April 27 2023 (a 17 minutes read)
This is a reworking of the original draft of the chapter I wrote for the “Advances in
Information Architecture” book. It discusses the ongoing evolution of information architecture in terms of a shift — under the pressure of social
and technological change — from the “classical” information architecture of the 1990s and early 2000s to a “contemporary”
If you cite or quote this text, please refer to the original published chapter: Resmini, A. (2021) Classical to Contemporary: An M3-based model for framing change in information architecture. In Resmini, A., Rice, S. A., and Irizarry, B. (eds) Advances in Information Architecture. HCI Series. Springer.
“Reframing Information Architecture” was the title of the first Academics and Practitioners Roundtable  which was held in 2013 as one of the pre-conference workshops at the 14th ASIS&T Information Architecture Summit in Baltimore, Maryland. The goal of that Roundtable was twofold:
That first Roundtable, and the global conversation it instigated, contributed to push our understanding of information architecture beyond that “illusion of the web as a library” (Resmini 2014, p. v) and the idea that “library and information science is the only relevant body of knowledge for the discipline” (Lacerda et al 2018).
This reframing opened the field to new methods, concepts, and contributions coming from a host of different disciplines — architecture and systems thinking among others — vastly broadening the concept of what the “information spaces” that the field is interested in are. It formalized, or made cohesively visible, how information architecture had been slowly moving “away from the single artifact, the website” to consider the “entire product or service ecosystem” — parts of which might not be online or even digital at all — and strategic and organizational problems that are far removed from the idea of just producing “blueprints for the web” (Wodtke 2003).
The 2013 Roundtable resulted in the publication of a book by the same name in 2014, documenting the event and presenting eleven contributions that illustrated the changing nature of information architecture practice and research. One of the chapters, “Information Architecture as an Academic Discipline”, was written by Flávia Lacerda and Mamede Lima-Marques. Lacerda discussed it at the Roundtable, describing how Van Gigch and Pipino’s Meta-Modeling Methodology (Van Gigch & Pipino 1986), often shortened to M3, could be used as a way to frame the debate on how to advance the discourse on information architecture (Lacerda & Lima-Marques 2014).
Lacerda and Lima-Marques’ revisitation of the M3 offered a way to interpret and explain the relationship between the practice and theory of information architecture that resonated with the discourse within and around the field: it provided a model that could be used to understand past failings and to prevent making new ones, making the Roundtable a real “round table” where both the industry and academia had a place and a voice. Unsurprisingly, it became the conceptual backbone for all eight Roundtables that followed between 2014 and 2021.
The M3 is a conceptual model meant to explain how innovation processes work: it adopts a social perspective and applies to any field of knowledge (fig. 2). It is structured around three individual layers that operate at different speeds and create a system of flows and stocks (Meadows 2008) that influence each other via their input and their output.
In the meta layer, the layer concerned with epistemology and the one slowest to change, we find the fundamental concepts that structure the worldview of any given field. This layer identifies valid “episteme”, knowledge, and differentiates it from “doxa”, opinion. When big changes happen, called paradigm shifts, this layer changes: a good example is the shift from the idea that the Earth is the center of the universe to Copernicus’ heliocentric, or star-centered, cosmology (Floridi 2014, p. 87).
The object layer, that of science, is where systematic studies organize and stratify knowledge through observation and experimentation and where the theories and models that are used “to describe, explain and predict problems and their solutions” are crafted (Lacerda & Lima-Marques 2014). This is where, for example, Darwin’s theory of evolution or Anderson’s 2004 “long tail” reside in the M3.
The application layer, that of practice and the fastest to change, is where we find the solutions to everyday problems encountered in a field. It’s where practitioners operate as they exercise their craft — surgeons, shopkeepers, and of course designers — and where concepts from the meta level and theories from the object level are concretely applied.
These three layers feed into each other: through time and accumulation, the “solutions to problems” of the application layer become “evidence” for the object layer. Here, they might change theories and become evidence for the meta layer, producing in the long term epistemological change.
