Online Place Branding: Is Geography ‘Destiny’ in a ‘Space of Flows’ World?

: Places have been promoting their attractions throughout history for almost a century now and place marketing started capturing the attention of economic researchers in the early 1990s. Although the globalized space has become a major interdisciplinary field of study in the past few decades, we still do not have a definite solution for measuring or predicting the changes it brings. The aim of this study is to propose an instrument to help scholars quantify the Web 1.0 and Web 2.0 components of the online presence of place brands and their spatial dynamics in a broader global context. By taking the ‘space of flows’ approach to frame globalization, we questioned whether geography is still destiny in the context of online place branding communication. To answer the question, we developed a category grid and applied it in the content analysis of 82 official country websites. We found that geographical position is still important in conditioning the adoption of Web 1.0, and does not make a significant difference in the adoption of Web 2.0. However, the strongest predictors of the adoption of Web 1.0 are not the ones related to geography, but to socio-economic inequality. It is therefore possible for the theorizations of the advantages of the ‘space of flows’ globalization to be suitable only for Web 2.0, Web 1.0 still being under the influence of the ‘space of place’. We discussed the implications of adopting the network society theory in place branding communication in order to nuance the understanding of the transitions from Web 1.0 to Web 2.0 and from Web 2.0 to Web 3.0 in place marketing.


Introduction
Although places have been promoting their attractions and their images throughout history for almost a century, through rivalry and competition in North America during the late nineteenth and early twentieth centuries [1] and place marketing developed as specialized domain in the early 1990s [2], the 'spatial turn' from social sciences revived the concept of place and gave it new nuances over the past decades, causing the present globalized space to become a major interdisciplinary field of study. In the sociology field, globalization has been conceptualized as a network society with its 'space of flows' that reduce the influence of the geographical space. From the branding perspective, Holt [3] proposes a new field of analysis specific to brands, advocating for a sociology of branding, in order to be able to describe and explain contemporary politics and society, as "a distinctive mode of capital accumulation. [...] it generates profits by creating and then exploiting various kinds of social dependency. A brand becomes an economic asset for the firm when people come to count on the brand to contribute to social life, when it is embedded in society and culture" [3] (p. 300). Similarly, in the marketing field, it has been theorized that globalization brings significant changes and articulates place branding communication, specifically in the online environment. From this point of view, brands are economic signals circulating in the economy, which allow companies to establish credible information about their offers. Brands guarantee that the products have a certain quality and reliability and that their manufacturer is present in case something is not functioning properly in the market [4] (p. 356) In consumer psychology, which is a dominant paradigm in the marketing process, brands are understood as devices that simplify consumer decision making and lower search costs [4], and from the perspective of behavioural psychology, the process of branding a place is seen as an attempt "to build positive place images by (re)constructing and projecting identities, building expectations around worthwhile place experiences" [5] (p. 33). However, we still do not have a definite solution for measuring or predicting these changes in order to understand place marketing better. Therefore, the main contribution of this paper is to propose an instrument to measure the adoption of Web 1.0 and Web 2.0 in country branding communication. To increase the predictive validity of our instrument, we tested four hypotheses derived from the network society theory and its recent reviews.
In order to nuance the understanding of the spatial dynamics of the Web 1.0 and Web 2.0 components of the online presence of place brands in a broader global context, we focused on how the local public authorities involved in place brand management choose to use official websites to communicate the online identities of their countries. In accordance with the network society theory, we questioned whether geography is still destiny in the context of online place branding communication, and conjectured that geographical position is less important than globalization factors for adopting digital technological advancements, and that even if geography might no longer be destiny, socio-economic inequality could still be.
To justify the research question and its potential answers, in the Literature Review section, we show there is a need for conceptual and theoretical clarification of place branding that could be addressed by adopting concepts from sociology of space and virtual communication. In this way, we lay the ground for a new theoretical framework in which place branding communication is an instance of the network society. Following this new theoretical framework, in the Materials and Methods section, we argue for introducing a new research instrument that could be used to measure place branding communication and a research design to assess its reliability and validity based on already established measurement indexes in fields of communication, marketing, and sociology. By applying this new instrument to 82 existing place websites, in the Results section, we answer the research question and test the hypothesis derived from its conjectures. Then, we discuss the implication of our findings by contextualizing them into the proposed theoretical framework. We conclude by stating the theoretical and practical contributions of our study and its potential applicability in future studies.

