Testing Times: Introduction Pilar Orero Universitat Autònoma de Barcelona David Hernández Falagán Universitat Politècnica de Catalunya _________________________________________________________ Abstract Citation: Orero, P. & D. Hernández That COVID-19 touched all walks of life is an understatement. With the Falagán (2021). Testing Times: risk of sounding frivolous, compared with other impacts, COVID-19 had Introduction. Journal of Audiovisual Translation, 4(2), 1–5. direct implications in research, and particularly in funded research https://doi.org/10.47476/jat.v4i2.2021.2 activities with a strict schedule. Luckily, in the field of audiovisual 07 Published: October 30, 2021 translation we do not require any live samples or animals to be fed Funding: This issue has been partially while in lockdown. Still, experimental programmed tests with people funded by the H2020 projects SO-CLOSE required alternative approaches. This special issue presents the social (GA 870939), REBUILD (GA 822215), TRACTION (GA 870610), HELIOS (GA distancing challenges faced in user-centric research methodologies 825585), and MEDIAVERSE (GA 957252) when human interaction is required. and by the Catalan Government SGR2017-113. Key words: COVID-19, testing, validation, user-centric research, Copyright: ©2021 Orero & Hernández Falagán. This is an open access article accessibility, accessible filmmaking. distributed under the terms of the Creative Commons Attribution License. This allows for unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.  pilar.orero@uab.cat, https://orcid.org/0000-0003-0269-1936  david.hernandez.falagan@upc.edu, https://orcid.org/0000-0002-9600-2304 1 Testing Times: Introduction 1. Introduction The lack of foresight and preparedness for COVID-19 proved cruel in all walks of life. Research was no exception, particularly when funded research activities had to meet established deadlines. Deadlines had to be kept and fast remedial action was required to find alternatives to physical spaces for shared interaction: from meetings to testing. Some Media Accessibility experimental research is user-centric and is framed within the Human Rights context of the United Nations Convention on the Rights of Persons with Disabilities1 (CRPD) “nothing about us is without us.” The end user in Media Accessibility used to be classified by their medical condition. Subtitles for the deaf focused on persons with hearing loss, and audio description focused on persons with sight loss. This clinical classification is no longer part of the demographic information for experiments in Media Accessibility, where the end user is now defined by their capabilities (Agulló et al., 2018). Reading capabilities, for example, are a good definition for testing subtitles, leading to a wide number of profiles related to literacy, learning, sight, reading in a different writing alphabet, etc. (Jiménez-Andrés, in this issue). Though it may seem strange, audio description should be tested based on hearing capabilities, since this accessibility service is sound based. It is important to define the end user’s capabilities and understand their needs and expectations towards meaningful technology development and validation. User-centric research methodology starts and ends with the end user and requires an intense interaction – which till March 2020 entailed contact face to face. Sharing a physical space had to be adapted to the many limitations imposed by lockdown and social distancing policies. Those policies were too changeable to allow for planning tests. The number of people who were allowed to meet indoors changed across time, countries and within the same country. Luckily Internet came to the rescue and offered a new kind of physical space to be shared safely. Internet was always available, and migration to this new testing environment was non- negotiable. Meetings moved to communication platforms, and so did focus groups, ethical requirements, questionnaires, and tests. While we had already started moving tests online, as with the Sign Language questionnaire (Velasco et al., 2019), there were no standalone processes which would be fully based online. Generating stimuli for analysis also moved to web-based platforms that offered instant visualisation (Hughes et al., 2020). Finally, some audiovisual content had to be generated and be accessible, and for that some quick and dirty guidelines had to be provided. These were required for some research projects such as Helios, Rebuild, So-Close, Traction, and Mediaverse. 1https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with- disabilities.html 2 Journal of Audiovisual Translation Volume 4, issue 2 2. Life Goes On, but Now It Is Better COVID-19 gave the final push to compulsory online interaction. Physical presence was now a thing of the past: we had to communicate and work from home. The coexistence of online and face-to-face (F2F) events had always tilted towards the latter. In some meetings or conferences there could be the exceptional remote intervention, often with poor synchronisation or imperfect sound quality. It is true that in education full online courses in audiovisual translation had been on offer for decades (Amador et al., 2004), and a MOOC (Massive Open Online Course) on Media Accessibility had been created taking into consideration accessibility (Orero & Tor-Carroggio, 2018). However, till the early months of 2020 F2F had been the most popular option when training. While two decades ago setting up an online MA programme in Audiovisual Translation required creativity, today all translation tools are web based, and media can be fast and easily downloaded from a server. It was the mindset of those who in 2020 were still sceptic that needed resetting. COVID-19 triggered the universal format change to a new e-lifestyle “leaving no one behind” (LNOB). Unaware of COVID-19, the