, the Internet enables us to study the “globalization” of an author. Techniques and methodologies similar to the ones used in the previous (p)scientometric study will be used for the analysis of the semantic space of “Tibor Braun” as a search term. The globalization of “Tibor Braun” can then be visualized by using a simulation.
Tibor Braun has been the Editor of the journal Scientometrics since its first appearance in 1978. On the occasion of his 60th birthday in 1992, a special issue of the journal entitled Psientometrics was issued, as a Festschrift in his honour. My contribution was an analysis of words used in the titles of his articles journals included in the Science Citation Index; I took the co-occurrences and co-absences of words as an indicator of the semantic space of this author (Callon et al.,1987). The study of the semantic space of a single author was previously suggested by Small (1978).
At that time, the analysis revealed three factors that could be designated in terms of the intellectual organization of the semantic space of the author, whereas a fourth factor was identified as “meta-scientific.” The words that loaded on the latter factor refered to Braun’s scientific activities in a more general sense (e.g., “recent,” “trends”), but these words loaded negatively on the other three factors. The reader is refered to the previous study for methodological details and the substantive results of the scientometric analysis.
Methods and Materials
In this study, I repeat the analysis using the Internet as the search domain. Rousseau (1997) has explored the structural differences and similarities between cybermetric and scientometric research techniques. In my opinion, the search engines at the Internet provide us with one window or another for a geometrical representation, while the system under study can be considered as developing algorithmically (Leydesdorff & Curran, 2000; Bar-Ilan, 2001 ). A complex dynamics can be mapped from different angles, but these perspectives can be orthogonal, and accordingly the representations may be nearly incommensurable (Leydesdorff & Van den Besselaar, 1997). Thus, the problem of mapping the Internet provides a meta-factorial problem: how many three-dimensional geometries can meaningfully be distinguished in this (four-dimensional) space-time? (Leydesdorff, 2001) The latter can be considered as a “virtual” reality.
In this study, I limited the search to AltaVista as a search engine and used “Tibor Braun”—in quotation marks—as the unique search term in the query. On 22 February 2002 this query was repeated three times with time intervals of several hours (Rousseau, 1999). All these searches provided me with 37 pages as a result. This domain was downloaded; the headers and footers that refer to information about the search engine were then removed from the text. The text was further organized in a database manager so that “key-word-in-context” (KWIC) searches became possible.
673 unique words occurred in these pages a total of 1627 times. Single occurrences (a total of 417), words which can be considered as typical of the web environment (like “http,” “translate,” etc.), abbreviations, dates, single and dual character words, and common particles (“and”, “the,” “of”) were removed from the file. By deleting the “s” as a final character when it occurred in a word, plurals and singulars were automatically equated, and the deviation of the third conjugation singular of the verb (e.g., “he walk(s)”) was also suppressed.
The remaining 161 words were further reduced by disregarding the words that occurred only twice in this dataset (e.g., “Leiden”, “Leydesdorff”, etc.). A matrix was then constructed consisting of the remaining 66 words as the variables and the 129 (non-empty) lines of the original download from AltaVista as the cases. This matrix was subjected to factor analysis and multi-dimensional scaling using SPSS.
Twenty-nine eigenvectors of this network of co-occurrences and co-absences of words (Hesse, 1980;  Law and Lodge, 1984;  Leydesdorff, 1995 ) have an eigenvalue larger than one. Thus, the distribution is not pronounced. In order to maintain comparability with the previous publication, I chose to factor-analyze initially in four (orthogonal) dimensions. The results are shown in Table 1.
Factor One can be designated as “Spanish.” Obviously, Spanish words have a tendency (in this file) to co-occur specifically. Factor Two and Three are highly correlated, but they differ only to the extent that both “Tibor” and “Braun” correlate negatively with Factor Two and positively with Factor Three. These two factors refer mainly to Braun’s work as a chemist. Note that “Chemistry,” “Science,” “Fullerenes,” “Nuclear,” and “Approach” correlate positively with “Tibor” and “Braun” in both dimensions.
Factor Four represents Braun’s interest in science studies. The words most prominently loading on this factor refer to a workshop of the Network of European Centers in Science and Technology Studies (NECST) that was organized in Amsterdam from 21-24 May 1996. Tibor Braun was an invited speaker at this workshop.
