Structuralism and a Pattern Language for Online Environments
I scrolled through first Instagram and felt my self-worth shrink even as my eyes gazed hungrily at skinny, bikini-clad influencers on exotic beaches. I looked on Facebook at articles of people who I disagreed with, but was silent and scrolled past. I flicked left and right on soon-to-be-forgotten men.
I felt distracted but utterly unfulfilled. I thought of the short story “I Have No Mouth, and I Must Scream” by Harlan Ellison in which a computer that has taken over the world tortures a group of humans. Networks have indeed taken over the world, but instead of outright torture we are caught in the webs of systems that change us and our world but without the words to put exactly what is wrong, or the words to change them. This world might be titled “I Can Scream but I Have No Words.” Words build bridges and new worlds. Screaming breeds divisiveness and fear. Without words to place what is wrong, little can be changed.
I thought I could find these words in the field of information architecture. Today there are two common approaches to the study of Information Architecture: Little IA and Big IA.
Little IA emerged from library science, and is focused on using labels, navigation, and taxonomies to help people navigate swiftly through websites. This narrow focus on navigation misses the multitude of other dynamics that are influenced by the design of information systems.
Meanwhile Big IA emerged in the consulting boom of the 1990’s and was positioned as highly process-oriented. Information architects in this camp utilize card sorting, tree testing, and user interviews to create tailored site maps for organizations. Big IA’s fixation on process has meant that companies need to hire consultants for each new product, which might help with job security, but has also stymied the creation of reproducible structures that can be used across different products.
I could not find these words in Big IA or Little IA, and so when I strove to build I relied on what existed (Facebook for X), which meant I was always tied to existing systems. Or, I was drawn on by the process of iteration. Each time getting incrementally closer, but it felt slow. Slowly, I built structures that I used in my own work to conceptualized information problems and design information systems. This is what I share with you now: a third way to look at information systems.
This approach to information architecture is called Structuralism. Structuralism is the belief that there are repeatable patterns for information structures that can exist outside of any one product or system, and a core goal of information architects should be to discover these patterns and understand how they interact with each other.
This approach is different from Little IA in that it situates the information architect as the designer of all dynamics online, not just navigation. It is different from Big IA in that it is focused on finding structures that hold true across systems instead of being tailored to specific products, much like a singular language can be used to write innumerable books.
This paper lays out some of the first “words” in this generative grammar, much like Christopher Alexander did in his book “A Pattern Language” for cities, where he described basic building blocks of city spaces, such as plazas, cafes, and sidewalks. But the online world is new, and the development of distinct “words” will go on for years. Instead of creating an exhaustive list, the concentration within this paper is on the methodology to create new words and how to combine them into larger structures.
This information grammar can help information architects design online systems more confidently as they have clearer ways to move from the interaction desired to the information structure created. A language of information structures I also believe can create a new era of democratized design, personalized networks and augmented spaces.
Part One delves into further background on Little IA and Big IA. Part Two speaks about Christopher Alexander and Richard Saul Wurman whose ideas on physical spaces impacted the creation of this language. Part Three breaks down basic components of the language. Part Four analyzes the existing network of Instagram through the lens of structuralism. Part Five looks at the implications of this language for the field of Information Architecture and the larger world.
This section provides a deeper overview of the approaches of Little IA and Big IA currently take to the approach of information architecture. Understanding these approaches can better orient readers to the shift towards structuralism.
The current perspective of Little IA practitioners’ approach to information structures is similar to the perspective urban planners in the 1950’s brought to the city. Both perspectives share the desire for efficient navigation. This viewpoint in urban planning manifested in the development of many new highways and car-oriented policies. The most notorious of these efficiency-oriented planners was Robert Moses, who created the Brooklyn-Battery link, the Triborough Bridge, and the Cross-Bronx expressway. These projects decimated urban neighborhoods and the communities that lived there.
This seems like an unfair comparison. When an information architect reorganizes a website, their actions hardly have the same life-altering repercussions. But the fixation of information architects on navigation has done something else: it has meant that many of the potential structures of a vibrant online world have never even been conceived. I do not mean that these structures have not been built; there will be new, different websites created for centuries. I mean that there are important, fundamental ways to interact in the online world that have not been thought of yet. It is not just that the online world is like a new city that only has the basic infrastructure of roadways set up. It is like a world where we have no comprehensive understanding of how to build the kinds of places that function like a house, or a public park, or a town hall.
