TS. Tijdschrift voor tijdschriftstudies. Jaargang 2013

[p. 73]

TS Tools

How to design a database of historical periodicals that can be used for research and teaching? In this section, Liying Sun and Matthias Arnold (members of the Research Cluster ‘Asia and Europe in a Global Context’ at Heidelberg University) present a newly established database of Chinese women's magazines.¹

Fig. 1: Home page of the database.

1 The website can be found at womag.uni-hd.de. For full metadata access and larger images please contact the project at [e-mail]. Sincere thanks to Prof. Barbara Mittler, Prof. Joan Judge, and Doris Sung for their invaluable input and to Maarten van Gageldonk and Maaike Koffeman for their help with the revision of this article.

International, interinstitutional, interdisciplinary: a history of the project

The idea to create the database ‘Chinese Women's Magazines in the Late Qing and Early Republican Period’ (or ‘WoMag’) developed from the research project ‘A New Approach to the Popular Press in China: Gender and Cultural Production, 1904-1937.’² Funded by the Canadian Social Sciences and Humanities Research Council and the German Humboldt Foundation, it brought together a group of established researchers from

2 For more information please refer to www.yorku.ca/ycar/womag.html.

[p. 74]

various academic fields. They formed a network of scholars, whose aim it was to combine different disciplinary approaches and develop new methods to examine Chinese women's journals. The team developed what they have begun to call ‘horizontal,’ ‘vertical,’ ‘integrated,’ and ‘situated’ readings of journals. By horizontal reading they mean close reading all material (text, images, advertisements) included in one issue or a series of issues of a particular journal. Vertical reading refers to studying one particular theme across different journals. Integrated reading means reading journals against other journals, for instance within the same periodical genre. Finally, situated reading integrates the study of other source materials, e.g. biographical or historical sources. A forthcoming edited volume will provide a variety of in-depth essays applying these methods to particular case studies.³

Fig. 2: The page to display magazine covers for Linglong (1931-1937).

3 Joan Judge, Barbara Mittler and Michel Hockx eds., A Space of Their Own: Women and the Periodical Press in China's Long Twentieth Century, forthcoming.

Conceptualization and database features

The group of scholars selected four important women's journals as research objects, Nüzi shijie (Women's World, 1904-1907), Funü shibao (The Women's Eastern Times, 1911-1917), Funü zazhi (Ladies’ Journal, 1915-1931), and Linglong (Elegance, 1931-1937, fig. 2). These journals cover an important period of socio-cultural change from the 1910s to the 1930s, and reflect the rise and development of a gendered periodical culture. In order to better read the periodicals through the above-mentioned methods, the scholars required a digital tool which would not only store the material, but also provide substantial metadata and cross-referencing options. Originally conceived as a research guide

[p. 75]

with a collection of indexes supplementing the printed volumes, the WoMag database became a collaborative online editing platform and research tool. From the project's methodological perspective, it was tailored as a tool that would facilitate both the horizontal and - as far as possible - vertical, situated and integrated reading.

Intellectually supported by the research project, the database was conceptualized and developed by the Heidelberg Research Architecture (HRA) with additional funding by the Cluster of Excellence ‘Asia and Europe in a Global Context.’ Its implementation consisted of three phases. The first was concerned with the data structure, starting with collaborative recording of metadata in spread sheets. We recorded as much information as possible, including authors and article titles, images and captions, advertisements and illustrations, front and back matter. This holistic approach deepened our understanding of the content and mediality of the periodicals. These pre-structured data were analyzed and translated into the conceptual design of a relational database, which was then implemented in PHP/ MySQL on a Linux server. Once a stable back-end was established, new data could be added directly into the system. The second stage focused on optimizing the visual data, aiming to present complete issues. Implementation of a front-end and optimization of usability and editing workflows formed the third stage. For more than a year new data were uploaded. A number of advanced editing tasks required the development of special interfaces. For example, we needed to rearrange pages according to the original printed page order. Since not every issue contains a complete set of pages, we wanted to be able to indicate where a page was missing or intentionally left blank (fig. 3). Other interfaces were added for assigning category labels (e.g. the article genre in Chinese characters, Pinyin transcription, and English translation (fig. 4), and organizing the workflow.

Fig. 3: Back-end interface to arrange pages, fill missing or blank pages, and assign copyright information.

[p. 76]

Fig. 4: Back-end interface to add or edit ‘article genre’ in Chinese, Pinyin transcription, and English.

Fig. 5: Search result for ‘body’ across all magazines within articles only, sorted by Chinese title.