This means that the “guiding” role of the epistemology level, which acts as a set of constraints that establish what is valid but also “acceptable” knowledge in a field in any specific historic moment, is counterbalanced and slowly undermined by a constant trickling up, bottom to top, of practices that first make it to the object level, where they are validated, and then possibly to the meta level where they ultimately challenge the current understanding of what “knowledge” is.
In doing so, the M3 provides a way to address “the existing gap in Western culture between ‘episteme’ (science and ‘knowledge that’), which is highly valued and respected, and ‘techne’ (technology and ‘knowledge how’), which is seen as secondary’ by making them part of a systemic process” (Lacerda et al 2018). Bridging these, making them part of one single conversation, is a rather important part of the ongoing efforts to reconcile the different views and goals that the practice and research and education in information architecture express as of today.
Looking back, a period of rapid transition in information architecture can be identified beginning around 2008 or 2009, and made to roughly coincide with the commercialization of the Apple iPhone and the subsequent introduction of the App Store. It is in these years that a coherent and continuative conversation begins (Alfrink 2007; Rosati & Resmini 2007; Hinton 2008; Potente & Salvini 2008; Hobbs 2008; Wodtke 2009; Resmini & Rosati 2009) that consciously and deliberately transcends the original formulation of information architecture as a way for “understanding and conveying the big picture of a web site” (Rosenfeld and Morville 1998, p. 14).
This period of transition follows what could only be called a period of epistemological crisis for the field (Resmini 2013), coinciding with the initial wave of interactive web application, Tim O’Reilly’s Web 2.0 (2005), and the widespread diffusion of folksonomies (Quintarelli 2005).
In “Pervasive Information Architecture” (2011) and in “A Brief History of Information Architecture” (2012), Luca Rosati and I, revising and extending earlier seminal work by Ronda León (2008), posited that it is possible to distinguish three initial knowledge streams contributing to the genesis of information architecture: an information design one, rooted in Wurman’s work from the 1970s onward (Wurman 1989; Wurman 1997); an information systems one, which became prevalent in the 1980s; and an information science one, that of Rosenfeld and Morville, which operated a “synthesis” and became mainstream in the 1990s. We called “classical information architecture” the information architecture of the years between the early 1990s and the early 2000s, and “pervasive or ubiquitous information architecture” the information architecture that emerged from the period of crisis outlined above (2004–2006), pointing at seminal contributions such as Morville’s post-Argus “Ambient Findability” (2005).
With the non-trivial benefit provided by an additional decade of hindsight, I suggest that the reframing of information architecture that happened in the 2000s should be understood in terms of the more reasoned chronology detailed above. I also posit that “contemporary” is a more neutral and suitable term for what we called the “pervasive and ubiquitous” period. Then, I propose a differentiation between classical and contemporary information architecture modeled in accordance with the conceptual framing of the M3 (fig. 3).
At the meta layer, classical information architecture is anchored to a digital-centered view of the world, which is disembodied, kept behind screens, and still rooted in the cultural dominant of the time, postmodernism (Resmini & Lindenfalk 2021). This is for example visible in the conceptual construction of the object of design as an individual, finished, authored artifact (Kirby 2009) — the website — but also in its acceptance of “open and ambiguous structures” (Beardon 1994) as part of its own foundations (Bowker & Star 1999).
At the object layer, that of theory, the relevant components are being borrowed primarily from library and information science and, to a lesser degree, from graphic design and computer science. It is interesting to note the absence, in the practice, of those information systems component that were part of the early foundations of the field: something that will lead later more business- and information technology-oriented commentators to distinguish between “web information architecture” and “enterprise information architecture”, this latter representing an evolution of the research and practice threads of the 1980s (Leganza 2010).
At the application layer, information architecture is primarily a website-only affair, with software applications, either desktop or remote, being a very limited subset of the practice.