Without these spaces online we will increasingly be locked into a consumerist, monopolistic world because we have only designed the online spaces that create this one kind of world. The online world will increasingly look like the top-down, artificially constructed cities of Abu Dhabi or Xiang’an with ample stores (Amazon, Zappos), spaces for passive consumption (Hulu, Netflix, Youtube), and places to show off (Facebook, Instagram). People can descend into rabbit holes of their own opinions with the help of algorithmic feeds, be followed across the internet’s streets by cookies, and craft posts that have no long term value for themselves or anyone else.
The online world lacks emergent pockets for creative play, cafes for respectful debate, newsstands where people can purvey a broad range of opinions, reading nooks for quiet reflection. As we need public spaces in the physical world, so too do we need these kinds of spaces online.
But unlike the physical world, there is no Jane Jacobs to rally communities to fight for the cadence of the sidewalk. This is in part because we are not quite sure what a digital sidewalk entails. Information architecture has not yet given people an effective way to enunciate how their desired interactions online translate into digital networks, and it is hard for people to fight for what they cannot explain.
The first step to create this language is to push information architects to look outside of the topic of navigation. Information Architects should concern themselves with the design of all the potential interactions created by information structures such as empathy, creativity, community, play, debate, amongst many others.
The second step is to look outside of research on specific user groups or navigation maps tailored to specific products to develop structures that exist exist outside of any one user and exist across many products.
These patterns have yet to be developed by Big IA because it is an approach that is beholden to the demands of the user. This captivity is a result of where the discipline is situated between business development and user experience design within companies.
Information architecture’s location near business development has been heavily impacted by the current lean start-up methodology and the idea of iterative design. In iterative design, product teams develop a minimum viable product, test it on users, develop the idea further, and test it again. This ideology is opposed to the development of continual structures that exist outside of the user. This process also results in product teams searching for structures that emerge slowly and uniquely to each project. This development process makes it harder for designers to see similarities in online structures across products.
While users should be an important part of design decisions, the people who use websites do not have a language to talk directly about the interactions they want. What many designers face is that users say things like “I want a Facebook, but for my neighborhood community”, “I want Uber, but for doctors”, “I want Snapchat, but for seeing events around me.” To liken it to cities, it is like someone saying “I want the Mall of America, but for my neighborhood,” “I want Times Square, but for my city,” “I want the East Village, but in the Midwest.” In each of these, people are using specific examples of places or networks that get across a piece of what they want. What they want is something that might exist in each of these places, but there is currently no modular way to describe pieces of online interactions.
Big IA is captured on the other side by user experience design. This kind of design is highly focused on how an individual experiences a website or app. One of the basic concepts in user experience design is affordances, or visual indicators of what actions people can do. There is also a focus on user-oriented principles: provide users feedback for each action they take, make buttons large enough and visually consistent, align user expectations to website actions. All of these ideas are important to get an enjoyable experience, but it does not go deeper to explore how the structures of the various pages create different dynamics between people and contribute to larger organizational structures.
The creation of a cohesive design language would give users more power to directly express what kinds of interactions they want to have. Now, we turn to the development of this language.
This section looks at two important individuals, Christoper Alexander and Richard Saul Wurman, whose ideas of modularization and functionalism created the basis for this new language.
The design of this language begins with Christopher Alexander and the modular approach he took to the city. In his book “A Pattern Language” he broke down “common problems which occur over and over again in our environment, and then describes the core of the solution to that problem, in such a way that you can use this solution a million times over” (x, A Pattern Language). In this book he collected 253 patterns such as neighborhoods, shopping, sacred sites, promenades, row houses, parks, that can be endlessly customized and connected together in different ways to form different kinds of cities.
These patterns were highly focused on physical features. For example, a Terrace on the Street pattern called for a “terrace, slightly raised, connected to the house, and on the street side” (xxxvi).
The pattern of Parallel Roads meanwhile should be roads that are situated 100 to 400 yards apart. Pedestrian pathways should be kept “at right angles to the parallel roads and raised above them where the two must run parallel” (130).