Literature Review
Recently, three flagship reviews of place branding literature have emerged. First, Papadopoulos and Heslop [6] searched the literature on place marketing and place branding published between 1952 and 2001 and identified 799 publications and the perennial quality of the field. Second, Hanna and Rowley [7] reviewed literature about 'place' and showed that place branding shifts the focus from tourism to branding and business, highlighting recent transformations of the field. Third, Oguztimur and Akturan [8] analyzed 146 articles from four disciplines in their systematic review (tourism, marketing, urban planning, and geography) on the topic of city branding published between 1988 and 2014, 70% of which have been published after 2010, which could mean a revival of the field. This paper builds on the opportunities and limits identified in the aforementioned reviews as well as in other recent published papers [9,10] and aims to fill the gaps and extend place branding literature in ways explained as follows.
The results of the previously cited reviews are compatible with the idea that place marketing relates to place branding, which includes country branding, regional branding, and city branding [11,12]. They resonate with Andersson and Ekman's definition of place branding as "the use of various marketing practices to promote a place's attractiveness for business, tourists, residents and students" [13] (p. 41). Hence, in various levels of our analysis, we took into account all three manifestations of place brands (city, region, country/nation) and did not reduce place branding to destination branding [14], considering online branding communication to be a tool for addressing various public audiences, that has an increasing potential and is adapted to today's global challenges.
The main conclusion of Oguztimur and Akturan's systematic review [8] was that there is a need for conceptual and theoretical clarification and consolidation of place branding which can be achieved by surpassing the mono-disciplinary approach and being more open to theorizations from sociology and anthropology. While in the market-oriented approach there is a consensus that branding helps places stay relevant on the global market [15], which is increasingly dynamic and completive [16], social sciences warn about the risk of commodifying places by consciously ignoring acute social and economic issues [17]. In this debate, Oguztimur and Akturan [8] recommend following a spatialized politics of flow, instead of a place-based politics of competition. Therefore, we framed our study based on a sociological perspective, namely Castells' [18] theory of network society, with the subsidiary concept of 'space of flows'. In addition, we integrated concepts from the field of communication studies, like the static 'web 1.0', the dynamic 'web 2.0' [19,20]; the semantic 'web 3.0' [21], and specific socio-economic indexes, like the KOF Index of Globalization [22], the Human Development Index [23], the Web Index of the World Wide Web Foundation [24], the GINI index [25], and the classification of countries based on income provided by the World Bank [26]. The components of this trans-disciplinary approach will be clarified in the following subsections of the present paper.
In the marketing literature, it is commonly accepted that websites have become fundamental communication tools for any type of brand [19] and social networks [10,27]. G. Christodoulides states that "brand is a brand regardless of context. What changes is the enactment of the brand" [20] (p. 143). When the brand concept was initially adopted in the online environment, the sites became "fundamental tools of communication" [19] (p. 315) for any type of brands, by using factors such as physical delivery and returns, locating the brand and speed of download, site appearance, navigation, personal support and differential reward [28]. Some years later, all of these aspects have become hygienic factors (i.e., responding to basic, natural, identifiable needs in Web 1.0), in the sense that they became almost compulsively used, but along with those other factors emerged, beyond immediate brand control defined by the top-down strategy. Branding now occurs at the next level (defined by the Web 2.0 concept), through blogs, widgets, user generated ads, social networks (e.g., Facebook, Myspace), podcasting, videocasting, content sharing (e.g., Scribd), video sharing (e.g., YouTube), etc. [20]. This means that "an effective online presence is vital to a brand's internet success" [29] (p. 300). Web 2.0 is viewed as a sum of "technologies that facilitate social communication" [19] (p. 316), providing users, with these Web 2.0 specific applications, the new possibility to "even interfere with the brand's values" [20] (p. 143).
In the particular field of place marketing, Oguztimur and Akturan [8] used thematic analysis to define branding communication and online branding as two of the five dimensions of place brands. They classified official place websites as indicators employed by researchers in their analyses, as the use of digital platforms to collect data in the field increases, but also in other subjects of research [9,30]. This is why our study focuses on how official websites build the online presence of various country brands. On one hand, official place websites are considered a top-down approach [31], representing the way authorities involved in place brand management choose to use new technologies to communicate the online identity of a particular place. On the other hand, scholars [32] emphasize the need for bottom-up studies, based on the participation of community members [33,34]. To solve this contradiction, our mix approach focuses on both top-down and bottom-up dimensions of official place websites, by including Christodoulides' [20] and Lincoln's [34] distinction between the static unidirectional online communication tools (Web 1.0) and the dynamic bidirectional ones (Web 2.0) used in online place branding communication.
The content analysis conducted by Oguztimur and Akturan [8] in their systematic review revealed that the studies in the field from the last two decades display some methodological limitations. Half of them are qualitative research papers based on a limited number of case studies, which causes limited generalization, and they tend to privilege highly economically developed places, mostly from the Global North. According to the same source [8], as regards to city brands, scholars focus mainly on eight cities, namely Berlin, Amsterdam, Birmingham, Bradford, Manchester, Shanghai, London, and New York. Similarly, while conducting a social and semantic network analysis to develop a new place branding measurement tool, Sevin [21] only focused on two US global cities, namely Boston and New York. In order to reduce the aforementioned limits, we took into account that the place brands favored by the literature in the field are already important nodes in the global network society [35], and that market-oriented approaches on place branding tend to ignore acute social and economic issues [17]. Hence, in order to avoid the risk of reproducing place privileges and obtaining results of limited applicability, we decided to conduct a quantitative content analysis on an extensive sample of 82 country websites. We made the selection based on a classification proposed by the World Tourism Organization that defines six geographic regions (Europe, Africa, the Americas, Eastern Asia and the Pacific, the Middle East, and Southern Asia). We described the details of our sampling decisions (as well as the configuration of the pilot study that included region and city websites in addition to country ones) in the Materials and Method section of this paper.
As an increasing number of recent studies approach emerging topics like Web 3.0 [21], one might question our decision to focus on Web 1.0 and Web 2.0. We believe that the transition from one type of web to the other takes time and is an incremental one, which is also the opinion of other scholars [19,20]. Considering the unequal online place branding communication global context, in which some countries do not even have an official website while testing the semantic web, is more of an emerging exception than a rule. In order to cover the entire online communication landscape, we wanted to configure a suitable tool for measuring and predicting the first transition, which would be adaptable enough to allow the incorporation of the second one in the future. In addition, as scholars have started to worry about the replicability crisis and the biases in the application of established methods and tools in science in general [36] and social sciences in particular [37], we acknowledge there is a need for more studies that accurately apply established methods to a variety of samples to produce results that stand the test of time. Nonetheless, the methods used in the present paper as well as the obtained results could be the starting point for future trend analysis based on longitudinal data, used to describe and predict the complex dynamics of the adoption of the three types of web in online place branding communication.
To summarize, we brought evidence, based on extended systematic reviews of the literature in the field of place marketing, that there is a need for conceptual and theoretical clarification of place branding which can be addressed by: (1) taking into account various audiences; (2) surpassing the limits of the market-oriented approach by being more open to theorizations from social sciences and following a spatialized politics of flow; (3) combining the top-down with the bottom-up approach; (4) the lack of quantitative research methods and extended samples taking into account not only notorious place brand from the Global North; and (5) the need for replicability in the field by creating contexts for longitudinal studies and trend analysis to be conducted in the future. To answer these limitations, we propose a new interdisciplinary framework combining theorizations from communication, marketing and social sciences which, as far as we know, were not used before in place marketing contexts. These theorizations will be synthesized as follows.

