|Designing interfaces to support contextual interpretation and open conversations with users|
Marialaura Ghidini: I’d like to start by asking you a bit about the genealogy of your work. What brought you to investigate publishing and archiving online?
Lozana Rossenova: It’s not a very linear path, but it started with an interest in fine art and design, that led me to pursue my undergrad studies which combined studio art, art history and design studies. During my undergrad I became particularly interested in information and system design — I wanted to study the typographic details that allow us to make sense of visual information: tables, indexes, footnotes, marginalia. My MA (at the University of Reading) combined traditional typography and communication, with more avant-garde practices. For example, I studied book and publication design under Fraser Muggeridge and John Morgan, designers who work exclusively with art and cultural clients. I wrote a dissertation on designing contemporary art catalogs, under the supervision of Dr Ruth Blacksell, whose own research focused on the ways in which typography and different forms of published documents were used by 1960s & 70s experimental writing, performance and Conceptual Art practices, in their shift away from fixed to more subjective and fluid forms. Dr Blacksell’s work introduced me to Aspen magazine, the Whole Earth Catalog, Art-Language, among others, where the forms and operations of the published document were open-ended, iterative and involved the readers in (inter)active ways. The work of Stuart Bailey, another Reading alum, with Dot Dot Dot, Dexter Sinister, and later The Serving Library, which is a project defined as “an archive that publishes, and a publisher that archives” was also highly influential. I was looking at understanding and developing complex design systems but through the perspective of art practice, modes of information circulation, and the idea of the ‘active’ reader vs a passive consumer of information. I did some DIY publishing experiments during my MA that resulted in a collaborative project titled Inland Editions and a publication of a wide range of essays on libraries and alternative library and archival spaces.
This was around 2011-12, the time when iPads came out, Kindles were incredibly popular, and digital books were on the radar of most publishers and publication designers. So after my MA, I did a lot of digital publications for commercial publishers like Taschen, or various educational and non-profit clients. I was becoming more interested in the intersection of publishing and accessing information digitally, but I was sceptical of being too closely tied to the devices… I preferred to design for what at the time were called ‘HTML-readers’ — essentially accessible websites, where publications can be read and browsed outside the boundary of a PDF or Epub file. The boom-and-bust (within just few years) of some of the costly iPad creations of some of the commercial publishers I was following, or working with, gave me an early taste of the consequences of technical obsolescence and dependencies. One of the last design jobs I did before going back to school to pursue a PhD was a digital archive for a big research programme at the LSE (London School of Economics). The project was very exciting, as it contained a wide diversity of source information — from newspapers and coffee table books to complex interactive data visualisation. It allowed me to explore how device-agnostic, web platforms can make information more accessible, and showed the benefits of taking digital archiving into consideration during the production and publication of information, and not just relegating it as an afterthought. It also highlighted the limits of trying to solve complex information design problems within the confines of a commercial contract.
At the time, 2016, I found the position for a PhD researcher with Rhizome being advertised and decided to apply. That project was set up as a collaborative partnership between Rhizome and the Centre for the Study of the Networked Image at London South Bank University, and focused on redesigning Rhizome’s online archive of net art. At first, it may have appeared to be a stretch to move from a career focusing mostly on design and publications, albeit digital, towards a research around a very particular online archive, the ArtBase. But on closer inspection, there were many intersections, and points of productive friction, between the worlds of publishing — and particularly publishing as artistic practice — and the online archive of net art. Not least the fact that although net art emerged as an artistic movement in the 90s, it has a long lineage of connections to earlier avant-garde and anti-institutional performance- and publishing-related art practices — practices concerned with the circulation and ‘making public’ of various types of visual, verbal or text-based information. I think I’ll stop here, as I’m sure we’ll have a chance to draw out further the common threads between the worlds of publishing and design on the one hand, and archives, networks, and software on the other, throughout the rest of this conversation.
M.G.: I’m sure! You have extensively worked with preservation of digital art for your practice-based PhD. In this respect, I am interested in your design approach and the fact you see preservation (and therefore archiving) as an act that is not only technical, but social and collective; highlighting the idea of context rather than content. Which role do social aspects and collective effort/knowledge play in your work?