These ideas of modularity impacted computer science with the creation of modular programming. Developers who write to subdivide a large computer program into separate sub-programs. A module, like one individual pattern, can be “used in a variety of applications and functions with other components of the system.” Modular programming is the idea that instead of creating one large program in which every section is linked to every other, the program is created from many different small programs. These smaller programs make it possible for developers to make high quality components, like menus, headers, and sidebars, that one developer can create and then many developers can re-use. This saves developers a lot of time because it means each individual does not need to reinvent the wheel.
Modular programming as similar to Alexander’s modular approach to the city. Both enable individuals to break large concepts of city or computer program, into smaller, reusable parts that enable development to happen faster. This is the first goal of this new language: it will enable people to design faster.
But Alexander concentrated on the physical components of the city. The size of the sidewalks, the orientations of the streets, the height of buildings. This is akin to interaction designers talking about scrolling, profiles, button sizes, image wheels, posts. These are parts of websites, but they speak to the visible parts of websites rather than the interactions they enable.
Alexander broke up cities into modular components.
Richard Saul Wurman in his “Time for Cities” speech outlined how we “ask for parks instead of places for recreation.” We use the known words of parks, but are actually trying to convey the desire for several different kinds of interactions, such as brief positive interactions amongst strangers, desire for information on new happenings in the community, creative play with close friends.
As Alexander broke cities into modular pieces, Wurman oriented these pieces around function and interaction patterns between people, rather than the physical objects that make up the system.
When you begin thinking in terms of the interactions that go on in a place, rather than
These are all typical
Modularity of Alexander’s helps people better recombine concepts, while the goal-oriented orientation of places that Wurman puts forward helps people to rethink what kind of spaces can exist.
Goal-oriented building can create such places as:
We now have some of the basic structures of the online world: Facebook, Reddit, Instagram, Email. Many people when they
building blocks of the online world: pages, posts, feeds, groups, and many others. The problem here is that these modules are much like the park benches of the physical world. They talk about the physical of what people can interact (what the user sees), but they do not speak about the interactions that people have there. This makes it harder for individuals to move from the kind of community they want to create, to understanding how to do this.
In the “real” world we can easily identify what play is. But in the online world, EVERYTHING is artificial. There is, unlike the real world, no direct links. The idea of benches, tables, sidewalks, have been engrained so deeply in our heads that it is hard to conceive that there was once a point where those objects were not commonplace.
This section delves into the fundamental components of this new language. It then uses this methodology to analyze the existing infrastructure of Instagram, a popular social network.
The basis of the language are objects. Objects are anything that can be conceived of as one. A salad can be one object. A chair can be considered one object. A restaurant can be considered one object. What is considered one changes based off of what kind of network you want to create and how you decide to organize information. If you want to create a website that helps people decorate their rooms, one object would be a single room. If your goal is to track energy usage of buildings, one would be a building. If you want to map out various neighborhoods in your city, one would be a neighborhood.
Objects can be added, subtracted, changed, and structured in many different ways. Object groups are multiple objects that are linked together in some way.
Objects are connected to each other via channels. Channels are connections between objects that exchange information. Channels can be designed so that they give very little information or a lot of information. Telegraph machines send only dots or dashes through the channel of air waves. Meanwhile, on Skype those same air waves can send video with depth, color, and audio.
Channels can be designed with incentivizers. Incentivizers fine tune the actions people undertake in a system. Some incentivizers constrain the actions of a user, like prompts that impact how people answer. Other incentivizers push forward users to do more of an action, like the Reddit karma incentivizes people to post content on the platform.
The combination of a person or object, a channel, and the incentivizers are called basic dynamics.
(Person or object) + channel + incentivizer (optional) = basic dynamic
A user that does an action and gains feedback from this action constitutes a full loop. Generally, platforms strive to have more activity on them, or more loops.
Blocks are the combination of groups of objects and basic dynamics with singular goals. If you have a one-goal system, like a simple calculator, an alarm clock, a basic weather app are all examples of one-goal systems. These systems are also similar to coding in terms of being one function, or a city in terms of
All systems are also made with goals. The primary goal of a network heavily impacts design decisions of a network, and how designers decide what objects to measure. Some online spaces have very open goals, where the biases, context, and thinking of the user heavily impacts the interactions on the network. Other platforms are goal-constrained, where users have very specific options they can take through a platform.