Place Marketing and Country Branding
The interest for brands and the research around them arose from the intermingling of the management, marketing and strategy spheres, which are generally aiming to propose pragmatic models regarding the "effects" produced by brands, through quantitative analyses [38]. Other studies, following the constructivist paradigm, which reflect the "nature" of the brand, "as a living entity (with a personality with which we can form a relationship and that can change and evolve over time)" [39] (p. 12), indicate a reconsideration of the general functioning mechanism of brands and the focus shift from products and their producers to the response and services provided to the consumer, in understanding the process of building brand value [40].
For economists, brands are a form of information, focusing on marketing strategies [4]. These are economic signals circulating in the economy, which allow companies to establish credible information about their offers. Brands guarantee that the products have a certain quality and reliability and that their manufacturer is present in case something is not functioning properly in the market.
The concept of place marketing emerged in the early 1990s as a distinctive theoretical and research field with the work of Kotler et al. [2] that "include a presentation of best practices and a theoretical approach" [41] (p. 6). By adding the term 'place' (or sometimes 'geographical' that corresponds to mean cities, metropolitan areas, regions, communities, states, countries or nations) to the 'marketing' term, it can lead to a type of specific product to which an ordinary process will be applied, considering that there is no practical difficulty in replacing physical products with places [42] and the use of techniques normally associated with the creation of classical product brands [43]. The marketing popular model of '4p' (product, price, placement, promotion) [44] and "its performance in the procedure of place/city marketing" [45] (p. 369) or the adoption of some "strategic marketing management tools and conscious branding" [46] (p. 253) or designing a place to satisfy the needs of its target markets [47] are key ingredients of place marketing.
Dolea and Țăruș [48] acknowledge the prominent role played by international public relations in building place brands, "representing the systematic and planned effort of a company, institution or government to establish mutually beneficial relations with the audiences of other nations" [48] (p. 18). In fact, strong arguments can be made in favor of the claim that "public relations strategies for internal and external audiences are the ingredients of a recipe for creating the national image. A positive image is considered an economic gain, as it generates tourism, creates cordial relations with other governments and increases that country's chances of benefiting from aid" [48] (pp. [19][20]. Place branding can be used as an example on a public-private partnership platform for tourism development and citizens' engagement [49].
Starting from the perspective that country branding is a field of theory and practice that aims to measure, build and manage the reputation of countries [43], we can appreciate that this field is similar to place branding, if we accept that, in this case, the country is the "product" of our analysis, compared to the other perspective of place branding, in which space becomes "product". Considering the definition of the branding process as a way to differentiate a certain product from its competitors, we can extrapolate the reasoning and similarly attribute to country branding a strong impact on the country's ability to win in international competition [50,51].
Place marketing has become a prominent feature of the economic development strategy, in terms of place development and country branding recognizing the brand's ability to position itself in the minds of users, rather than being a creation entirely controllable by a marketing function, as "the unique, multi-dimensional blend of elements that provide the nation with culturally grounded differentiation and relevance for all of its target audiences" [51] (p. 15)

'Space of Flows' in the Network Society
The network society theory states that we live in a global economy based on international networks of economic exchange and fluxes of online communication, in which the significance of physical space, also called 'space of place' is decreasing [35]. More exactly, recent advancements in information technology are transforming the geographical space into a space of flux so that access to informational networks becomes more important than local particularities of specific geographical spaces [18]. Since in the network society spaces no longer confine within fixed national territories and become flows of images, sounds, symbols, interactions, experiences, goods, capitals, and technologies, we aimed to question whether geography is still destiny in the context of online place branding communication. While trying to answer this complex question in accordance with the network society theory, we conjectured that geographical position is less important than globalization factors for adopting digital technological advancements in place brand communication.
In addition, we believe that the model of the network society theory is only an idealized version of the future, in the sense that in the present time, even if geography might no longer be destiny, socio-economic inequality could still be. In order to clarify these answers, we will explain what we mean by digital technological advancements, globalization and socio-economic inequality factors.

