Infrastructures. The Digital Orientalist’s 2022 Conference

The Digital Orientalist’s conference is approaching. Join us on June 25th for a series of talks presenting new ideas and projects, around the theme of “Infrastructure.” Below are listed the titles and abstracts for the papers that will be delivered on June 25th, for the DO 2022 conference on the theme “Infrastructure.” Speakers are listed in alphabetic order. 

You can register here. An updated schedule will be published soon.

Invited speakers


Ya-Hwei Hsu, Academia Sinica. Social Changes of Liao, Song, and Jin Dynasty as Seen through Tombs 
The Tang-Song transition theory has stimulated much debate and prompted scholars to reassess the social, cultural, and economic change/transformation following the collapse of the Tang dynasty. Along this line of discussion, the fact that the Song dynasty first coexisted with the Kitan Liao dynasty and then the Jurchen Jin dynasty has often been overlooked. The Song occupied South China ruling mainly the Han Chinese while the Liao and Jin’s territories stretched from North China to today’s Mongolia and Siberia populated by peoples of different ethnic origins. The environmental condition led to the contrasting lifestyles of the South and North, i.e., sedentary vs nomadic. The political systems they implemented were also different. The changes observed in the Song dynasty may not be applicable to the Liao and Jin. Moreover, the dynamics between the South and North regimes might have played a role in those changes observed and explained under the Tang-Song transition paradigm. To address this cross-regime, structural issue, we need a comparative research of the Lian, Song, and Jin. With this goal in mind, we design a database to digitize Liao, Song, and Jin tombs excavated by archaeologists in a comprehensive and systematic way. The information recorded includes the tomb structures, mural and carving decoration, and all burial goods. In the talk, I will demonstrate a comparative study of the Liao/Northern Song and that of the Southern Song/Jin by looking at the ways in which epitaphs were used in tombs of different social strata.

Jann Ronis, Buddhist Digital Resource Center, Boston. Title TBD. 
Daniel Stökl, École Pratique des Hautes Études. Title TBD. 
Kaitlyn Ugoretz, University of California Santa Barbara. Do We Have the Technology?: Living Into Infrastructural Imbalances to Rebuild a Better Academy.
 
Griet Vankeerberghen and Raja Sengupta, McGill. Representing the Mobility of Historical Agents through Digitization and Topographic Path Finding: the Diary of Shi Rao (d. 10 BCE)
Shi Rao, a low-level official who served in the commandery offices of Donghai, left a diary in his tomb that includes much information about his whereabouts over the course of the year 11 BCE. Our presentation will seek to reconstruct, as precisely as possible, how Shi Rao traveled through the landscape of what is now northern Jiangsu and southern Shandong. In the centuries BCE, the natural landscape in the area (esp. the course of rivers) was substantially different from how it presents now. It was also, already by Shi Rao’s time, a landscape with a long history of human intervention, with many old settlements and a heavily traveled road network. Using Tan Qixiang’s atlas, that depicts the situation based on records that date variously to 8 BCE and 2 CE, we digitized the river networks, and added modern topographic data. This allows us to look at how the landscape could have impacted Shi Rao’s travels.  In the presentation we will use Agent Based Modeling and Least Cost Path analysis to determine the likely routes of Shi Rao’s travel; in addition, we will reflect on the modes of transportation (boat, carriage, horseback) that were most appropriate for various segments of Shi Rao’s travels. This presentation is part of a larger-scale project on mobility in Western Han (201 BCE–8 CE).

Participants


Elizabeth Bishop, MECAM/Texas State University
Mapping African Students’ Organizations via ArcGIS: Paris, 1955 
Did Muslim students from North Africa, learn anything from the experiences of students from Sub-Saharan Africa who were in Paris at the same time? Theoretical and methodological tools developed by geographers to analyze spatial formations and spatial patterns reveal unexpected correlations in historical documents regarding African students’ organizations in Paris during 1955-1960. Inspired by Ali Rahnema’s monograph (2014) and his use of maps from Tehran to narrativize crowds’ sequence of movements (and the institutional factors behind such movements through public space), and building on “Behind the 1956 Algerian Strike” (Bishop, Chow, and Lasode 2021) this study responds to this challenge with an integrated analytic approach. UGEMA member Zahir Fares’ memoir, Afrique et Démocratie (1993/2000) devotes an entire chapter to African intellectuals in Paris, and how they supported Algeria’s nationalist struggle for independence through the UGEMA, L’Union Generale des Etudiants Musulmans Algeriens. I compare Zahir Fares’ memoir with Mohamed Harbi’s recently-recorded interviews (2021), Belaïd Abdesselam’s and Amady Aly Dieng’s monographs (both 2011), and that of Houari Mouffok (1999), consider these in the light of Diop’s ” student avant-garde” (1997, p. 114); and Adams’ assertions regarding Christian student groups (1998, p, 63). Drawing on newspaper reports, oral history interviews, still photographs, and newsreels with data gathered from archives on three continents, this ArcGIS Story Map draws as well on current aerial and satellite views of the Cité Internationale Universitaire de Paris and other key locations, contributing to the development of digital humanities as well as structuring one chapter in a forthcoming monograph.

