Selected Papers of #AoIR2024: The 25th Annual Conference of the Association of Internet Researchers Sheffield, UK / 30 Oct - 2 Nov 2024 TO SCREENSHOT OR NOT TO SCREENSHOT? TENSIONS IN REPRESENTING VISUAL SOCIAL MEDIA PLATFORM POSTS Joseph S. Schafer University of Washington Brett A. Halperin University of Washington Sourojit Ghosh University of Washington Julie Vera University of Washington Introduction The rise to prominence of visual social media platforms (VSMPs) including TikTok, Instagram, and YouTube has led to increasing amounts of research attention directed to these platforms. As research engages multimodal platforms, representing their content (including text, audio, image, and video components) increasingly becomes both important and complex. The AoIR Internet Research Ethics (IRE) 3.0 guidelines stipulate that “we need to elaborate an ethics addressing the distinctive issues clustering around the production, sharing, and thereby research on visual images” (franzke et al., 2020). In this paper, we begin making such an elaboration, describing considerations necessary when representing screenshots. We provide an overview of four current approaches to representing VSMP posts and annotate their tensions. Prior Examinations of Screenshots The act of screenshotting is non-neutral, and produces its own media artifact. One notable discussion of this artifact format is (Frosh, 2018), which provides an in-depth historical and media form analysis of the screenshot. This details how the screenshot’s design both invisibilizes its existence as its own media object, and makes it function as an evidentiary artifact (see also (Jaynes, 2020)). Other researchers have discussed how Suggested Citation (APA): Schafer, J. S., Halperin, B. A., Ghosh, S., and Vera, J. (2024, October). TO SCREENSHOT OR NOT TO SCREENSHOT? TENSIONS IN REPRESENTING VISUAL SOCIAL MEDIA PLATFORM POSTS. Paper presented at AoIR2024: The 25th Annual Conference of the Association of Internet Researchers. Sheffield, UK: AoIR. Retrieved from http://spir.aoir.org. Suggested Citation (APA): Schafer, J. S., Halperin, B. A., Ghosh, S., Vera, J. (2024, October). To Screenshot or Not to Screenshot? Tensions in Representing Visual Social Media Platform Posts. Paper presented at AoIR2024: The 25th Annual Conference of the Association of Internet Researchers. Sheffield, UK: AoIR. Retrieved from http://spir.aoir.org. screenshots can be used to hold structural powers accountable (Jenkins & Cramer, 2022). Prior research has discussed screenshots’ privacy implications, and how researchers can learn from fan archives’ practices (Drouin, 2021) and online remix cultures (Fiesler & Bruckman, 2014). (Shore & Prena, 2023)’s recent study of users’ screenshotting behaviors reveals how non-researchers conceptualize screenshots’ privacy risks. Further, we find (Cagle, 2021)’s discussion of researching nonconsensually-captured images informative. Cagle argues that both looking at and sharing these images can be ethically fraught, and urges for obtaining informed consent from those represented along with ‘ethical ekphrasis’ as text-only or text-in-addition description of images (Cagle, 2021). It is important to note issues of representation in research images have significant histories. One example is the “Lena” image of a Playboy Magazine model, used in early computer vision research without the model’s consent (Mulvin, 2021a, 2021b). Current Practices of VSMP Screenshot Inclusion To ground our discussion of tensions in screenshotting VSMPs, we synthesize common practices in existing research. Representations of these media tend to come in four primary forms: 1) text-only descriptions, 2) unaltered screenshots, 3) partially-altered screenshots, and 4) drawings. Text-only descriptions do not represent posts visually, but through verbal descriptions, such as (Vázquez-Herrero et al., 2022) and (Kennedy, 2020). Unaltered screenshots are not subsequently edited before being included in research outputs, such as (Maddox, 2023), and (Bhandari & Bimo, 2022). Partially-altered screenshots have visual elements blurred or altered, such as obscuring creators’ faces (e.g. (Lundy, 2023)) or blurring account usernames (e.g. (Khoury et al., 2023)). These alterations can include blocking less personally-identifiable information, such as (Vizcaíno-Verdú & Abidin, 