Globalization as Interconnectedness and Spread of Innovation
The network society with its 'space of flows' is a sociological way of explaining how globalization works, by focusing on an international and virtual web for diffusing information. For this reason, and to bridge the economic field, we decided to make use of existing globalization indexes in order to calculate how ready each country is for the future network society. We used the KOF Globalization Index [22], as it has been used in more than 100 studies [52,53] and the Web Index [24], calculated by the World Wide Web Foundation. Marketing theory and practice is adapting to the new global context by gradually assimilating the digital technological advancements. Over the past decades, branding communication has been expanding online in two stages in order to deliver the promised brand experience to a global audience. During the first stage of the World Wide Web's evolution, called Web 1.0 or the browserbased web, branding communication took advantage of controllable factors such as locating the brand, speed of download, personal support, differential reward, physical delivery, returns, site appearance, and navigation. However, during the second stage, called Web 2.0 or the interactive web, new factors outside of the brand's immediate control emerged as brand communication started to use blogs, widgets, viewer-created ads, groups, podcasting and video casting, or video sharing [20]. In brief, the transition from Web 1.0 to Web 2.0 meant switching from consuming a static website to producing and interacting with content on various, usually interconnected web platforms.
Nowadays, internet marketing communication strategies have become the norm not only for promoting tourist destinations [54] and managing place brands [55], but also in finding a balance between globalizing and localizing place identities [56]. This is why local authorities have recently started to pay attention to the potential of Web 2.0 for place branding communication [57]. For the future, it is expected that place marketing will migrate towards other emerging web platforms such as Web 3.0 (the semantic web or the web of context) which could offer a more customized brand experience, Web 4.0 (the web of things or the symbiotic web) which merges interconnectivity [58,59], artificial intelligence and offline objects of daily use, or even Web 5.0 (the web of thought) involving "emotional interaction between humans and computers by allowing the Web to recognize users' emotions and reactions" [60] (p. 64).

Inequality in Human Development and Income
Although globalization and the network society have the potential to become equalizing forces, scholars [61,62] have recently highlighted that "the digital environment can reproduce or mitigate inequalities that have been molded and routinized in the physical environment" [61] (p. 580). Almost two decades ago, Schiller critiqued Castells' theory of network society by highlighting that old patterns still persist in this so-called new society, in the sense that "the social and economic inequalities characteristic of market economies tend to widen rather than close" [63] (p. 48). Castells [18], who warns about the power of the network society to booster inequality and polarization, also acknowledged this idea. To address this problem, some scholars proposed various composite indexes to measure various forms of inequality in the network society, like informational inequality [64]. In the cited papers, inequality was measured either at the micro or at the macro level. Therefore, we launched the conjecture that, despite the opportunities provided by information technology allowing place branding communication to surpass geographical barriers, there might still be country-level inequality factors holding the process back. More exactly, we drafted the working hypothesis that the category in which a country is included based on income and human development might influence the adoption of Web 1.0 or Web 2.0 for place branding communication. To measure inequality, we used the Human Development Index [23], the most recent classification of countries based on income calculated by the World Bank [26], and the Gini Index [25].

Network Society as a Trope
We propose a new theoretical framework according to which, the network society can be understood as a trope that makes visible the shape of a communication process between the local, empirical, geographic space (which provides fixed spatial contexts for sensemaking) and the global imaginary place of information (in which sensemaking is decontextualized). The interconnection between space and place is mediated by languages and facilitated by digital technologies to produce recontextualized associations. In this way, Web 1.0 represents the content of communication; Web 2.0, the audiences and their arenas; Web 3.0, the meaning; Web 4.0, the interconnection between the physical space and the information space, and Web 5.0, the cognition and emotion. Hence, the distance between words in a text makes visible semantic differences in the same way in which the geographic distance between physical spaces produces different cultural contexts and sensemakings. This process can be visualized as an abstract texture of flows shaped by the movement between space and information, in a given period of time, which involves sensemaking, based on physical and imaginary objects and contexts that trigger cognitive and emotional associations. The physical geographical space is fixed in concepts such as cities, countries, regions and made dynamic by the economic exchange of goods and capitals, constrained by space-based socio-economic inequality; while the place of information is fixed in texts and communication protocols, such as Web 1.0, Web 2.0, Web 3.0, Web 4.0 and Web 5.0 channeled by digital technologies, and extended by flows of images, sounds, symbols, interactions, and experiences. The network society has the potential to create new places of information coagulated into displaced territories of associations, built from recontextualizing already acquired space-based associations (Web 3.0). In this paper, we used the trope 'space of flows' to express the possibility of the global imaginary place of information.
Place branding communication can be considered as a prototype of the network society described above, one of its many possible manifestations, chosen because the link between physical space and information place (ascribed meaning) is already established in our local and global imaginaries. However, this trope is only a heuristic to imagine a possible version of our social future in a world strongly confined by geographical space and time limitations and conditioned by politicoeconomic logics which produce inequality and is measured by already established indexes. In this paper, we used the trope 'destiny' to express these confinements of the local, empirical, geographic space. The network society is the globalization context (measured by already established indexes with economic, social and political dimensions), in which place branding communication occurs, following not only a communication logic (measured by already established indexes of Internet access), but also a market-oriented one, producing social effects (measured by indexes of human development, of income/GDP inequality) and guided by political logics, identifiable in specific geographic spaces (regions, countries, cities) with specific promotion vehicles (such as their official websites) adopting in various degrees top-down (Web 1.0) or bottom-up approaches (Web 2.0). Even so, if one takes into account that innovation might emerge from combining already existing tools and contents, these indexes can be used to produce tools that enlarge the possibility of predicting the dynamics of globalization.
Hence, the research question of this paper, already stated in the title, is whether the global imaginary place of information (space of flows) is confined or not by the local, empirical, geographic space (destiny). We conjectured that the transition to a bottom-up approach (Web 2.0) is incremental, therefore conditioned by a top-down approach (Web 1.0); which is confined by geographic and socioeconomic inequalities; and both transitions (to Web 1.0 and from Web 1.0 to Web 2.0) are contextual (depending on various dimensions of globalization: social, economic, political, informational).