L.R.: Thank you, that’s a great question. With my PhD research, I aimed to unpack the question of user access in born-digital archives, and more specifically how to support informed user interaction and potential for intervention in the archive, through a situated design practice.
Here net art and its specific mode(s) of operation, as well as the organisational history of the ArtBase, both informed and influenced my design approach and methodology. To be performed and experienced, net art depends upon alignments between hardware and software environments, network protocols, and crucially an active user — without user interaction, the works cannot be rendered at all. Therefore to preserve net art works in the ArtBase, Rhizome developed the concept of reperformance-as-preservation. This paradigm positions the figure of the user as central to preserving the works and the social context around them, which extends beyond the internal operations of the institution, the archival team, and even the artists. Here, preserving social memory and contextual knowledge about maintaining and using a specific piece of software, or about performing a specific interaction within a particular graphical user interface, become important processes part of the social fabric of the archive and concern various communities of practice — including both internal stakeholders and external users. But most archival systems lack capacity to record this type of memory or collective knowledge in meaningful ways through their data schemas, and also lack visual strategies to display this in a user interface.
In my PhD thesis, I argue that conceptualizing the archive as a network of socio-technical relations is productive not only from the point of view of social sciences, science and technology studies (STS), or critical archival theory, but also for the practice of designing archive infrastructure and interfaces. I argue that the design process needs to be understood not (just) as an expert-driven act of translation from one social world to another — a term often used in the human computer interaction (HCI) and user experience (UX) design field, indicating the transformation of stakeholder requirements into designed artefacts that also meet user expectations — but rather as an act of building networks of connections that facilitate many-to-many relations. To make this more concrete, the core idea of the methodological framework I developed in my PhD practice is that besides the technical and infrastructural aspects (the data model, the database and the user interface), the design practice needs to take into account the processual dimension, i.e. the social context of the archive network, namely — processes of classification, maintenance and use. And furthermore, the technical dimensions are always already operated by and operating upon the social processes. The way different communities of practice classify specific concepts or materials into descriptive and contextual metadata, carry out maintenance activities or perform user interactions in the archive, inform the design of the interface, the choice of database and the structure of the data model.
Finally, the design of the technical and infrastructural aspects of the archive also aim to make the social processes more visible and readable to all users and stakeholders involved in the network of the archive, and therefore open possibilities for a more collaborative, collective form of preservation.
M.G.: Your report Design Landscape: Online Collection Interfaces (2020), presents a detailed analysis of the way viewers/users interact with various interfaces of online exhibitions. How does studying interfaces help with understanding our digital landscape?
L.R.: If I can pick up from the thread in the previous answer — I think studying interfaces helps us understand just how limited, and closed to the potential for intervention, or collaboration most forms of digital interactions are at the moment, which amounts to a very homogenous digital landscape… homogenous not in visual terms — I think there is still a great variety of interesting visual design (even if we account for the dominance of certain trends such as the colourful, bubbly illustrations gracing most tech startup websites these days); but the possibilities for more complex, less-prescriptive interactions remain rather limited within most digital products.
I was also looking primarily at digital design examples, not art (within explicitly artist-led contexts there is more space for experimentation), so that needs to be taken into account. The example interfaces were of course designed within specific contexts and with a specific audience in mind, but those designed artefacts are also what tends to set user expectations. For example, the more interfaces with very large ‘search’ bars we encounter in the digital landscape, the more these encounters set certain expectations — i.e. that archival and collection interfaces need to be primarily focused on ‘search’ interactions, and that the ideal user in those contexts is always looking for something specific. Although this assumption may not always be true, digital interfaces reinforce certain assumptions (about users, and use scenarios) over others, and over time this leads to standardized interactions that move the entire digital landscape in certain directions whether we think it’s the right direction or not. So the study of digital interfaces, I think, can be used as a form of ‘barometer’ for the general state of digital landscapes.
M.G.: Which are the types of interactions you have encountered, and what do they tell about design standards, or perhaps the history of their development?
