Humanities and Social Science Research; Vol. 8, No. 4; 2025 ISSN 2576-3024 E-ISSN 2576-3032 https://doi.org/10.30560/hssr.v8n4p108 108 Published by IDEAS SPREAD Audience Labor in Chinese Social Media: A Case Study of Film Reviewers on Douban Yucong Li1 1 Macau University of Science and Technology, Macau Correspondence: Yucong Li, Macau University of Science and Technology, Macau. Received: July 24, 2025; Accepted: August 4, 2025; Published: August 5, 2025 Abstract With the rapid development of internet technologies, the new media industry has swept into people’s lives at an astonishing pace, bringing about significant changes in global labor relations. Drawing on the theory of digital labor, this article examines audiences within China’s social media by focusing on film reviewers on Douban as a case study. It reconstructs the labor process involved in writing film reviews and explores the issue of autonomy within this context. The study finds that while using Douban, film reviewers invest substantial time and effort into producing cultural content for the platform. In this process, although they appear to possess a certain degree of autonomy and choice, film reviewers are in fact subject to platform control and exploitation. They are often compelled to navigate a delicate balance between adhering to platform rules and asserting their individual agency. Keywords: digital labor, audience labor, movie reviews, autonomy 1. Introduction With the rapid advancement of internet technologies, the new media industry—characterized by its strong penetration and wide-reaching influence—has profoundly transformed people's everyday lives and modes of social interaction. Emerging media forms such as social networking sites, short video platforms, and online communities have become essential arenas for information acquisition, opinion expression, self-representation, and social engagement. At the same time, the reshaping of labor by digital platforms has drawn increasing attention from the academic community. Research on the transformation of global labor relations is flourishing, and the group referred to as “digital laborers” has garnered growing scholarly interest. Researchers have begun to focus on ordinary users who engage in content creation on digital platforms, seeking to reconceptualize the meaning and boundaries of labor in the digital age by examining its processes, value production, mechanisms of control, and degrees of autonomy. Douban is a Chinese platform that offers recommendations, reviews, and price comparisons for books, films, and music albums, as well as information about urban cultural life. As one of the most representative Web 2.0 websites in China, Douban aims to help users discover content suited to their interests and connect with like-minded individuals. Douban Movie, a key component of the platform, serves as the largest online community for film sharing and reviewing in China. It attracts a wide range of film enthusiasts who not only consume content but also actively contribute to the platform's development and maintenance. These users engage in activities such as rating films, writing short reviews, tagging keywords, posting updates, composing in-depth reviews, and participating in discussion groups. On the film rating and review system—the most popular feature of Douban Movie—even newly released films often feature dozens of detailed reviews that generate significant user engagement and discussion. Focusing on the audience labor groups embedded within China’s social media, this study selects film reviewers on Douban Movie as its primary research subject. By investigating the specific processes through which users compose film reviews on the platform, the study further explores the question of autonomy in digital labor. The study raises two core research questions: (1) How do Douban users write a film reviews? (2) How much autonomy do film reviewers possess in the process of writing reviews? 2. Literature Review In fact, the phenomenon of audience labor in the media predates the advent of the Internet. As early as 1977, Dallas Smythe proposed the concept of the ‘audience commodity’ arguing that television viewers are unpaid laborers who are sold as commodities to advertisers [1]. This perspective has gained significant traction over the past decade. In today’s digital landscape, the interpretation itself has become a form of labor, increasingly mediated and expressed through digital media. Theorizing digital labor—understood as labor that either produces or utilizes hssr.ideasspread.org Humanities and Social Science Research Vol. 8, No. 4; 2025 109 Published by IDEAS SPREAD digital technologies—helps illuminate its inherent problems, limitations, potentials, and contradictions (Sandoval et.al., 2014) [2]. The conceptualization of digital labor must not ignore its physical and manual dimensions (Sandoval, 2013) [3]. As Fuchs (2015) [4] points out, defining digital labor requires attention to the various forms of mental and physical work involved in the production, circulation, and consumption of digital media. In his analysis, digital labor encompasses a wide range of workers, including Congolese miners, Foxconn factory workers, Indian software service employees, and even Facebook users. He argues that users of advertising-driven social media platforms become digital laborers, as their activities generate both productive and surplus value, which is subsequently commodified and sold on the market. Chen (2015) [5] further theorizes digital labor by emphasizing the discursive structures surrounding data value and labor value in the internet era. She defines digital labor as an interactive process with digital communication technologies that has already been monetized. This definition differs from traditional notions of work, as it encompasses fragmented and transient encounters in the digital and networked environment—interactions that are typically not recognized as labor. These include online activities such as blogging, watching videos on YouTube, or browsing posts on Facebook. Scholars have mostly advocated for an inclusive understanding of digital labor, given the production, circulation, and use of digital media required for various forms of mental and physical labor. Scholars have largely emphasized the profit-making strategies adopted by digital media companies, particularly their exploitation of digital media users. Nixon (2016) [6], using Google as an example, argues that most digital media companies seek to regulate users’ activities as cultural consumers in order to maximize profits. The Googlization of everything represents an attempt to control the broad spectrum of digital cultural consumption, which is understood as a form of digital audience labor exploitation. 3. Research Method This study employs two primary research methods: online observation and semi-structured interviews. Observation is one of the most fundamental methods through which humans perceive the world, and it is also a key technique in scientific research. As Chen (2000) [7] notes, unlike casual observation in everyday life, scientific observation is a purposeful and systematic activity undertaken by the researcher. Online observation refers to the process in which a researcher enters a specific digital environment to systematically and continuously observe online interaction behaviors. By analyzing these behaviors, researchers can gain insights into users’ psychological states (Zhang & Lei, 2012) [8]. Based on the research questions, this study focuses on observing the behaviors of film reviewers in the Douban Movie community, including their personal pages, data profiles, and textual contributions. Interviews, on the other hand, are a form of purposeful and rule-guided conversational inquiry where researchers collect first-hand data from participants through verbal communication (Chen, 2000) [7]. Depending on the degree of structure imposed by the researcher, interviews can be categorized as structured, unstructured, or semi- structured. This study adopts semi-structured interviews, wherein the researcher maintains a moderate level of control over the interview format while allowing participants to engage freely. A general interview guide was prepared in advance to ensure coherence across interviews. A total of eight film reviewers were selected based on the following criteria: they have been active on Douban for over four years, have watched more than 1,500 films, and have over 1,500 followers(For confidentiality, the participants are referred to as Interviewees M1 to M8）. In the data analysis stage, this study employed thematic analysis to systematically identify and interpret recurring patterns and meaningful themes related to film reviewers' digital labor practices. The researcher first transcribed interviews and online observation notes, then applied open coding to extract initial codes reflecting practices such as review preparation, writing strategies, interaction habits, and perceptions of autonomy. Based on this initial coding, the study categorized related codes into core themes such as pre-review preparation, writing labor, platform interaction, and limited autonomy. This coding process provided a solid analytical foundation for understanding the mechanisms and subjective experiences of digital labor within the context of user-generated film reviews on Douban. 4. The Labor Process and Autonomy of Film Reviewers 4.1 Operational Habits On Douban Movie, users can assign star ratings to films, which are then aggregated and analyzed through algorithms to generate a final numerical score. In addition to ratings, users can write both short comments and long-form reviews. Upon opening the page of any given film on Douban, one can see the number of short and long reviews available. Typically, short comments greatly outnumber long-form reviews. This raises the question: How hssr.ideasspread.org Humanities and Social Science Research Vol. 8, No. 4; 2025 110 Published by IDEAS SPREAD does one go about writing a long-form review? What kind of preparatory work and considerations are involved in the writing process? Insights into these questions emerged through interviews with participants M2 and M5. Participant M2 has authored several original film reviews that have ranked first among the most popular reviews for certain films. He explained that his writing habit involves first summarizing the plot, and then analyzing it from various perspectives, such as storyline, direction, acting, and behind-the-scenes production. Generally, before writing, he collects relevant news articles about the film, incorporating insights from professional critics to substantiate his own views. His reviews are known for their professionalism and depth, which contribute to their appeal among Douban users. Participant M5, a financial journalist and professional writer, noted that before reviewing a new film by a particular director, he would systematically