Research Aim, Objectives and Hypotheses
The aim of this study is to develop an instrument to help scholars identify and quantify the Web 1.0 and Web 2.0 components of the online presence of place brands and their spatial dynamics in a broader global context. The general objective is to show how the public authorities involved in place brand management choose to use official country websites to communicate online identities. While formulating the specific objectives of the study, we extended Sevin's [21] analytic framework (Define-Measure-Visualize, DMV model) into DRMVP model, as follows (See Table 1): 1. Define and Refine-elaborate a coding scheme for analyzing the online presence of place brands and refine it through a pilot study assessing reliability and validity; 2. Measure and Visualize-apply the coding scheme to 82 country websites and graph the results; 3. Predict-test the hypotheses in order to predict the adoption of Web 1.0 or Web 2.0 in place branding communication in a more globalizing context and to assess the predictive validity of our instrument.

Methods and Techniques of Data Collection
We considered three main arguments when choosing content analysis of country websites as our main data collection method. Firstly, place brand analysis in the online environment is a relatively new field [19,33,65] that has no need for unanimously accepted models in the academic community or by practitioners. Secondly, content analysis can have a significant role in examining the manifest content of messages, while being an ideal tool for examining websites [66]. We also agree with Gibson and Ward [67] that the main advantage of this conceptual scheme is that it provides the framework for a standardized and flexible tool, so that as web innovations or new features of virtual presence are developed, they find their way into the scheme.

Data Sample
We used the database of the World Tourism Organization, UNTWO [68], as a sampling frame, including all the countries for which the organization offers information, regardless of their membership status-whether they are Member States of the Organization (156), Associated members (6) or Observers (2)-and using a classification [68] by geographic location: Africa, East Asia and the Pacific region, the Americas, the southern region of Asia, the Middle East and Europe. We considered the UNTWO website as an official and credible source of information, with global recognition. After setting up the database of 164 countries and 250 web addresses present in the UNTWO portal, representing the universe of research, we followed the stage of selecting the sources of communication, as a procedure used in criterion sampling, "review and study all cases that meet some predetermined criterion of importance" [69] (p. 176), considered by researchers relevant to the problem studied. For this purpose, we sought to specify the selection criteria, eliminating the following cases: (1) where it was proved that the address/addresses specified did not work [70]; (2) because of the lack of the web address of the place brand; (3) that generated a redirect to another web address; and (4) without English content [70][71][72].
Therefore, all other official countries websites (82 cases) became part of the investigated body and primary data source, presented in Appendix Table A1. Furthermore, in Appendix Table A2, there are detailed information about number and distribution of countries by geographical region.

Measurement
Four concepts were measured: place branding communication, globalization, inequality, and geographic position and are presented in Table 2.

Place Branding Communication
To measure place branding communication (PBC), we calculated two indexes (the adoption of Web 1.0 and the adoption of Web 2.0) based on our content analysis results.
Regarding the methodological criteria for conducting the empirical investigation to measure PBC, we used as a quantitative research method the content analysis on the official sites of the place brands (i.e., the country brands), in order to "record the presence or absence of attributes" [73] (p. 136), applying the procedure of frequency analysis on the system of analysis categories [74]. Under this procedure, the analysis unit was considered "the page" of the official websites.
The starting point in applying the procedure for establishing the characteristics according to the coding scheme, is the Homepage; this page being considered "a central gate to Web-based communication" [75] (p. 205), with most visitors deciding whether they will continue to browse a site based on their impressions of the homepage [76]. In addition, by using the Homepage as a unit of analysis, sites of different sizes can be compared, using the same degree of size, as a measure to "equalize differences between small sites with few links and larger sites with many links" [77] (p. 14). However, given the character of the analysis aimed at identifying web and graphical characteristics rather than textual content, it was necessary to use, outside the Homepage of the websites, other secondary pages, which allow tracking the presence or absence of these elements, as this methodological option is explained by R. Opoku and R. Hinson, who directed their research by defining the unit of analysis of both the Homepage and "four levels down the hierarchy of information" [70] (p. 122).
By analyzing and interpreting the data derived from the application of a coding scheme to the 82 selected cases, we identified variations regarding place branding communication managed in the online environment. At this point of research, we needed first to determine which characteristics are associated with the adoption of Web 1.0 in place branding communication and which one with the Web 2.0 environment; we structured the Web 1.0 and Web 2.0 parameters (see Appendix Table A3), constructing a summative index in SPSS for each environment, detailed and explained in the following section of the article.

B. The Assessment of Reliability and Validity
In developing our analysis grid and in assigning the value, code, or score to each indicator, we followed Hwang, McMillan, and Lee's [77] advice. More exactly, we focused on surpassing the limit that "[h]uman intuition can intervene in the process of developing coding schemes. Although all research is subject to some human intuition, the process of content analysis allows relatively more room for the researcher's intuition than do some methods such as surveys or experimental designs" [77] (p. 13). To assess test-retest reliability, we conducted a previous pilot study in which the category grid was tested on one website at different time intervals (April 2016 and July 2017) and calculated the Pearson correlation coefficient between the two data series (r(43) = 0.720, p < 0.05). Moreover, to assess inter-coder reliability, two independent observers encoded in July 2017 data and Cohen's K (Kappa) coefficient was calculated between the two data series (K(43) = 0.769, p < 0.05). Therefore, both test-retest and inter-coder reliability for the category grid used for content analysis are high.
Although it has been asserted that "each activity on the website contributes to place branding, which results in a certain image of the place" [33] (p. 283) or, at least, the development of the website's potential for branding communication [78], we did not take it for granted. First, for each of the two concepts, a summative index was calculated and normalized to vary between 0 and 100, based on which country ranks were assigned. Secondly, we used country ranks based on the FutureBrand Country Brand Index [79], the Bloom Consulting Country Branding Ranking Tourism [80], computing the Spearman's rank correlation coefficient. While the correlation between the Web 1.0 rank and the FutureBrand Country Brand Index rank was weak ( (44) = 0.327, p < 0.05), the one between the Web 2.0 rank and the FutureBrand Country Brand Index rank was moderate ( (44) = 0.537, p < 0.05). Additionally, we found that the adoption rank of Web 2.0 in place branding communication strongly correlates with the Bloom Consulting Country Branding Ranking Tourism ( (73) = 0.606, p < 0.05). The weaker correlations for the adoption of Web 1.0 mean to suggest that it has evolved into more of a precondition of place brand communication than a dimension per se and due to its increasing use (as the results will show), it does not discriminate among place brands as it has become the norm. However, the strong correlation of the adoption of Web 2.0 with other country brand indexes is a powerful argument for the concurrent validity of our instrument.