Marie Bizais-Lillig University of Strasbourg, Benjamin Guichard BULAC, and Adeline Joffres Huma-Num
HUMA-NUM consortium DISTAM, or How a national infrastructure supports a DH-structuration effort in area studies
Successive French governments have supported the development of shared infrastructures to promote open access academic publishing (OpenEdition), digitization of academic collections (Persée) and digital humanities (HUMA-NUM). With regard to the HUMA-NUM infrastructure, it has set up a grid of generic services accessible to any researcher in France as a means to simplify the management of FAIR data. The grid includes Nakala repositories with DOI, IIIF services, Heurist and Omeka-S databases among other services. In order to meet the needs of specific communities within the large spectrum of the humanities, HUMA-NUM welcomes initiatives from groups of researchers and data scientists who want to establish consortiums. 12 consortiums are presently supported by HUMA-NUM. These consortiums are meant to structure and strengthen academic communities who face precise issues when applying DH methodologies in their specialized fields. The Digital Studies Africa Asia Middle-East (DISTAM) consortium was created in January 2022 to cope with specific questions in area studies, for example: text mining and text recognition for non-latin scripts; broader access to multilingual catalogs, corpora and metadata in digital libraries; management of geographical data; application of open science principles to data from culturally or politically sensitive fields. This presentation will bring together an actor of HUMA-NUM and two members of DISTAM. Together, they will underline how different levels of structuration help spraying DH through academic communities. They will show how the goals of a nationwide infrastructure are carried by a smaller group and how this group promotes common tools, shares methods and trains a community.

Wanyun Cao, University of Edinburgh
Social Media Ranking and Reactivity: based on Weibo Celebrity Power Ranking
Celebrities are often considered to be highly influential and visible, and their buzz often acts as an engine of commercial value for the entertainment industry. As such, quantifying and ranking the achievements of this group has been an important topic of recent research. Among the many celebrity rankings produced by specialist media based on a relatively single measure such as box office performance and number of plays, social media-generated celebrity rankings stand out with the help of the complex algorithmic logic of the platforms and the data labour of fans. This article extends the reactivity framework used to explain the mechanics and impact of university rankings to the realm of social media rankings, arguing that the concept of reactivity is also at play in social media rankings and has reshaped the entertainment industry in China. After a brief description of the history and measurement of the Weibo Celebrity Power Ranking, I explain the adaptation of reactivity in social media rankings and how it differs from traditional ones, which includes a discussion of two powerful mechanisms and three important effects. This framework shows how the platform’s algorithmic logic is changing the norms of online fandom practice and influencing the values of the entertainment industry, suggesting why it is important for scholars to more carefully investigate, and study the impact of social media generated assessment systems.

Joseph Dennis, University of Wisconsin-Madison, Department of History
The Construction and Uses of the Books in China Database
This paper introduces the Books in China Database (BIC), a web tool created by Joseph Dennis of the University of Wisconsin Department of History, and Dr. Chen Shih-Pei and Calvin Yeh of the Max Planck Institute for History of Science (MPIWG). BIC is an open-access research and visualization tool that allows users to explore the circulation of books in China from the 1200s to the early 1900s. BIC is based on over 32,000 lines of data about books and other texts held by over 700 local school libraries and government offices. Dennis extracted information from Chinese local gazetteers using the MPIWG-developed software Local Gazetteer Research Tools. Most data comes from lists of books in school libraries and commemorative records of book acquisition. Each line of data is based on one book that has an associated latitude, longitude, date, and storage location. Many records contain information on method of acquisition, cost, donor name and position, etc. The data set became the back end for BIC: https://www.booksinchina.org/bicdb/#/ With this tool scholars can find and visualize information related to the circulation of books in particular times and places in China. The tool has chronological and spatial visualization tools and a comparison function that allows exploration of titles relative to each other. The web site is now online and the documentation and user’s guide will be finished by June 25, 2022.

Tuğba Zeynep Şen Boğaziçi University Social Policy Forum, and Engin Onukİstanbul Bilgi University
Vernacular Discourses of Transnational Online Misogyny: Studying Turkish Incels
Rapid advancements in communication technologies have made the diffusion of concepts and ideas easier than ever. Online communities cut across borders and cultural contexts, and in many cases create vernacular and cultural hybrids. One case in point is the Incel community and its Turkish incarnations. The term Incel refers to ‘involuntary celibates’, (mostly) young men who have become disillusioned with modern dating due to their inability to have successful romantic or sexual relationships. This is considered a result of women gaining the upper hand in social relations because of feminist movements, which are deemed to be the detriment of society. Frequently linked to misogynist rhetoric and violent tendencies, Incels have been the focus of increasing scrutiny and censure, including the banning of all related subreddits on Reddit. These communities have developed an extensive system of in-group cues and terminology, almost exclusively developed within the English-language communities. This jargon has been adopted almost verbatim by the Turkish community, creating a system of reference which is both deeply contingent on a very specific context and very decontextualized for members of the out-group by being transplanted into a foreign language, creating two layers of impenetrability for researchers and readers. This presentation will explore the difficulties of working with online corpora in non-English languages which are so inextricably linked to its foreign counterparts and the obstacles this presents for the study and dissemination of online phenomena for and within local scientific communities and audiences.