study that director’s past works, as well as the performances of the cast involved. Based on these interviews, the study observed that the preparatory work involved in writing long-form reviews typically includes researching relevant materials, reading other users' reviews, and watching a substantial number of films. Moreover, additional labor often occurs during the viewing process itself, such as taking notes while watching. Participant M5 mentioned that he had watched some films multiple times, meticulously noting down various details, including emotional reactions that might later be used in his writing. He also described carefully planning the structure of each review, reflecting extensively on what to write, how to write it, and scrutinizing every word and sentence. These findings suggest that writing long-form film reviews involves a fragmented and time-consuming process, particularly due to repeated film viewing. While the act of publishing a review may appear to conclude the labor process, this study argues that interactions with other Douban users following publication should also be considered part of the labor involved. These interactions require additional personal time and mental effort from the reviewers. User engagement generally takes two forms: wide-ranging public interaction and more limited interaction within personal networks. According to the interviews, reviewers tend to favor the latter. Participant M4 indicated that he frequently leaves messages for his close contacts to share his latest thoughts. Similarly, Participant M6 expressed a preference for engaging with friends rather than unfamiliar users, valuing feedback from acquaintances more. The study found that such small-scale interactions are often more in-depth and frequent. Additionally, these interactions may extend to other social media platforms or even offline activities. 4.2 Temporal Intensity This section evaluates the temporal intensity of film reviewers’ engagement with Douban from three dimensions: time spent browsing the platform, time required for writing, and the often-overlooked fragmented time involved. Interview data reveal that most film reviewers habitually log into Douban daily, typically spending between one to five hours browsing. Notably, their frequency of login often exceeds once per day. This frequency tends to increase with the number of years a user has been registered on Douban, primarily due to the accumulation of a stable social network. For instance, participant M5, who has used the platform for seven years, noted that after forming friendships on Douban, he gradually developed a stable circle of social contacts, which prompted him to log in frequently to check for updates and messages from friends. Writing a long-form review takes two to three hours, although this time is not fixed. Participant M4 stated that writing time can fluctuate based on their emotional response to the film—about two hours in smooth situations, but three to four hours if the film inspires a deeper emotional reaction and the reviewer aims for a more polished piece. Furthermore, hitting creative bottlenecks is common among reviewers; due to factors such as workload or lack of inspiration, it may take several days to complete a review. Through in-depth interviews, the study also identified a significant amount of labor time that is fragmented and thus easily overlooked. For instance, time spent reviewing the logical flow and correcting typos after completing a draft is often unaccounted for by reviewers themselves. M5, a journalist by profession, repeatedly emphasized his fast-writing speed and high level of textual control—claiming an output of 1,600–1,700 words per hour. However, based on prior discussions about his preparatory work, the study found that the actual time he spends on writing far exceeds this. After drafting a review, M5 typically revises the logical structure and performs line-by- line proofreading to ensure coherence. Much like the interactive behaviors discussed earlier, this kind of fragmented labor time constitutes an often ignored but essential part of the film reviewer’s work process. 4.3 The Autonomy of Film Reviewers Douban Movie is the largest film-sharing and review community in China, attracting a vast number of film enthusiasts. As its rating system gains widespread recognition and dissemination, Douban’s reputation for “strict” scoring has drawn public attention and has come to be regarded as a symbol of authority and fairness. Therefore, hssr.ideasspread.org Humanities and Social Science Research Vol. 8, No. 4; 2025 111 Published by IDEAS SPREAD exploring film reviewers’ autonomy in the process of writing reviews is critical to understanding both Douban’s rating and review system and the labor process of content contributors. The concept of autonomy is often associated with the political sphere. Pressures on autonomy usually stem from political or commercial power (Nygren, Ostrowska & Anikina, 2015) [9]. This study finds that Douban’s review censorship system primarily affects reviewers’ creative practices. Participant M3 noted that, after several years of encountering censorship due to sensitive keywords, he has become well-versed in navigating Douban’s moderation policies. Over time, he adapted his writing habits, consciously avoiding political, violent, or sexual themes, and now focuses only on more entertainment-oriented films. At the same time, autonomy is a fluid and shifting concept (Sjøvaag, 