[p. 77]

In their article ‘The Rise of Periodical Studies,’ Latham and Scholes present five important guidelines for the digital archiving of periodicals.⁴ Both in terms of its conceptualization and the realization of its functionalities, the WoMag database responds to their suggestions. We have used original issues whenever possible. The database presents images of all pages from cover to cover; if a set of a magazine is not complete, we have either tried to find the missing pages through interinstitutional cooperation, or inserted a mention at the location in question. We included an individual section for advertisements, collecting metadata such as titles and headings, keywords, product categories and trademarks. We recorded the verbal parts of advertising in full text (usually in Chinese) and made it searchable. And finally, the database displays the number of hits for each query, and highlights the searched words in the results (fig. 5).

4 Sean Latham and Robert Scholes, ‘The Rise of Periodical Studies,’ PMLA 121-2, 2006, 517-531 (524).

Research and teaching

The database was designed for both research and teaching. Each of the four journals has been studied by both senior and junior scholars. Having extensively worked on the Funü shibao for her forthcoming book Republican Lens: Everyday Knowledge, Gender, and the Periodical Tress in China, Joan Judge used the database to conduct multi-layered readings of everydayness in the 1910s and published several articles on the subject. Barbara Mittler examines how gender relations are reflected in media in the long twentieth century, and the WoMag database provided her with rich visual sources for her study on tropes of the New (Wo)men in Chinese women's magazines. As part of her dissertation entitled Redefining Female Talents: Women Artists in Republican China, Doris Sung examined how the gendered journals opened up a new public space for women artists in the first three decades of the twentieth century. Liying Sun focused on the transcultural journey of the Western nude photographs published in Linglong as part of her dissertation research entitled Body Un/Dis-covered: Luoti, Editorial Agency and Transcultural Production in Chinese Pictorials (1925-1933).

Combining her experience in constructing the WoMag database and her research interests, Sun designed two related courses for master's students. One introduces the four methods of reading periodicals by taking Linglong as an example; the other trains students in analyzing periodicals by filling out the various database fields. Each field corresponds to a question developed by the researchers. Both courses placed the WoMag database at the center of the teaching and learning activities. The students enjoyed using the database, and their feedback in turn has greatly helped us to improve it.

Current status and future plans

Since its launch in 2009 the database has grown substantially. Currently, it comprises some 453 issues from four magazines, adding up to 55,644 pages in total, including 5,333 ‘missing’ or ‘blank’ pages. Metadata of 8,989 articles, 8,322 images, and 3,787 advertisements were recorded, 5,727 persons were identified and assigned to 14,416 items. We

[p. 78]

have defined 330 keywords and assigned them to items 13,209 times.⁵ Although still far from comprehensive, the database already offers new insights to periodical scholars. Since large parts of the metadata are provided bilingually, a broader audience can use the system for comparative purposes.

During the past three years, we realized that generating high quality metadata not only required substantial funds and manpower, but also an interdisciplinary dialogue. The commitment to collaborative research and database development was crucial to the success of our project. One of the exciting ‘by-products’ of the WoMag database is the large number of historical ‘actors’ it brings to the fore - persons, groups of persons, and organizations, many of them hitherto unknown. We recorded names and name types (e.g. ‘pseudonym’ and ‘variants’) and listed their respective roles (‘publisher,’ ‘editor,’ ‘author,’ ‘painter,’ ‘photographer,’ ‘calligraphier,’ ‘mentioned in article,’ ‘depicted on image,’ and so on). We have also recorded events and additional biographical data as they appear in the magazines. By relating a person's name to events, social activities and portraits, we hope to be able to map biographies on a larger scale, especially those of people who are absent from existing lexica. The systematic tracing of a selected group of more influential agents will be the focus of a follow-up project.

Indeed, the WoMag database is developed further in a new project entitled ‘Early Chinese Periodicals Online’ financed by the Chiang Ching-Kuo Foundation. The three-year (2012-2015) international collaborative project between members of Taiwan's Academia Sinica (Yu Chien-ming, Lien Ling-Ling, Hu Siao-chen) and the University of Heidelberg (Barbara Mittler) is now devising a larger system based on the model of WoMag database, which aims to present more than 200 additional gendered journals and entertainment newspapers as well as their metadata. The WoMag database will be at the core of this vast, comprehensive project.

•> liying sun and matthias arnold are members of the Research Cluster ‘Asia and Europe in a Global Context’ at Heidelberg University. Project leaders are joan judge, barbara mittler and grace fong. Project members doris sung (York University) and annika joest (University of Heidelberg) also made crucial contributions to the founding and development of the WoMag database. The programming was conducted by wolfgang häberle.

5 Statistical figures as of February 18, 2013.