A multi-goal system has many different goals that people can complete. An example is Facebook, where people can do numerous things like message, post, shop, all on the same platform.
A block consists of a combination of different groupings that combine together with a goal. A platform consists of many different interaction blocks that are all linked together to form a larger goal.
This language creates a clearer place for Information Architecture within the design professions. As user experience design is the bottom-up study of how people interact with online environments, information architecture is the top-down design of information structures.
Currently, interaction designers, or user experience designers, create wireframes of Instagram’s screens which they test with users.
UI of Instagram Screens:
These wireframes are then connected by information architects into site maps.
Site map of Instagram:
There is a gap between the user’s experience and the general links between pages. There is the dynamics between people, and between people and objects, that are not expressly taken into account. While on some level designers might understand these flows, the lack of explicit statement about the designed interactions makes it harder for
This language is interaction-focused, which can help designers in organizations more confidently test out different concepts, and can help urban planners better design physical spaces for interaction rather than the current zoning we have now.
The rise of a structural information architecture language will help information architecture practitioners design faster, in an approach called modular network design. Modular networks in turn will impact what can be designed, giving rise to personalized networks and evolving, augmented physical spaces. This section sets forward how this can be accomplished and the kind of world this can create.
Recombinable architectures can democratize design in the form of modular networks. Like Google attempted with Project Ara’s modular phone blocks, where the design of the phone allowed people to enunciate what kinds of interactions they wanted with their phones and change them based on what they wanted, so too do I believe this modularization on a digital level will create more flexible, personalized online networks based on people’s individual needs and interests.
Currently Reddit, Facebook, Instagram, dominate the time people spend online. In the near future, people will have their own individual networks, which will change throughout the day, change based on their location in physical space, and change based on each interaction they have with the people around them.
Example of this in online space:
(diagram of modularly built network).
Link to physical spaces. It can help urban planners design better — -which will be necessary as online networks have an increasing impact on our interactions in urban spaces.
This design language I also believe will impact physical spaces with the creation of augmented spaces. Public plazas can have associated digital infrastructure that changes over the course of a day, a season, and is reactive to the people that are there.
I imagine you will be able to walk through your neighborhood and you see a cute dog at the dog park, so you post a picture to the Community Board. There you see an announcement for a bake sale next week. You navigate to the Community Board’s Events page and read through the list of what is being made. You put down that you will bring brownies.
A news story on the Community Board has gotten enough attention that it has been bumped up to the County level. You think the issue is not that important, so you click that it is over-sensationalized. Anyone can submit to the Community Board, and anyone can read stories immediately in their area. You have to pay to see stories outside of your zone
You get to the park. You have a class on plant flora and fauna, and they have a library of different sign-posts of the plants there. You can swipe through on your phone through the various different seasons and see an overlay of what the most common plants are. You spend an hour walking through the park and seeing what plants are there. You make sure to hit the landmarks where your professor has placed down notes. At the end of the hour, you see an odd mushroom that has not been catalogued. You take a picture of it and add it to the class library.
You are also a big gamer. You’re currently playing a city-wide game of Running Man, where your Fitbit points translate into ammunition. You’re alerted that someone in the game is nearby. You search around and see a guy walking his dog up ahead. You’ve been walking for the last hour so you are charged up! You take aim on your phone and fire! It’s a direct hit. You see him look up and you smile. He pretends he’s taken a direct hit. After the show you both wave good-bye at each other and pass on.
Places can have digital layers, like a layer of information about the nature of a place, a game layer that connects people in the plaza in a specific game, a layer of community where people can debate issues that impact the community. Each person will develop their own networks, which they can specify how they will interact with their friends and the world around them. These building blocks I believe can create a more creative and interactive world.
My next step for this is to get this article published and begin development of a book of many different interaction patterns. Then I want to start building these modular networks and augmented spaces at MIT’s Media Lab.
I have included links to articles that have impacted my thinking on this subject:
Structuralism and a Pattern Language for Online Environments
Research & References of Structuralism and a Pattern Language for Online Environments|A&C Accounting And Tax Services
Source
0 Comments