UN motto LNOB is the central transformative promise of the 2030 Agenda for Sustainable Development and its Sustainable Development Goals (SDGs). While the UN looks to “eradicate poverty in all its forms, end discrimination and exclusion, and reduce the inequalities and vulnerabilities that leave people behind and undermine the potential of individuals and of humanity as a whole”, this is also one of the main objectives shared by all the projects where accessibility had to be taken into consideration at the time of generating media content towards testing. Following Romero-Fresco (2018), the decision was made to follow accessible filmmaking principles. COVID-19 has helped to highlight the existing potentialities and tools for communication and interaction. The many voices against online formats had to quickly adapt to what they had rejected based on no evidence. Media accessibility in both content and interaction became again a central issue across all demographics, including migrants (Jiménez-Andrés, in this issue). To date (2021) no communication platform is fully accessible in its interaction or in the services it allows, e.g., a shared screen for a Power Point Presentation and a Sign Language interpreter or speaker. Even if most online communication platforms allow the use of subtitles, these are mostly automatic subtitles and only for English, minority languages pose a major challenge for subtitling, especially in the case of minority languages (Oncins & Delgado, 2019). These were some of the issues faced when setting up online experiments in the aforementioned EU research projects (Okyayuz, in this issue). The silver cloud with COVID-19 has been the fast compulsory training of all people in the use of online communication platforms, and the crash course in web usability (W3C). Everyone was exposed to the reality of interaction with web applications, and accessibility organically took centre stage (Oncins, in this issue). In the past, when defining technology requirements, some interaction issues had been flagged and procrastinated, now suddenly they were analysed and solved, and even data protection had to be quickly updated to the new conditions (Goberna, in this issue). What the lived experience demonstrates is that the pandemic has precipitated a digital future in which we were thrown into a context of interacting that had largely been unfamiliar. An online 3 Testing Times: Introduction context which often fails to ensure equal access, but whose infrastructure promises unbeatable potential to build universal-access communication and interaction tools (Hernández-Falagán, 2020). Probably in many cases the risks involved in online testing and evaluation operations closely resemble the risks involved in simultaneous migration to a virtual workspace: control of data and users; relocation of participants with the risk of exploitation of disadvantaged regions; obligation of users to provide their own infrastructure (connection and devices). However, as this special issue demonstrates, digital space and online platforms offer creative opportunities to face such risks and replace (when not already coexisting) the physical spaces in many forms of interaction and communication. References Agulló, B., Matamala, A., & Orero, P. (2018). From disabilities to capabilities: Testing subtitles in immersive environments with end users. HIKMA, 17: 195–220. https://doi.org/10.21071/hikma.v17i0.11167 Amador, M., Dorado, C., & Orero, P. (2004). e-AVT: A perfect match: Strategies, functions and interactions in an on-line environment for learning. In P. Orero (ed.) Topics in audiovisual translation (pp. 141–153), Benjamins. https://doi.org/10.1075/btl.56 Goberna, C. (forthcoming). Respecting the GDPR in online interviewed focus groups. Journal of Audiovisual Transalation. Hughes, C., Zapata, M. B., Johnston, M., and Orero, P. (2020, December 14–18) Immersive captioning: Developing a framework for evaluating user needs [Conference presentation]. IEEE AIVR 2020: 3rd international conference on artificial intelligence & virtual reality , [please add place of the conference here], [state]https://aivr.science.uu.nl/ Hernández. Falagán, D. (2020). Collaborative pedagogy and social networks. Design in quarantine. VIII workshop on educational innovation in architecture (JIDA’20), Escuela Técnica Superior de Arquitectura de Málaga, 2020: Proceedings. https://doi.org/10.5821/jida.2020.9336 Jiménez-Andrés, M. (forthcoming). Researching vulnerable groups during COVID-19: Displaced populations. Journal of Audiovisual Transalation. Okyayuz, S. (forthcoming). Food for thought on conducting research projects on media accessibility during the COVID-19. Journal of Audiovisual Transalation. Oncins, E. (forthcoming). Accessibility in online user-based testing. Journal of Audiovisual Transalation. Oncins, E., & Delgado, H. (2019). Respeaking in minority languages: Development of a Catalan automatic speech recognition system. SKASE Journal of Translation and Interpretation, 12(2), 37–49. Orero, P., & Tor-Carroggio, I. (2018). User requirements when designing learning e-content: interaction for all. In E. Kapros &M. Koutsombogera (Eds.), Designing for the user experience in learning systems (pp. 105–121). Springer. Romero-Fresco, P. (2018). The accessible filmmaking guide. Arches Mark. http://archersmark.co.uk/wp-content/uploads/2019/04/Accessible-Filmmaking- Guide2018.pdf 4 Journal of Audiovisual Translation Volume 4, issue 2 Velasco, J. P., Bosch, M., Martin, C. A., Menéndez, J. M., Orero, P., Soler, O., & Álvarez, F. (2019). Design and development of sign language questionnaires based on video and web interfaces. In P. Biswas, P. Orero & T. Metin Sezgin, (Ed.) Intelligent interaction design [Special Issue]. Artificial Intelligence for Engineering Design, Analysis and Manufacturing , 33(4), 429–441. doi:10.1017/S0890060419000374 5