Panel Outline

Media practice today is not confined to a specific medium, platform, topic or language. In times of intense “media convergence” (Jenkins 2006), of the “meta-medium” computer (Manovich 2013), and of ubiquitous digital networks, digital culture appears to happen everywhere at the same time. Videogames are streamed on Twitch.tv, with the support of donations by “patrons” on respective crowdfunding platforms, next to streams of the U.S. congress, which are commented on in the same manner as game play scenes (see Figure 1 and 2).

Media practice appears more dispersed and transgressive of media boundaries, of national boundaries, and of language boundaries. Platforms can be regarded as attempts at providing clarity. At the same time, they restrain, channel and control information and user practices (Stalder 2016; Itō [伊藤] 2016). In this sense, contemporary media practice is both unbound and highly controlled. On this basis, one of the future challenges for digital humanities is how to theorize, capture and analyze such media practice adequately. This is a challenge for media theory and methodology, and one that requires new research infrastructures and tools. Focusing on digital games and Japanese visual media, this panel aims to discuss these challenges and suggest avenues for solving them. The panelists present their own approaches to making sense of contemporary digital media culture, and of making use of various available technologies and tools from the field of data science. Paper 1 discusses the possibility of modeling the heterogeneous data resources and using linked data technologies for making them accessible to human and computer-aided analyses. Paper 2 analyzes large sets of fan-created data for revisiting Hiroki Azuma’s assumptions about the creation of otaku characters in Japanese anime, manga and games. Paper 3 explores possibilities for understanding language-based differences in game culture, comparing text data from a wide range of game cultural data resources.

In combination, these contributions can hopefully present a crucial problem that contemporary media research is facing, and relate our search for solutions to the work of other areas of digital humanities that face similar issues.

(Kazufumi Fukuda) For the Development of a Data Model for Tracing “Play” with Video Games and the Media Culture

In considering the current video game culture, it is hard to ignore the existence of other related media. One of the great examples is the case of “Dark Souls 3” (FromSoftware, 2016). Given the game’s difficulty, large game world, and the variety of gameplay it offers, arguably, few players have never seen the YouTube video of the game. At the same time, players will often open strategy books and browse strategy sites as well. Interestingly, this game has an online competitive play element. Therefore, some players may send information and communicate with other players via SNS such as Twitter. Thus, it can be said that the current video game play experience exists among video games and related media. These are topics that have been discussed with keywords such as media mix (Steinberg 2012) and convergence culture (Jenkins 2008). We think that various resources can be linked by constructing a data model that defines multiple media via video game works. Numerous instances of them already exist on the Internet, either as user-generated contents or as metadata for materials held by institutions around the world. By connecting these resources, it becomes possible to analyze a wide range of play traces occurring in video games and related media. The purpose of this study is to discuss future directions for constructing such a data model and a research platform based on it.

Data models for describing video game materials and works have been discussed by many scholars since the beginning of the 2010s (i.e., McDounogh et al. 2010, Jett et al. 2016, Author et al. 2018, Hoffmann 2019). Besides, some researchers (Aarseth 2014, Juul 2016) examine possible ontologies to define game characteristics. However, the existing works take different approaches. The former adopts a model construction approach that assumes the creation of bibliographic data and its adaptation. On the other hand, the latter is a discussion about the “character of the game” and the “mode of existence” based on the framework of the existing philosophy, and does not take data creation into consideration. Based on this difference, it is clear that a future subject for game studies with data models is the inclusive connection between a bibliographic data and the philosophical framework the author mentioned above.

Prior work has also been focusing on creating metadata for an entity/relational data model. These data are published in the online catalog "RCGS Collection (Fig 1)" owned by Ritsumeikan Center for Game Studies. Moreover, the data were contributed to "Media Arts Database (Fig 2)" known as a comprehensive catalog service of media arts provided by the Agency for Cultural Affairs. In the RCGS Collection, linked data are published as dump data. In addition, SPARQL endpoints are provided, enabling data retrieval and analysis using a query language. Also, this dataset gets some of the authoritative data of the work from Wikidata (Author 2019). Wikidata is characterized in that it is connected to many external sources. Thus, it can be also said that their meta data contribute to the connection with external related resources.

Finally, we would like to discuss the problems and methodologies necessary to build a data model to properly describe the game-centric media culture mentioned at the outset. As a major premise, it is necessary to establish a scope in order to select resources to be modeled, because there is a huge amount of media related to games. To take shape in our research, the case study approach may be most effective, because it allows us to collect various materials related to the playing experience from a selected game work. By analyzing these materials, we can extract requirements and develop models. It is preferable that multiple cases be selected, and that they cover as much “ground” of what we perceive as game culture as possible.

In conjunction with the case studies, a more specific designation may occur when discussing the relationship with the game. Therefore, it is also necessary to consider a model for designating the part of the content that constitutes the game work itself.