2023)’s replacing “like” counts alongside blurring creators’ faces. Our final example of screenshot practices is when researchers draw, rather than screenshot, posts. Examples include (Abidin, 2018), (Stephenson et al., 2024), and the conference presentation of (Biggs et al., 2023). Alongside image formats of VSMP screenshots, we also see diversity in what screenshots are taken. While most previously-referenced papers include a single screenshot per video, some of the four-screenshot composites in (Vizcaíno-Verdú & Abidin, 2023) may be of subsequent frames within one video. This reveals important methodological questions — what frame do you choose to screenshot moving images? How do researchers convey cuts between different video parts, or choose a “typical” frame for a particular camera setup within a video? How do static portrayals reduce nuance and encode researchers’ biases by freezing complex subjects at selected moments in time? Compositing multiple screenshots to account for temporality, or using artistic techniques such as blur lines, could help address these challenges. Ethical Tensions Regarding VSMP Screenshots in Research As the IRE guidelines (franzke et al., 2020) argue, representing multimodal data is particularly ethically fraught. Below, we outline three value questions researchers should reflect upon when including VSMP screenshots (or not): what is (in)visibilized, what is consensual, and what is legal, when representing VSMP posts? Similar to (Schafer et al., 2023), we do not prescribe guidelines or resolutions, but instead describe critical considerations. First, researchers should address the tension between the values of attribution and anonymity. Posts by VSMP creators represent significant creative labor. On one hand, invisibilizing this labor through anonymization can harm creators by not crediting them. On the other hand, visibility can elicit further scrutiny or harassment (Alcoff, 1991). Researchers have demonstrated creators often navigate calibrated visibility using complex folk theories (DeVito, 2022), and this navigation has been discussed in prior AoIR research (Stegeman et al., 2022). Adding unintended visibility could cause harm by undermining creators’ efforts to be seen only in specific contexts (Dym et al., 2019). Visibility is also problematized for those studying harmful media on VSMPs, from media captured nonconsensually (Cagle, 2021), to misinformation or otherwise objectionable content, when researchers might want to avoid giving this content the “oxygen of amplification” (Phillips, 2018). Another tension for researchers to navigate is around degrees of informed consent, particularly since VSMPs are known to have long, convoluted end-user license agreements (EULAs) that most users do not read before joining (Fiesler & Proferes, 2018). Since most users do not know the rights they sign away as they join VSMPs, researchers have additional responsibility to take care in representing user-generated content in their work. Researchers could consider procedures of dynamic consent (Tseng et al., 2024), and contact users whose content they are interested in highlighting, to discuss consent to use and represent such content in their papers. However, there are temporal challenges associated with dynamic consent (Halperin & McElroy, 2023) and mechanisms needed to accommodate changes in consent preferences over time, particularly when dealing with sensitive subject matters (Halperin et al., 2023). After an article’s publication, researchers no longer have easy ways to remove screenshots. While consent may be difficult to adapt or dangerous to acquire for some cases, such as in (Fuchs & Meikle, 2018), considering these processes and contextual degrees of risk when representing VSMP posts is warranted. Our third tension exists within navigating legal frameworks surrounding VSMP posts and their usage. Since posts constitute intellectual property, including screenshots could implicate researchers under copyright law. For instance, some fair use guidelines for researchers caution screenshots may be used “only for the purpose of critique” but not to “illustrate a point,” and that redrawing a figure does not equate to permissible use (Association for Computing Machinery, n.d.). Additionally, platform terms of service regarding researcher data access and republication