Globalization
Globalization (G) was measured using the KOF Globalization Index [22,52] with its three dimensions (economic globalization, political globalization, social globalization) and the Web Index [24] with its four dimensions (universal access, freedom and openness, relevant content, empowerment).
Regarding the KOF Globalization Index: economic globalization measures economic flows, political globalization measures restrictions, while social globalization measures information flows [81].
As for the Web Index, universal access refers to investing in affordable access to high quality internet infrastructure and in the education and skills that citizens need in order to use the Web efficiently. The freedom and openness dimension measures the right to information, opinion, expression, safety and online privacy. The relevant content dimension assesses the extent to which different stakeholders can access information that is relevant to them, in the language that they are most comfortable using and via platforms and channels that are widely available. Empowerment is linked to the difference that the Web makes to people and the extent to which its use by stakeholders produces positive change in society, economy, politics and environment.

Socio-Economic Inequality
Socio-economic inequality (SEI) was measured using three country classifications derived from the Human Development Index values [23], the World Bank's country classifications by income level [26], and the Gini Index [25] values. The second classification is derived from the fact that the World Bank separates four country income groups based on GNI per capita: high income (above $12,235/capita), upper-middle income (between $3,956/capita and $12,235/capita), lower-middle income (between $1006/capita and $3955/capita), and low income (below $1,005/capita).

Geographic Position
Geographic position (GP) was measured using the country classifications used by the World Tourism Organization, UNTWO [68]. It operates with six geographic regions (Europe, Africa, the Americas, Eastern Asia and the Pacific, the Middle East, and Southern Asia)-detailed in Appendix Table A1.

Statistical Analysis
After calculating the descriptive statistics, we applied the multiple linear regressions analysis (particularly the ordinary least squares regression) by following Hayes [82]. We ensured that the conditions of regression application were met (especially normality, homoscedasticity, absence of multi-collinearity, absence of extreme values and influential cases). We identified two statistically significant regression models, one predicting the adoption of Web 1.0 in place branding communication, and the other predicting the adoption of Web 2.0.

Results
This type of instantaneous (static) analysis [83] of an average number of websites, at a given period of time, allows comparisons and grouping of sites. Thus, we could observe that the websites of country brands tend to present a richer content from a graphic point of view, to the detriment of the text, according to other authors [31], as a main factor in capturing the attention of visitors in the online environment. Animated images or dynamic image galleries capture the most representative visual aspects of a country, because they promote tourist attractions, current events by selecting quality photos, at a high resolution (but reduced in size for adaptation to the online environment) in most cases. Although, in terms of presentation and appearance (as a dimension of adoption of Web 2.0), video materials (as a characteristic of the dynamism of the site) are used in over two-thirds of cases, we expect an increase in importance in their future, when the authorities that manage the country brands will use the advantages offered by the video content sharing platforms (e.g., YouTube, Vimeo) and other Web 2.0 applications in the social media category.
We were able to identify a higher degree of similarities between the sites, by repeating the same features, compared to the differences between them. In general, the virtual portals of country brands offer attractive content, from a visual point of view, with a trend from text to interactive and multimedia content. The commercial aspect is undersized on the official web pages (the frequency of online banners containing advertising messages being very low, for example: Ghana, South Korea, Vietnam), the only noticeable presence being the component of the upward flows of information: offers (e.g., shopping, rentals, reservations, orders, discounts, etc.), with a higher frequency of occurrence.
The results shown in Figure 1 suggest  The regression model 1c from Table 3 accounts for 73% of the variance of the adoption of Web 1.0 in country branding communication, while the inclusion of predictors related to the country's geographic position improves the prediction significantly. The adoption of Web 1.0 can be predicted based on globalization (namely, social globalization), inequality (human development, GINI, country income category), and geographic position (namely, if the country is in Eastern Asia and the Pacific, Europe, or the Middle East). If we compare the Beta standardized coefficients from Model 1c, we can see that the strongest predictors of the adoption of Web 1.0 are indicators of socio-economic inequality (the level of human development, if a country is a low-income economy or a lower-middleincome economy). Referring only to predictors as indicators of geographic position, the fact that a country is from Europe is the strongest one. The regression model 2a from Table 4 accounts for 44% of the variance of the adoption of Web 2.0 in country branding communication, while the inclusion of predictors related to the socioeconomic inequality and the country's geographic position does not improve the prediction significantly. However, Model 2c has a statistically significant predictor, which is an indicator of geographic position. Hence, the only identified predictors for adopting Web 2.0 in country branding communication are the three dimensions of globalization (empowerment, web relevant content and web social globalization). Testing the hypotheses: 1. As there are no country websites using the Web 2.0 facilities more than the Web 1.0 ones, it seems that the adoption of Web 2.0 for place branding communication is conditioned by the adoption of Web 1.0. This finding was also confirmed by the results of regressing Web 2.0 on Web 1.0 (R 2 = 0.14, F (1, 80) = 13.10, p < 0.05; b = 0.375, t(80) = 3.62, p < 0.1); 2. Model 1a from Table 3 is not statistically significant (R 2 = 0.176, F (7, 44) = 1.128, p > 0.05), while Model 2a from Table 4 is (R 2 = 0.442, F (7, 44) = 4.191, p < 0.1). In other words, the second hypothesis was only partially confirmed, in the sense that the country globalization only predicts the adoption of Web 2.0 and not the adoption of Web 2.0 for place branding communication.
However, these predictors of Web 2.0 are stronger than the previous ones; 3. The computation of R 2 Change from Model 1a to Model 1b from Table 3 shows that it is not statistically significant, which means to imply that the data does not support the hypothesis that the socio-economic inequality in the country improves the prediction of the adoption of Web 1.0 for place branding communication based on country globalization; 4. The computation of R 2 Change from Model 1b to Model 1c from Table 3 shows that it is statistically significant, which confirms the first hypothesis that the country's geographic position improves the prediction of the adoption of Web 1.0 for place branding communication based on country globalization and on country socio-economic inequality.