I interpret the results of this mapping of the semantic space of the words “Tibor Braun” on the Internet as exhibiting the environments of his working relations. The center of these relations is generated by a dense cluster of words that refer to his work in Budapest, Hungary. This work has two dimensions that we will be able to analyze below in greater detail using multi-dimensional scaling techniques. Two other axes become visible on the Internet, notably one with Spain—in Spanish—and one with The Netherlands.
Since Factor Two and
Three are not different except for the position of “Tibor Braun,” I decided to
pursue the analysis further in three dimensions in order to obtain a global
picture. By clicking on
Figure One, these global relations can be visualized in relation to the daily
rotation of the globe:
The multi-dimensional scaling solution of this data provides us with more detail about the inner cluster of words that are located at the center of the factor solution. This picture (Figure 2) shows a virtual axis that can perhaps be characterized as cognitive (‘science’ is placed at the top of this axis). The horizontal axis can be associated with Braun’s publication behaviour. This axis seems to become more specific as one follows it to the right side of the representation.
Figure 2. MDS-mapping of “Tibor Braun” at the Internet
(S-stress = 0.199)
The stress-level of the two-dimensional representation (S = 0.199) converges in only five iterations and cannot then be further improved. These stress-levels legitimate the prediction that “Tibor Braun”’s activities are rather robust and sustainable. We wish him at least another decade of professional contributions.
Compared with the previous study, I suggest that the results of these Internet searches represent the practice of the active scientist “Tibor Braun,” whereas the representation based on the ISI-database—the Science Citation Index—exhibited the intellectual interests of the author in much more detail. The two databases are different in terms of the codification of the title words (Leydesdorff, 2001). The higher degree of codification of title words in the Science Citation Index allows for insights into the intellectual organization of the semantic space of the author, while the Internet reveals the patterns of his global relations.
 Loet Leydesdorff (1992). ‘The Organization of the Semantic Space of an Author,’ Psientometrics 60(1), 19-24.
 Michel Callon, John Law, and Arie Rip (Eds.) (1986). Mapping the Dynamics of Science and Technology. London: Macmillan.
 Henry Small (1978). ‘Cited documents as concept symbols,’ Social Studies of Science 7, 142-7.
 Ronald Rousseau (1997). ‘Sitations: an exploratory study,’ Cybermetrics 1, Issue 1, Paper 1, at http://www.cindoc.csic.es/cybermetrics/articles/v1i1p1.html
 Loet Leydesdorff and Michael Curran, ‘Mapping University-Industry-Government Relations on the Internet: the Construction of Indicators for a Knowledge-based Economy,’ Cybermetrics 4 (2000), Issue 1, Paper 2 at
 Judit Bar-Ilan (2001). ‘Data collection methods on the Web for infometric purposes — A review and analysis,’ Scientometrics 50(1), 7-32; at http://www.kluweronline.com/issn/0138-9130/contents
 Loet Leydesdorff and Peter Van den Besselaar (1997). ‘Scientometrics and Communication Theory: Towards Theoretically Informed Indicators,’ Scientometrics 38, 155-74.
 Loet Leydesdorff (2001). A Sociological Theory of Communication: The Self-Organization of the Knowledge-Based Society. Parkland, FL: Universal Publishers; at http://www.upublish.com/books/leydesdorff.htm .
 Ronald Rousseau (1999). ‘Daily time series of common single word searches in AltaVista and NorthernLight, Cybermetrics 2/3 (1999), Paper 2 at http://www.cindoc.csic.es/cybermetrics/articles/v2i1p2.html .
 Mary Hesse (1980). Revolutions and Reconstructions in the Philosophy of Science. London: Harvester Press.
 John Law and P. Lodge (1984). Science for Social Scientists. London, etc.: Macmillan.
 Loet Leydesdorff (1995). The Challenge of Scientometrics: The development, measurement, and self-organization of scientific communications. Leiden: DSWO/Leiden University.
 Loet Leydesdorff (2001). Indicators of Innovation in a Knowledge-based Economy. Cybermetrics, 5 (Issue 1), Paper 2, at http://www.cindoc.csic.es/cybermetrics/articles/v5i1p2.html .