L.R.: There were several common tropes that I encountered throughout most interfaces: prominent search bar features, timelines or other forms of clickable data visualizations, as well as the most ubiquitous design pattern known as “overviews and previews” in digital collection interfaces. The latter is essentially a way to describe the all-too-familiar grid of thumbnails. The grid pages present users with overviews of the whole collection, and the previews (or the thumbnails) give users a quick visual cue as to what to expect before clicking on a specific artwork (that is collection item) link. The overviews and previews were common among both collection and exhibition interfaces. In my report, I describe them as a form of born-digital salon wall, and in my thesis I elaborate on this idea a bit further. I argue that typically, online collection interfaces rely on metaphors from the social world of physical institutions, such as the metaphor of web pages as (virtual) galleries and thumbnail grids as salon walls, where static image files are put on view for the contemplation of visitors (not users). In looking at examples of online exhibitions, I also discovered two common forms what I refer to as “the white cube approach” or the “virtual 3D gallery environment” in the report. The white cube approach utilized grid layouts, large amounts of white space and affective flat images (i.e. representative screenshots) as stand-ins for the artworks, which were then usually available to explore via clicking on these screenshot previews. The 3D environments also sought to recreate the white-walled gallery experience, but with added interactions around a virtual 3D space inside the user’s own browser. Even in 3D space, however, the artworks’ representation remains flattened to 2D images, or at most videos.
These few examples point out a key aspect in the development of design standards, namely the use of metaphors with connection to the physical world (these are easy to spot everywhere across the desktop and the file/folder system we use in contemporary operating systems). And also the progression of metaphorical use in design from more abstract towards more literal application (i.e. from 2D thumbnail walls to 3D environments) with the historical improvement of hardware (graphics cards) and software (browser capabilities). On the one hand, there is nothing fundamentally wrong with this — metaphors are helpful in creating familiar environments that users grasp more easily. But the metaphors used in most online collection interfaces (and the underlying content management systems) facilitate a linear and static experience, which excludes non-standardized practices that extend beyond the opening of a thumbnail preview to view the high-resolution digital reproduction of a painting, for example. For the most part, existing design metaphors in collection interfaces cannot account for the processes involved in the preservation and presentation of the networked and interactive properties of net art.
M.G.: Are there significant examples of project that you think interfere/subvert the way we traditionally interact with artworks online that stand out in your research?
L.R.: If we take a broader view on art representation online (not net art specifically), I think the notion of generous interfaces, i.e. interfaces that present users with different ways to browse and not just search through collections, has been an important way to interfere in more traditional approaches to collection interactions. The generous concept was originally developed by designer and academic Mitchell Whitelaw, and significant further developments of the concept can be seen in the work of influential designers and researchers like George Oates, Florian Kräutli, and Olivia Vane. Generous interfaces tend to feature different forms of data visualizations that not only highlight new ways for users to browse and discover works in online collections, but can also reveal new connections across items, or related contextual metadata such as people, institutions, artistic movements, etc. The adoption of linked data alongside data visualization in projects such as Connect Vermeer (also featured in my report) provide interesting opportunities to extend the generous approach beyond the client-side (i.e. the execution of interactive code on-the-fly inside the user’s own browser based on user interactions), and to encode various data interconnections already at the data model and database level. The latter opens further opportunities to connect data across silos (of institutional websites and individual web pages) and extend the capabilities of a user to access art-related information across different sources through a single interface. However, all of this is still only applied to collections of largely digitized material rather than born-digital art.
In terms of representing net art works, some of the more significant examples from the body of work I researched include an early, and now largely defunct, online exhibition titled Kingdom of Piracy. Curated by Shu Lea Cheang, Armin Medosch and Yukiko Shikata, it was originally commissioned by the Acer Digital Art Center [ADAC] in Taiwan for ArtFuture 2002. This project reimagined the presentation of the works amid a stylized blue “sea”, where vertical scrolling allows users to discover (and here I am quoting the curatorial concept of the project) “links, objects, ideas, software, commissioned artists’ projects, critical writing and online streaming media events”. Although most of the links are no longer accessible, the exhibition and curatorial statement remain relevant and poignant examples of both the issues net art addresses (information economies on the web, intellectual property, ownership and authorship, among others) and how these issues challenge and subvert user expectations.