Discussion
Through this paper, we proposed a new theoretical framework of the network society as a trope that makes visible the shape of a communication to redefine interdisciplinary online place branding and to answer whether the global imaginary place of information is confined by the local, empirical, geographic space. To answer the research question, we also proposed a category grid helping scholars define and quantify the Web 1.0 and Web 2.0 components of the online presence of place brands through content analysis. In a preliminary pilot study, we found the test-retest and inter-coder reliability of this instrument to be strong. After applying it to analyze the content of 82 official country websites, our results correlated with other country brand indexes, as a measure of concurrent validity. To provide further validity-supporting evidence for our instrument, we framed it through the theory of network society and the concept of 'space of flows' [18,35] and showed how the two analyzed dimensions of place branding communication can be predicted by country-based measures of globalization, socio-economic inequality, and geographic position. Besides proposing an instrument to assess how the public authorities involved in place brand management choose to use official websites to communicate a country's online identity, we also brought arguments supporting three out of the four hypotheses we formulated, reflecting the spatial dynamics of online place brands communication in the contemporary global context.
To sustain an in-depth discussion about the research findings, from the point of view of the visual presentation of place brands, the site must capture the attention of users, creating a "good first impression" [31] (p. 104), by applying: (i) design elements: placing the logo on all pages of the website and implementing in the Homepage a gallery of representative images (e.g., Kenya, Namibia, Guatemala, Slovakia) and interactive maps (e.g., Botswana, Madagascar, Malawi, Aruba, Bahamas, Dominican Republic); (ii) easy identification of the managing authority of the brand and its representatives (e.g., Kenya, Madagascar, Zambia, Japan, Bosnia and Herzegovina). By communicating the most relevant information (i.e., about the place, public announcements, answers to frequently asked questions, personalized documents, depending on the audience), the design of the site must allow quick access to them, with as little effort as possible, using features of their functionality and adaptation to the Web 2.0 environment (e.g., South Africa, Malaysia, Norway, the Netherlands, Slovakia). The effectiveness of the site can be improved by frequently updating information (e.g., Norway, Russian Federation, Rwanda), fast loading of the Homepage, keeping a small size in KB (e.g., Yemen, Bhutan, Botswana, Uruguay), and versions of the site accessible in other languages of international circulation. To increase the time spent online and the return rate of users, more actions must be implemented: (i) by generating customized versions, as accessibility options: by the presence of the "text only" feature for all sites, font size; (ii) providing access and development of user-generated content through the integration of social media services (e.g., South Africa, Seychelles, Thailand, Switzerland, Greece) and the promotion of place brand applications for smart terminals (e.g., Kenya, Guatemala, Panama, Austria, Croatia, Norway, Turkey) and (iii) by implementing features or offering services different from other brands: Live Chat or Info Line (e.g., Aruba, Bahamas, Mexico, Puerto Rico, Malaysia, Poland), electronic newsletter (e.g., Australia, the Netherlands, Turkey, Egypt), videos (e.g., Colombia, Ecuador, Puerto Rico, Greece) or video broadcasts via webcams (e.g., Thailand, Bulgaria, Jordan), or personalized documents depending on the audience (e.g., Botswana, Greece, South Africa, San Marino, Bosnia and Herzegovina).
To answer the research question whether geography is destiny for adopting Web 1.0 or Web 2.0 in place branding communication, our results converge to suggest that geographical position is still important in conditioning the adoption of Web 1.0. This means that the top-down approach to place branding communication (based only on fixed communication contents, such as images, sounds, symbols, interactions, and experiences) is still confined by the local, empirical, and geographic space. However, results show that physical space does not make a significant difference in the adoption of Web 2.0 in comparison with the prediction power of globalization and inequality indicators. Hence, geography is no longer destiny when place branding communication adopts a bottom-up approach, so that the content of communication is dynamic and allows for interactions with various publics. This component of branding communication does not depend on physical space constraints (fixed in concepts such as cities, countries, regions), but on the globalization context and its socio-economic effects. Even so, it should not be ignored that the adoption of Web 2.0 is predicted by a country being in Eastern Asia or the Pacific (this predictor being the strongest one in the overall model), contexts in which physical space is still influential. Our results also highlight that, in comparison to other geographic regions, being a country from Europe or from Eastern Asia and the Pacific are the strongest predictors of adopting Web 1.0 in place branding communication. These two geographic contexts are the ones in which top-down place branding communication is more anchored to physical space concepts, such as cities, countries or regions.
Even so, only the adoption of Web 2.0 in place branding communication follows the model of 'space of flows', being predicted only by dimensions of country globalization, among which WEB empowerment is the strongest predictor, while the model including the geographical position does not improve the prediction significantly. This means that, by knowing the difference that the Internet makes to people and the extent to which its use by stakeholders produces positive change in society, economy, politics and environment, one could predict the adoption of Web 2.0 in place branding communication. In other words, the possibility of configuring a global imaginary place of information for a dynamic and bottom-up place branding communication relies mainly on expanding the capacity of the web to empower its publics.
The socio-economic inequalities among countries condition the adoption of Web 1.0 in place branding communication. It is interesting to note that the strongest predictors of the adoption of Web 1.0 in place branding communication are not the ones related to geography, but to socio-economic inequality, namely the Human Development Index and the low-income economy. Even if geography is still destiny in the adoption of Web 1.0, inequality is also a significant predictor. This finding supports the conclusions of other scholars [61,62] who warned about the potential of the 'space of flows' to reproduce or amplify socio-economic inequalities, despite their potential to liberate from constraints of the local geographic space [18]. The unidirectional place branding communication is limited by the social effects of globalization, such as inequality.
Geography is still destiny for adopting Web 1.0 in place branding communication (especially for European countries), as well as the socio-economic inequality within the country (especially the level of the Human Development Index), while the adoption of Web 2.0 is facilitated by the network society, being predicted only by the dimensions of globalization (especially the level of the WEB empowerment sub-index). It is therefore possible for the theorizations of the advantages of the 'space of flows' globalization to only be applicable for Web 2.0 (or maybe Web 3.0 in the future), as Web 1.0 is still being under the influence of the 'space of place' and its localized socio-economic inequalities. The first transition from a fixed (Web 1.0) to a dynamic place brand communication (Web 2.0) is incremental (conditioned by a place branding communication fixed in geographic concepts) and contextual (conditioned by the access to a global communication infrastructure).
Based on these results, we can theorize farther that the second transition from a dynamic place brand communication (Web 2.0) to a semantic one (Web 3.0) could imply migrating towards constructing new territories of meaning which are not geographically bound nor market-oriented, but detached from localized culturally constructed meanings, and deconstructed into new assemblages. In this way, the geographic distance between physical spaces can be translated into the mathematical distance between various contents so that new territories of meanings can emerge when space is imagined as a language and communication as a movement which can be globally understood.
Moreover, as "Web 3.0 technologies facilitate cooperative processes that are integrative in the construction of new information and meaning" [84] (p. 6), in the future, place branding communication can migrate towards a semantic web (Web 3.0) which is a language territory in which physical distances are distances between meanings. This movement can be imagined as an informational deconstruction in which contents, interactions, and geographic spaces are merged and reassembled to move towards new global understandings beyond individual, situational, and structural constraints.
Furthermore, in terms of the newest web paradigm, according to a study based on the scientific literature published from 2009 to 2017 on Web 4.0, the concept is not clear, being "quite irregular and mutable… contrarily to previous wide-accepted definitions of Web 1.0, Web 2.0 and Web 3.0" [59] (p. 7044). The same author specifies there are five dimensions of this new web paradigm: symbiotic Web; Web of Things; Web of social computing; pervasive Web; and ubiquitous computing. All this enhances the impact of the social networks evolving towards an increasingly interactive web of intelligence networks governed by artificial intelligence but it is an open debate about the development of online place branding communication in this mixed context.