As I wrote in 2013, the classical information architecture period saw practicing information architects “choosing to define the discipline through the artifacts of the practice” — websites. This had major consequences, as it created a conflation between the field at large and its incidental, historically situated outcomes that, because of the practice-led nature of the field (Hobbs et al 2010; Hobbs 2019), spilled over into academic research and education and ultimately stifled development.
Contemporary information architecture — post-iPhone-revolution information architecture — is not a new or different beast. In Kuhn’s terms , it does not represent a discontinuity: no currently dominant paradigm has become incompatible with emerging phenomena and has been replaced by a new, different paradigm. It is rather an acknowledgment that not only practitioners of information architecture were (and are) extending their applicative reach to domains extraneous to those considered “valid” during the classical period, but that a parallel reflection, as I mentioned earlier, had (and has) been happening in information architecture research and was (and is) affecting the object and meta levels of the M3.
Contemporary information architecture sees its meta layer center on postdigital, rather than digital, experiences (Jenkins 2011; Berry & Dieter 2015). Postdigital is to be intended as the concretization of Negroponte’s far-seeing prediction of the upcoming banality of digital, written for Wired in 1998: as digital and physical increasingly blend, they form a continuous information space in which people move from a mobile app to a store, from a real-time display to a card reader, from a website to a classroom, without ever worrying, as long as everything works, whether something is digital or physical (Benyon 2014).
Because of the new relevance that the physical world, its context, and its affordances assume in the meta level, embodiment becomes a concern of information architecture and embodied experiences part of its design concerns. For example, when considering the information architecture of a Disney park (Arango 2018; 2022) or when establishing how a voice interface should contextualize conversations (Grochow 2020).
Postdigitality and embodiment, together with the anonymous mass co-production of information possible in the post-iPhone world, create in turn the conditions for supplanting postmodernism as the cultural dominant. Kirby (2009) calls this new cultural dominant “digimodernism”, a product of the effects of computing on society and culture. Digimodernism is characterized by anonymous co-creation and ownership, evanescence, haphazardness, and unfinishedness. For example, a social media platform such as Instagram is by definition an empty information architecture that requires contributions from its user base in the form of photos, comments, and social metrics such as views, comments, or hearts. It is also evanescent, in the sense that its content is in continuous flow, it “does not endure (and) it has no interest as a reproducible item” (Kirby 2009, p. 52), making it also unfinished, and haphazard, in the sense that its “future development (...) is undecided. What it will consist of further down the line is as yet unknown. This feels like freedom; it may also feel like futility” (ibidem).
At the object level, contemporary information architecture adds theories and methods borrowed from disciplines that address the increased complexity of its object of design to the theories and methods of the classical period. These have so far included, for example, architecture and urban planning (Arango 2018; Resmini & Rosati 2011), systems thinking (Morville 2015), cognitive science (Hinton 2014).
At the application level, practitioners engage with a vast array of experience ecosystems (Resmini & Lacerda 2016; Resmini & Lindenfalk 2021) that deal with both organizational (Merholz 2019) and social (Hobbs 2008) issues, and with digital / physical environments (Mandelli et al 2011; Rosati 2020).
Framing information architecture through the lens of Lacerda and Lima-Marques’ reinterpretation of Van Gigch and Pipino’s Meta-Modeling Methodology is not a point of arrival, it is a starting point.
The M3 has proved to be an immensely valuable tool for the information architecture community: it has allowed all actors involved to press their points in a systematic way while being aware of the mutual relationships between each other’s respective expertise and roles, and their relative weight at specific crossroads in the conversation.
Still, it should be treated for what it is: a model.
Classical information and contemporary information architecture, and their different characteristics at the meta, object and application level as outlined here, are also not meant to be taken as absolute categories. They represent a secondary, hopefully useful abstraction whose goal is to illuminate, under its own light of course, why information architecture today should be practiced, judged, criticized, theorized differently than in the 1990s, and thus facilitate a critical, reflective conversation on the development of the field.
We have gotten this far: the interesting question is what comes next.