Another important example for my research was Rhizome’s own Net Art Anthology exhibition. By the time I started my research the exhibition had already been largely planned, so I got to witness its temporal unfolding over 2.5 years. The most significant aspect of the exhibition — from an interaction design point of view — was the use of browser-based emulation to exhibit net art works in environments contemporaneous to the time of creation of the works. This type of contextualized user experience had previously only been available in physical exhibitions (using old or adapted hardware for example), but not in an online setting. So in this sense, the NAA exhibition was piloting a new type of interaction with net art, I believe.
Of course there are many brilliant examples of net art works, or artist-led projects that involve and evolve over multiple works (or variations of works), that subvert user expectations of what we are supposed to do when we encounter art online (many of them are in fact featured in the NAA). I have drawn inspiration from such works to develop my thinking about interfaces beyond the metaphors of windows or transparent screens. But in terms of specific design and interaction patterns, my research focused primarily on institutional examples of online archives and collections, and those I found rarely, if ever, attempting to challenge traditions.
M.G.: After the explorative times of the early web, followed by the sociality of the Web 2.0 and its more user-friendly interfaces, we now navigate a quite different digital landscape. The online environment is now often made by self-contained, commercially-driven and product-oriented worlds that dictate more strictly users’ behaviours. How do you think our relationship with interfaces is changing?
L.R.: Yes, the walled gardens of apps and other cloud-based delights! In many ways, I think what we have now is the logical continuation of a computing paradigm that was developed in the labs of early Silicon Valley corporations as far back as the 60s and 70s. I would like to bring in the concept of transparency here, because it plays a central role in my own research, and because I think it is central to the question of what is our relationship with interfaces and what it could be.
Within the field of HCI, transparency has traditionally been framed as a necessary condition of a good interface, one that reduces visual complexity in the interest of a user-friendly interaction experience. Transparency means the interface, associated software and hardware infrastructure, and all decisions framing that infrastructure should remain invisible, so that the user of the interface can focus only on the task at hand: typing a document, watching a video, etc. This may be counter-intuitive, since we typically associate transparency with making things more visible, open to scrutiny etc, especially in terms of e.g. governance. But in terms of interface design, the more ‘transparent’ an interface appears to its users, the more opaque the underlying infrastructure is rendered.
In Reading Writing Interfaces (a text I highly recommend), Lori Emerson traces the history of the transparent interface paradigm to the notion of ubiquitous computing — an early computing concept that envisioned the future of interfaces hidden in walls and office furniture — ubiquitous, yet invisible, transparent. The premise is that this interface and infrastructure are purely ‘technical’ constructs: value-free and neutral; and therefore, not worthy of being visible. Of course, this perceived neutrality is an illusion, which has been eloquently unpacked by many critical media studies scholars and net artists. And yet, I think with every decade of computing developments, this insistence on transparency, on hiding away the infrastructure — whether with buzz words (the cloud), or fashionable accessories (wearable tech), if not exactly the office furniture — endures and gets even more pernicious (see Olia Lialina’s excellent series of essays starting with Turing Complete User).
This is a long detour, but thinking through the concept of transparency and its relation to the user-friendly ideology is essential in understanding the business, as well as design development of digital products today. Transparency has its place in interface design — some level of abstraction is useful, as users do not always need to be aware of every single operation a computer executes in order to serve a particular GUI. But a blind pursuit of transparency (whether because obfuscation is part of the official business strategy, or because reusing Google’s material design framework, for example, is easy and saves time for smaller product companies) leads to extremely formulaic and limited patterns of interaction for end-users. Although we can still celebrate the ingenuity of young people to work within the limitations of apps like Instagram or TikTok towards highly creative outputs, an interesting trend has been popping up on my networks lately — university educators are struggling with incoming students who do not understand the basic principles of computer files and folders. The file and folder system of desktop computers is itself a layer of abstraction that makes interfacing with the computer more “user-friendly”. But at least there is still a connection to materiality — to bits and bytes having some sort of persistent storage, which has a concrete physical infrastructure. With the move towards apps on mobile devices, or even browser-based apps that operate with data in transparent ways (i.e. ways we don’t see unless we dig into Inspector tools and have a good understanding of server infrastructures), user interactions happen at ever higher levels of abstraction. This abstraction can and does prevent users from exercising agency over the way online products operate with our data. Enmeshed with personal interactions, this data extends to far more than just office documents in files and folders these days. So ironically, as our relationships with interfaces become more and more personal, we (collectively) as users become ever more detached, intentionally held at arms’ length, from the operations of these interfaces.