Conclusions
This paper aimed to create an interdisciplinary theoretical framework, based on which we developed a new research instrument, that we applied to 82 place websites in order to answer the research question, and applied the results to expand the proposed framework and inspire developing further research instruments that could continue our work. The novelty of our approach resides in proposing a different understanding of place branding communication by merging theories from communication, marketing and sociology using language tropes as heuristics. In this way, we showed how physical distance (from geography) can be interconnected with the distance between words, interactional agents, and meanings (from communication) to configure an abstract space inspired by the theory of the network society (form sociology of space) and taking into account already existing empirical effects (place websites developed by marketers and inequality indexes used by social scientists).
This study proposes an interdisciplinary theoretical approach for extending the marketing literature on place branding communication by incorporating the sociological literature about the network society and its 'spaces of flows'. It can lay the ground for a better understanding of the future transition to Web 3.0 in place branding communication. In this regard, we developed a theoretical framework in which the network society can be understood as a trope that makes visible the shape of a communication process between the local, empirical, geographic space and the global imaginary place of information. We assumed place branding communication as a prototype of the network society, chosen because the link between physical space and information place is already established in our local and global imaginaries. Based on this framework, we asked whether the global imaginary place of information is confined by the local, empirical, geographic space, and answered that geographical and socio-economic inequalities confine only the adoption of top-down approaches (therefore the process is incremental) and not the transition to bottom-up ones, conditioned mainly by social, economic, political and informational dimensions of the globalization (therefore, the process is contextual).
The main practical contribution of this study is to propose an instrument to help scholars identify and quantify the Web 1.0 and Web 2.0 components of the online presence of place brands and their spatial dynamics in a broader global context. We tested the four hypotheses to show the predictive validity of our instrument. However, a different geographical country categorization could be taken into account in the future, as well as other globalization and socio-economic inequality indicators.
Some key limitations need to be considered. First, due to using hierarchical regression analysis, several questions remain unanswered, particularly ones related to the complex interconnection of the measured concepts (place branding communication, globalization, inequality, and geographic position), in lack of testing for mediation or moderation. For future investigations, it might be suitable to apply structural equation modeling. Secondly, the generalization of these results is questionable, as the sampling procedure followed the country selection method used by the World Tourism Organization [68], which might be considered limitative. Similarly, our grid could be extended to incorporate the Web 3.0 dimension so that longitudinal analyses could be conducted in the future.