So Miyagawa, National Institute for Japanese Language and Linguistics
Digital Curation of Resources Related to Endangered Languages and Dialects in Japan 
This presentation will discuss the construction of a digital archive for endangered Japanese dialects and languages; for example, the Ainu and Ryukyuan languages. This digital archive is based on the Kotoba no Museum (Museum of Languages) project at the National Institute for Japanese Language and Linguistics (NINJAL) and designed according to the framework principles of FAIR (findable, accessible, interoperable, and reusable). To realize these principles, it effectively leverages the formats regarded as global standards in digital humanities. This presentation commences with an overview of the history and developments of the Text Encoding Initiative (TEI), the global standard for text structuring of textual materials; the International Image Interoperability Framework (IIIF), the global standard for interoperability framework for image materials; and the Dublin Core, a metadata standard. It progresses to discuss the use of IIF to display dialect-related maps, such as Nihon Gengo Chizu (Japanese Language Atlas), and TEI to markup materials with dialect glosses and dictionaries. We subsequently present a digital archive of Japanese crisis dialect-related materials currently under construction using Omeka S, a content management system (CMS) provided by the Roy Rosenzweig Center for History and New Media (CHNM) at George Mason University in the United States. The Omeka S promotes efficient collaboration to develop a digital archive using global standards such as TEI, IIIF, and Dublin Core. We examine how these global standard formats can be applied to existing tools and the contribution the digital archive can make to language preservation, education, and linguistic research efforts.

Sruthi V S & Annie Johnny
Indian Cultural Curations- Diving into the Archive of Heritage in the Digital
This paper argues that some Instagram pages act as artistic research outside academia and could also be considered a source material for digital humanities. We will be conducting a critical analysis of Instagram pages involved in participatory heritage curation of collective memory across two themes – cities and communities of India. Additionally, a textual analysis of captions, hashtags and comments on Instagram posts will also be undertaken. This is to determine how these pages are creating a larger digital historiography of the selected themes. These pages collate information that exists in a space which is not bound by academia or and disciplinary boundaries. The interest generated by these pages makes them more than just ‘creative curations’ as it provides significant primary and secondary information to followers regarding the featured objects/monuments. Often the page owners also source details from relevant books, academic papers and other relevant books and online resources bringing out at times even subaltern voices. An example is ‘angloindiastories’ (https://www.instagram.com/p/BpnoyHKltGD/) – a collection of crowd sourced photographs that are creating an archive, a narrative about Anglo- Indian history, a community that rarely has a visual presence in Indian cultural memory. Similarly, ‘heritage_haryana’ (https://www.instagram.com/heritage_haryana/) is a visual repository of known and unknown heritage monuments in the North Indian state, Haryana. The state is mostly known for the ‘millennium city’ and tech hub Gurugram (earlier Gurgaon) where some of the largest Indian corporate offices and local offices of international companies are located. Although the state has well known archaeological sites such as Rakhigarhi (Indus valley civilization site), crowds are usually drawn to holy places like Kurukshetra, known as the ‘Land of the Bhagavad Gita’. This page therefore gives much needed insight to heritage sites that lay unprotected and unresearched. Archiving or curating online used to happen through blogs and dedicated websites. However, they pale in comparison to the reach and influence of social media sites. Unlike the restrictions such as the ‘no touch’ policy at traditional art institutions, social media ‘curations’ provides the freedom to engage with the object. Hence, creating a space for collecting even that information that may usually slip away in interstices.

Jia Xie and Suk Man Yip, Heidelberg Centre for Transcultural Studies, Heidelberg University 
Making Ground Truth for Republican Chinese Newspaper: Taking Jing Bao as an example
Many researchers have explored the use of machine learning for optical character recognition (OCR), particularly in Europe and North America, and many projects are producing ground truth (GT) data for this purpose. This is different when it comes to non-Latin script (NLS) material. The Early Chinese Periodicals Online (ECPO) project at Heidelberg University has started to work on ways to produce machine-readable full text from historical Chinese newspapers.  ECPO makes use of different machine learning approaches including convolutional neural networks to develop a semi-automatical pipeline to produce machine-readable full text. We chose the entertainment newspaper Jing Bao (The Crystal, 1919-1940) as basis for our experiments.  Our presentation focuses on two main aspects: Firstly, we provide a description of our ground truth editing work. This includes assembling the editing team, organizing the workflows, establishing processing regulations, and ensuring quality control. Secondly, we discuss particular challenges in producing the GT sets, including issues in character encoding and problems with variant characters related to Unicode, and problems related to segmentation. We were able to produce two ground truth data sets comprising segmentation data (geometry GT) and textual/structural data (full text GT). We hope our experiences from the project can be helpful to others working with NLS material. Based on our work, we can point to some pitfalls and provide hints to avoiding them, making the use of machine learning more efficient.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s