M.G.: What does subverting given interfaces mean to you nowadays?
L.R.: I think subverting interfaces nowadays should be about users exercising agency in their interactions with the networked flow of information. And they do — even if they don’t have control over how corporations extract, store, sell and generally operate with their personal data — they can actively sabotage this through purposefully performing misleading acts. Most often we see this being enacted by artists as part of a performance (e.g. Jonas Lund’s performance Operation Earnest Voice which turned the Photographers’ Gallery in London into an “influencing agency” pushing propaganda to “reverse Brexit”), but it has been done by coordinated civil group actions as well (e.g. K-pop fans flooding Twitter to diffuse racist hashtags, among other forms of collective creative disobedience online). As much as I find this type of actions inspiring and noteworthy, as a designer of interfaces I can’t help but want more — more capacity for user agency built directly into the design of the interface, and the underlying infrastructure.
That’s why I’m interested in studying transparent interfaces, I want to design another form of interface; not necessarily purposefully difficult, or user-unfriendly interfaces, but rather interfaces that act more like mirrors, and not windows — mirrors that reflect the socio-technical context of the computer product and its user (as conceptualized by Jay Bolder and Diane Gromala in their text Windows and Mirrors). And of course, to give a concrete example, I can draw on my work with interfaces for digital archives and collections. With the redesign of Rhizome’s ArtBase, we chose to use a backend infrastructure that structures data following the principles of linked open data. This is quite different from most types of traditional databases which structure data in very prescribed ways. With linked open data, we have the possibility to develop a classification ontology around artworks, software, artists and organizations that is not fixed, but can grow organically: simultaneously accounting for all previous variations of the archival organization schema, and allowing new forms of classification to be added. There is also version control and the option for provenance tracking. These are not only backend functionalities, but aspects of the data structure that can be made visible to users in the frontend interface, too. And this is what the new ArtBase aims to do. Instead of flattening the user encounter with the work to a single canonical version and a set of static metadata fields, users can access multiple artwork variants operating at various degrees of functionality; they can access metadata statements and contextual narratives as well as the sources of these statements and narratives. They can click on metadata fields and read more about the choice of terminology, or classification. All this added potential for interaction certainly makes the interface less ‘transparent’. But instead of talking in terms what is or isn’t user-friendly, I think it is more productive to talk about whether this interface succeeds in its attempt to invite users to enter a a , to linger and spend time with the hyperlinked information and media rendered via their browsers. The new ArtBase interface is by no means perfect, or perfectly successful in its aims, but I’d like to think that with continuous community engagement and further design development, it can be a model for what Johanna Drucker refers to as humanistic interface design. Such a model would draw on critical theory from the humanities, in addition to and as an enrichment of traditional HCI methods, in order to support the act of interacting with an interface as an act of interpretation, as opportunity to enter a discourse, and to exercise critical agency, too. And on that note, I’d like to end with a quote from Drucker’s essay on Performative materiality:
“Objects exist in the world but their meaning and value are the result of a performative act of interpretation provoked by their specific qualities. To say that is merely to remind ourselves of what we already know: that we need to recover the lineage of critical theory that transformed the humanities from structuralism onward to understand digital objects and to design them.” (Johanna Drucker, 2013)
M.G.: Particularly during the Covid-19 pandemic, we have seen a tendency of replicating the experience of viewing art in isolation, instilling a sense of contemplation without interference from the outside world. This has given form to viewing rooms, 3D rooms, sound-filled navigable spaces. I found this odd since there are many ‘historical’ examples of interface design that respond site-specifically to the diversity of audience engagement online. What has been your experience?