2013) [10]. Some participants pointed out that, compared to other platforms, Douban still offers relatively greater autonomy, especially in allowing discussions around niche and marginal cultures. For instance, participant M8 observed that censorship on many online platforms is more stringent than on Douban, and that fans of subcultures have ultimately found a space for dialogue on Douban. Interview data suggest that in response to platform-level constraints, film reviewers must make various compromises to retain autonomy—such as altering creative habits or abandoning personal cultural interests. Although Douban encourages original content creation and topic-based discussions, reviewers, as digital laborers, are constrained by structural limitations while producing cultural content under restrictive creative conditions, thus contributing both traffic and value to the platform. 5. Conclusion In the past, watching television was considered labor. Today, being online is labor. This study has explored the labor process of film reviewers on Douban, demonstrating that the work they engage in is complex, time- consuming, and often invisible to both the reviewers themselves and the broader audience. The labor involved not only includes the time spent writing reviews, but also the preparatory work, which involves researching, watching films multiple times, and organizing thoughts. These elements, although essential, are often underestimated in terms of time and energy expended. Additionally, this labor includes subtle and fragmented forms of effort that are frequently overlooked, including proofreading, restructuring arguments, and maintaining user interactions. The ‘always-online’ nature of social networks, along with the accumulation of thousands of ‘likes’ and ‘shares’ enables advertisers not only to understand consumers purchasing patterns but also to access micro-level details such as their every move, buying habits, and geographic location. This phenomenon has been aptly described as exploitative (Brown, 2014) [11]. As a platform that operates entirely on user-generated content, Douban relies on film reviewers who invest significant amounts of personal time and effort to produce cultural products—without any form of compensation. This indicates that the content created by users is appropriated by the platform, thereby constituting a form of exploitation. The results of this study also highlight the dynamic between user autonomy and platform control. While Douban’s review system provides a space for autonomy, especially in discussing niche and marginalized cultures, it also exerts significant pressure through content moderation and censorship. Reviewers, therefore, navigate a delicate balance—adapting their content creation to fit platform rules while maintaining a sense of personal agency. This tension between autonomy and regulation is central to understanding the digital labor. As users, especially in creative and cultural sectors, continue to drive platforms' growth, their labor increasingly becomes a key factor in platform profitability. References [1] Smythe, D. W. (1977). Communications: Blindspot of western Marxism. CTheory, 1(3), 1–27. [2] Sandoval, M., Fuchs, C., Prodnik, J. A., Sevignani, S., & Allmer, T. (2014). Introduction: Philosophers of the World Unite! Theorising Digital Labor and Virtual Work-Definitions, Dimensions, and Forms. tripleC: Communication, Capitalism and Critique, 12(2), 464–467. https://doi.org/10.31269/triplec.v12i2.622 [3] Sandoval, M. (2013). Foxconned labor as the dark side of the information age: Working conditions at Apple’s contract manufacturers in China. TripleC: Communication, Capitalism & Critique, 11(2), 318–347. https://doi.org/10.31269/triplec.v11i2.481 [4] Fuchs, C. (2015). Digital labor: A comment on César Bolaño’s tripleC reflection. tripleC: Communication, Capitalism & Critique, 13(1), 84–92. https://doi.org/10.31269/triplec.v13i1.675 [5] Chen, Y. (2014). Production Cultures and Differentiations of Digital Labor. tripleC: Communication, Capitalism & Critique. Open Access Journal for a Global Sustainable Information Society, 12(2), 648–667. https://doi.org/10.31269/triplec.v12i2.547 hssr.ideasspread.org Humanities and Social Science Research Vol. 8, No. 4; 2025 112 Published by IDEAS SPREAD [6] Nixon, B. (2016). The old media business in the new: ‘The Googlization of everything’ as the capitalization of digital consumption. Media, Culture & Society, 38(2), 212–231. https://doi.org/10.1177/0163443715594036 [7] Chen, X. M. (2000). Qualitative research methods and social science research. Education Science Press. [8] Zhang, G., & Lei, L. (2012). Current status and prospects of online research methods. Psychological Research, 5(3), 20–25. [9] Nygren, G., Dobek-Ostrowska, B., & Anikina, M. (2015). Professional autonomy. Nordicom Review, 36, 79– 95. https://doi.org/10.1515/nor-2015-0018 [10] Sjøvaag, H. (2013). Journalistic Autonomy. Between Structure, Agency and Institution. Nordicom Review, 34, 155–166. https://doi.org/10.2478/nor-2013-0111 [11] Brown, B. (2014). Will work for free: The biopolitics of unwaged digital labor. tripleC: Communication, Capitalism & Critique. Open Access Journal for a Global Sustainable Information Society, 12(2), 694–712. https://doi.org/10.31269/triplec.v12i2.538 Copyrights Copyright for this article is retained by the author(s), with first publication rights granted to the journal. This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).