In this paper, we suggest solutions for these challenges and point at further work to be done in the future.

(Zoltan Kacsuk) Revisiting “Otaku: Japan’s Database Animals” twenty years on: Can we use fan created databases to test one of Hiroki Azuma’s central arguments?

Media fans have been compiling data on their favorite works long before the internet became the most important channel for communication (Jenkins, 1992; Okada, 1996; Yoshimoto, 2009). Naturally, with the advent of the internet this process of collecting and cataloging information by individual fans and enthusiast communities has only become easier and amplified in its scale. The level of detail afforded by these fan compiled data sources has also not gone unnoticed in academic research (cf. Hills, 2002). Although various online databases created by enthusiast communities have become the go to resource for checking information on hard to find media texts and artifacts, their use for large-scale quantitative research has yet to become widespread.

Harnessing the power of fan created databases for academic research is precisely the aim of the Japanese Visual Media Graph (JVMG) project, funded by the German Research Foundation’s (Deutsche Forschungsgemeinschaft, DFG) e-Research Technologies program. By learning from and building on the experiences and content produced and aggregated by enthusiast communities, the project aims to create a database and query tool primarily for academic researchers working on Japanese visual media. Furthermore, the JVMG project employs a linked data framework for processing the various data sources and for creating links between them. This approach not only allows for a great level of flexibility in data processing, but also facilitates the linking of the data to even further sources, especially linked open data. Although the project is still in its development phase it has already proven to be a powerful resource for approaching both new and long-standing questions in research on Japanese anime, manga, light novels and video games. To demonstrate the possibilities opened up by such an integrated database as well as the challenges entailed in working with data compiled by enthusiast communities this presentation will focus on examining one of Hiroki Azuma’s most well-known works through the lens of data on visual novels (a unique kind of narrative centered video game genre) and other types of Japanese visual media, such as anime and manga.

Hiroki Azuma’s Dōbutsu ka suru posutomodan: otaku kara mita Nihon shakai (Animalizing postmodern: Japanese society as seen from otaku), published in 2001, has been one of the most influential treatises on not only Japanese otaku, but also on the production and consumption paradigm defining Japanese anime, manga, light novels and video games in late modernity. The book’s impact on the discourse around otaku and the just enumerated domains is truly international thanks in part to the English translation, which was published in 2009 as Otaku: Japan’s Database Animals. With almost twenty years since the original publication in Japanese and more than ten years since the English translation was released, the concepts and fra to the lack of these types of quantitative research approaches in the concerned fields, however, the lack of available data that is both large and granular enough to use for such an analysis has also not been conducive to the emergence of quantitative examinations of this sort.

This presentation will thus both introduce the wider JVMG project, and provide an example use case showcasing the JVMG database’s affordances and limitations by examining one of the central points from Azuma’s above cited work – namely that “many of the otaku characters created in recent years are connected to many characters across individual works, rather than emerging from a single author or a work” (2009 [2001]: 49) – with the help of the database. The use case will serve to illustrate certain inherent limitations of databases compiled by enthusiast communities, and will also highlight how the JVMG database with its incorporation of multiple such databases can serve to help mitigate some of those shortcomings.

(Akito Inoue) Can we measure cultural differences in video games? Awareness of the problem from previous research

The relation between human consciousness and "language" has been a major point of dispute in social and human science. In what sense can we say that human consciousness also bound by the "games" it plays?

As a human behavioral environment, "playing video games" has several interesting characteristics. Compared to linguistic behavior, video games may be less affected by the "vocabulary" or "range of meaning" formed by a specific region. For example, if playing the same game, both African and Japanese children may have similar experiences. There are probably quite a few cases where the game is played in an identical way.

Based on this assumption, can we also say that the is game completely free from social-cultural differences? Most often, the answer can be “NO”, for the following two reasons. Firstly, games are localized and the content slightly differs from country to country (O'Hagan, M., & Mangiron, C., 2013). Secondly, even when playing the same game, research (i.e., Feng.at al, 2007) shows that there are usually social differences between players that affect the gameplay. For example, when men and women were asked to play the same game, a difference in gameplay tendencies was found in many studies (i.e., Linda Hughes, 1983, Feng.at al, ibid).

Although there is not yet much research on differences in gameplay by region, the RPG Studies suggest that there may be a significant difference in the strategies of raising the game character level (Endoh, 2019) based on different social environments. Cultural differences in gameplay can be presumed to exist because of the following three approaches that have been applied to measuring cultural differences concerning games:

(1) Differences in different versions of the same work (O'Hagan, M., & Mangiron, C., 2013)

(2) Differences in gameplay in different play environments of the same work (Pelletier-Gangon, 2019)

(3) Differences in the same version and environment of the game (Feng.at al, 2007)

Based on the above, our study aims to further develop awareness of the problem of (3). Particularly, we would like to clarify the cultural differences in games, especially those found in different language areas.