L.R.: To be honest, I think I entered the collective zoom fatigue during the peak of the pandemic and did not actually have the patience to engage much with the offerings of traditional art institutions. It’s funny that you mention that there has been this tendency towards the isolating experience, maybe that part I missed. Thinking back on 2020, there were maybe three of four online art presentations that really stuck with me, and interestingly all of them had some form of ‘sociality’, i.e. you experienced the space in relation to others, too. An example of this was the new online platform of Amsterdam’s Upstream Gallery, which brought together net art artists from several generations and hosted great online openings and guided tours wherein you could see and interact with others in the audience. I also remember the online programme/platform Pohflepp in Practice: Revisiting the work and world of artist & designer Sascha Pohflepp hosted by the HKW in Berlin, which ran on open source software developed by Trust Berlin. It also had explicit temporal unfolding (with live events) and the sense of being in the space together with others. In both of these cases interface design played an important role in the overall experience of the events, and while it supported the artworks/live events in an appropriate manner, it refused to be just a blank non-entity, like a virtual white cube. On the contrary, the distinctive aesthetics of the interface is what made these virtual spaces more memorable, and engaging to me, as a member of the audience.
M.G.: Which roles can interface design and interface studies play in today’s web in the context of artistic and curatorial production?
L.R.: As the examples I just mentioned show, interface design plays a significant role in the presentation of individual works, as well as group curatorial projects online. We have seen net art exhibitions being produced for both online and offline spaces for several decades by now, but as more and more work is being created with the explicit intention to be shown online (whether for COVID-related reasons, or in response to the growing commercial market for digital works), there is a need to think about new forms of presentation that take into account what high speed networks, powerful graphics cards and high-resolution displays offer that maybe wasn’t technically feasible 15, or even just 5 years ago. But I also want to pose the question as to what degree are these ‘new’ affordances equally distributed, or not; and what could be the role, or potential trade-offs, of alternatives to the high-speed/high-res model, like the movement for ‘minimal computing’, vis-à-vis energy consumption, for example. These appear to be infrastructure questions, but interface design has always been tightly connected to infrastructure, even if questions of aesthetics (e.g. is it minimalist or skeuomorphic; is it flat or multi-dimensional) tend to dominate discussions when we talk about design. So thinking about these interface-infrastructure interdependencies, while keeping in mind the constant danger of obsolesce, I believe that artistic and curatorial work should be based on informed choices that include considering how works can be presented and contextualized online, but also how can they be preserved and archived outside the context of the temporary online show. And given this multi-decade history of net art production and exhibition, both artists and curators can look back to key moments from the historical development of interfaces and recognize patterns that get repeated over time (propelled by hype-cycles), in order to make informed choices with regards to the development of new works or new exhibitions which aim to move the discourse forward, rather than perpetuate older debates. Of course this ability to ‘look back on’ depends on having reliable sources and archives. And here I would like to point to the ArtBase again, because besides an archive of net art, it is also a valuable archive of the evolution of interface design practice. Many of the works represented in the ArtBase engage with contemporaneous interfaces to various degrees, whether to critique or exploit their structure and design for creative purposes. So I also see the role of the new ArtBase not only as a model design framework, but as a research tool that can inform the development of future interface design paradigms.
I am also curious to flip this question around and think about what interface studies can learn from artistic and curatorial production. Let’s use a concrete example. This will potentially date my answer to a very limited context, but in the last few weeks (or few years, if you count the built up to it) we have been hearing the word “meta(verse)” a lot. A full discussion on this is beyond the scope of our conversation today, so I will only speak about a phenomenon I observed in my twitter bubble emerging pretty quickly. People with backgrounds in the arts, curation, and related academic fields, have been bringing up poignant examples from the past three decades and more, and getting the point across that this ‘new’ mode of interfacing with computers is not so new at all, and has in fact been the subject of both commercial flops as well as deeply imaginative and inspiring artistic production. Analysing why things flopped or what made certain artistic projects compelling, could inform how designers think about interfaces and make decisions for the future regarding what to design, or what not to design, how, and to what ends. There’s more I could say, but I think I’ll end here.
Thanks so much for allowing me space to think through these interesting and important questions with you for the Curating Online platform!