are important considerations for representing VSMP posts. However, not including screenshots, since drawings or textual descriptions may not fulfill screenshots’ evidentiary role, may present its own legal challenges, such as creators alleging they were inaccurately or even defamatorily represented. Omitting screenshots may also weaken researchers’ arguments, since their ability to demonstrate evidence via VSMP post representations would be diminished. Conclusions and Future Work In this paper, we described existing methods researchers use to represent VSMP posts, and some attendant ethical tensions in using screenshots. Beyond the tensions and overview we produced, this reveals important future research directions. These include interviewing researchers about their existing practices, exploring alternative representation methods, or studying users’ reactions to the VSMP post usage in research, similar to (Fiesler & Proferes, 2018). Acknowledgements The authors would like to thank Dr. Lauren S. Berliner, Nina Lutz, the students of University of Washington Cinema and Media Studies 597 “Digital Media Power and Practice,” and the reviewers for their thoughtful comments which improved this project. Joseph S. Schafer and Brett A. Halperin are supported by National Science Foundation Graduate Research Fellowships (Grant number DGE-2140004). Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the National Science Foundation. References Abidin, C. (2018). Internet Celebrity: Understanding Fame Online (Illustrated edition). Emerald Publishing. Alcoff, L. (1991). The Problem of Speaking for Others. Cultural Critique, 20, 5–32. https://doi.org/10.2307/1354221 Association for Computing Machinery. (n.d.). Guidance For Authors On Fair Use. Retrieved February 26, 2024, from https://www.acm.org/publications/authors/guidance-for-authors-on-fair-use Bhandari, A., & Bimo, S. (2022). Why’s Everyone on TikTok Now? The Algorithmized Self and the Future of Self-Making on Social Media. Social Media + Society, 8(1), 20563051221086241. https://doi.org/10.1177/20563051221086241 Biggs, H., Marcotte, A., & Bardzell, S. (2023). TikTok as a Stage: Performing Rural #farmqueer Utopias on TikTok. Proceedings of the 2023 ACM Designing Interactive Systems Conference, 946–956. https://doi.org/10.1145/3563657.3596038 Cagle, L. E. (2021). The ethics of researching unethical images: A story of trying to do good research without doing bad things. Computers and Composition, 61, 102651. https://doi.org/10.1016/j.compcom.2021.102651 DeVito, M. A. (2022). How Transfeminine TikTok Creators Navigate the Algorithmic Trap of Visibility Via Folk Theorization. Proceedings of the ACM on Human-Computer Interaction, 6(CSCW2), 1–31. https://doi.org/10.1145/3555105 Drouin, R. A. (2021). (Don’t) just screenshot it: Ethics of archiving fan misbehaviours. Journal of Fandom Studies, 9(1), 19–32. https://doi.org/10.1386/jfs_00030_1 Dym, B., Brubaker, J. R., Fiesler, C., & Semaan, B. (2019). “Coming Out Okay”: Community Narratives for LGBTQ Identity Recovery Work. Proceedings of the ACM on Human-Computer Interaction, 3(CSCW), 154:1-154:28. https://doi.org/10.1145/3359256 Fiesler, C., & Bruckman, A. S. (2014). Remixers’ understandings of fair use online. Proceedings of the 17th ACM Conference on Computer Supported Cooperative Work & Social Computing, 1023–1032. https://doi.org/10.1145/2531602.2531695 Fiesler, C., & Proferes, N. (2018). “Participant” Perceptions of Twitter Research Ethics. Social Media + Society, 4(1), 2056305118763366. https://doi.org/10.1177/2056305118763366 franzke, aline shakti, Bechmann, A., Zimmer, M., Ess, C., & the Association of Internet Researchers. (2020). Internet Research: Ethical Guidelines 3.0. https://aoir.org/reports/ethics3.pdf Frosh, P. (2018). Screenshot: The “Photographic” Witnessing of Digital Worlds. In The Poetics of Digital Media (pp. 158–223). Polity. Fuchs, C., & Meikle, G. (2018). “Dear Mr. Neo-Nazi, Can You Please Give Me Your Informed Consent So That I Can Quote Your Fascist Tweet?”: Questions of Social Media Research Ethics in Online Ideology Critique. In Routledge Companion to Media and Activism. Routledge. https://westminsterresearch.westminster.ac.uk/download/3771d54accbad32ff3aa d192b97046846bf53526a21984caf3c25b140ef88c26/160940/Routledge%20Co mpanion%20to%20Media%20and%20Activism.pdf Halperin, B. A., Hsieh, G., McElroy, E., Pierce, J., & Rosner, D. K. (2023). Probing a Community-Based Conversational Storytelling Agent to Document Digital Stories of Housing Insecurity. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, 1–18. https://doi.org/10.1145/3544548.3581109 Halperin, B. A., & McElroy, E. (2023). Temporal Tensions in Digital Story Mapping for Housing Justice: Rethinking Time and Technology in Community-Based Design. Proceedings of the 2023 ACM Designing Interactive Systems Conference, 2469–2488. https://doi.org/10.1145/3563657.3596088 Jaynes, V. (2020). The social life of screenshots: The power of visibility in teen friendship groups. New Media & Society, 22(8), 1378–1393. https://doi.org/10.1177/1461444819878806 Jenkins, B. M., & Cramer, E. M. (2022). Capturing Injustice: The Screenshot as a Tool for Sousveillance. Howard Journal of Communications, 33(4), 412–433. https://doi.org/10.1080/10646175.2022.2032884 Kennedy, M. (2020). ‘If the rise of the TikTok dance and e-girl aesthetic has taught us anything, it’s that teenage girls rule the internet right now’: TikTok celebrity, girls and the Coronavirus crisis. European Journal of Cultural Studies, 23(6), 1069–1076. https://doi.org/10.1177/1367549420945341 Khoury, C., Owen-Smith, A., Joh, U., Duan, Y., & Hemsley, J. (2023). Multi-Modal Crisis Discourse and Collective Sensemaking on TikTok. Proceedings of the Association for Information Science and Technology, 60(1), 203–212. https://doi.org/10.1002/pra2.781 Lundy, M. (2023). TikTok and COVID-19 Vaccine Misinformation: New Avenues for Misinformation Spread, Popular Infodemic Topics, and Dangerous Logical Fallacies. International Journal of Communication, 17(0), Article 0. Maddox, J. (2023). Micro-celebrities of information: Mapping calibrated expertise and knowledge influencers among social media veterinarians. Information, Communication & Society, 26(14), 2726–2752. https://doi.org/10.1080/1369118X.2022.2109980 Mulvin, D. (2021a). Proxy Justice (or The Lena Image, Part 2). In Proxies: The Cultural Work of Standing In (pp. 113–144). The MIT Press. https://direct.mit.edu/books/oa-monograph/5154/chapter/3403261/Proxy-Justice- or-The-Lena-Image-Part-2 Mulvin, D. (2021b). The Visual Culture of Image Engineers (or The Lena Image, Part 1). In Proxies: The Cultural Work of Standing In (pp. 73–112). The MIT Press. https://direct.mit.edu/books/oa-monograph/5154/chapter/3403261/Proxy-Justice- or-The-Lena-Image-Part-2 Phillips, W. (2018). The Oxygen of Amplification. Data & Society. https://datasociety.net/wp-content/uploads/2018/05/1_PART_1_Oxygen_of_Ampl ification_DS.pdf Schafer, J. S., Starbird, K., & Rosner, D. K. (2023). Participatory Design and Power in Misinformation, Disinformation, and Online Hate Research. Proceedings of the 2023 ACM Designing Interactive Systems Conference, 1724–1739. https://doi.org/10.1145/3563657.3596119 Shore, A., & Prena, K. (2023). Platform rules as privacy tools: The influence of screenshot accountability and trust on privacy management. New Media & Society, 14614448231188929. https://doi.org/10.1177/14614448231188929 Stegeman, H. M., Poell, T., Duffy, B. E., Jokubauskaitė, E., Cotter, K., Velthuis, O., & Meisner, C. (2022). PLATFORMS AND THE PRECARITY OF CREATOR (IN)VISIBILITY. AoIR Selected Papers of Internet Research. https://doi.org/10.5210/spir.v2022i0.12968 Stephenson, S., Page, C. N., Wei, M., Kapadia, A., & Roesner, F. (2024, May 11). Sharenting on TikTok: Exploring Parental Sharing Behaviors and the Discourse Around Children’s Online Privacy. Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems. CHI ’24, Honolulu, HI, USA. Tseng, E., Bellini, R., Lee, Y.-Y., Ramjit, A., Ristenpart, T., & Dell, N. (2024). Data Stewardship in Clinical Computer Security: Balancing Benefit and Burden in Participatory Systems. Proceedings of the ACM on Human-Computer Interaction, 8(CSCW1), 29. https://doi.org/10.1145/3637316 Vázquez-Herrero, J., Negreira-Rey, M.-C., & López-García, X. (2022). Let’s dance the news! How the news media are adapting to the logic of TikTok. Journalism, 23(8), 1717–1735. https://doi.org/10.1177/1464884920969092 Vizcaíno-Verdú, A., & Abidin, C. (2023). TeachTok: Teachers of TikTok, micro-celebrification, and fun learning communities. Teaching and Teacher Education, 123, 103978. https://doi.org/10.1016/j.tate.2022.103978