Digital design is in a good place as a practice and craft. There are well-established patterns, guides, recipes, and abundant educational resources. Standards and uniformity exist where needed while leaving space for creativity and experimentation Reflecting on the progress of digital design and the possible futures ahead exposes many remaining user experience challenges and uncertainties for designers, technologists, and product managers to wrestle with.
Framing these “Grand Challenges” helps bring purpose and clarity to nagging and unfulfilled needs that are not easily solved or overcome and have the expectation of hard work over a long time. There is no shortage of topics spanning many levels of abstraction to consider.
Take them as provocations or nonsense. Please question, debate, reject, refute, or take action on them.
Artificial Intelligence User Interfaces
We have anthropomorphized AI. It is ChatGTP, Siri, and Alexa, and soon, a proliferation of agents decorated with personalities, possibly named Smith. Making AI anthropomorphic by limiting interaction to a conversational model feels lazy, heedless, or, at worst, a simplistic first attempt at which we raise our hands and say, “Ta-da! It’s done.” Conversational interfaces, while natural and intuitive, are also inefficient and ambiguous. Conversation is not the apex of communication and should not be the stopping point but a fragment of a whole multi-model interface.
There is an innate desire (sometimes crossing over into a fetish) to make AI an intelligent, conscious being. We think of an AI as an entity separate from another computer system. For example, it is popular for a chatbot to be an intermediary between a user and customer support or, more technically, between a user and a corpus or technical support documentation The interaction with AI is through natural language, either written or spoken.
A (human) conversational interaction model is challenging to design and implement The predominant chat interface minimizes the design to a text input component – a square box. A harshly barren UI replaces visual design with AI personality design, and UX becomes an enormous systems design problem. There has to be more than this.
Data Visualizations
Users need data visualizations to help them synthesize, analyze, and make sense of data. They need extra knowledge from data, but instead, UIs apathetically say, “Here is the data; you figure out what to do with it.” Product managers and designers cannot expect users to have the skills of data scientists or data engineers, yet that is the burden imposed on users, especially when data is presented in tables. Data must escape the rows of table cells in which we imprison it.
Data should be exciting, vibrant, and interactive. Give life to data! Several great practitioners of data visualizations have shown us how to elevate data into something meaningful. The challenge is to follow their leadership and make data visualization pervasive and available in more digital experiences. Designers must be comfortable designing data visualizations, and development teams must build them proficiently. As inspiration, instead of making AI anthropomorphic, let AI remain compumorphic and be the driving energy behind data visualizations.
Technical Problem Spaces
Designing with technology is ubiquitous to the point we just call it design. Designing for technology is a grand challenge. When designing with technology, technology is a material used to create a thing, service, or product for some subject domain or problem space. The focus is on the subject domain and problem space, not the material. When designing for technology, the material is the subject matter. A comparative analogy is designing a chair using wood, plastic, or metal as the materials. When designing and building digital products and experiences, websites, apps, APIs, and sensors are the materials.
Products for technology accelerate the advancement of a technology or improve the efficiency of working with a technology. They are tools for scientists, technical researchers, software developers, designers, technicians, and operators. Better-designed tools and workflows in technical domains are catalysts for technical innovation. Products for technology are necessary for making the target technology robust and stable, in other words, for making the technology market-ready and solution-ready so that the technology can be used to create products and services for other subject domains and problem spaces – so that technology becomes a material to make other things with.
Products for technology are complex and pose unique challenges. The good news is that existing design processes, techniques, and patterns apply. Software developers, data scientists, and quantum engineers have the exact needs as any other user, albeit more difficult to speak to. Like other users, they too have difficulty expressing their wants, needs, and workflows, but they will conflate their deep understanding of the technology with the best UX and UI for working with that technology. They will be suspect of the design process and have a cognitive bias towards their existing (often self-made) tools while neglecting the arduous, highly inefficient, multi-step process that is “just how it works.”
The other obstacle is the subject matter. Suppose you are designing tools for GPU profiling, quantum computers, or machine learning data pipelines, which are deeply technical and complex subject domains. You must understand how the underlying technology or developer APIs work to create great products. Product discovery and design for technology-focused products are considerably more complex than back office products for enterprise logistics and orders of magnitude more complicated than an enterprise website.
Designing for technology is an underserved problem space with many interesting challenges and rewarding outcomes. The downstream benefits reach all users of digital products and are worth more design attention.
Resolve and Purpose
In review, the themes are maturing design around AI (as we’ve done in other paradigms like WIMP or touch), better and interactive data presentation, and making technology easier to use at the deepest levels. To resolve these challenges, designers must extend themselves beyond aesthetics and empathy and into potentially uncomfortable places of data modeling, curation, and analysis and deeper into the digital technology stack. It won’t be easy, but I’m highly optimistic we can make progress in these areas in the next year.
