Less is/and more
Dr Peter Thomas (then writing as founding director of HaileyburyX), writing with Lauren Sayer (former director of Digital Learning at Haileybury and now Executive Director, Digital Learning, Research and Innovation at Melbourne Girls Grammar), about less is/and more design principles and what they mean for online learning, technology and pedagogy.
Often summarised as “less is more” (which is in fact a quote from Mies van der Rohe; Dieter Rams’ book on his work is called “less and more”) the “good design is as little design as possible” approach has been dominant, especially in technology, and especially in digital technology, for at least the last decade — and much longer than that in the kind of industrial design practised by Dieter Rams.
Grab your iPhone 11, or Samsung Galaxy, or look at your Macbook or Surface Pro, and you have in your hands the ultimate less is/and more object.
The designers of those products are, of course, aiming for ease of use. Decades of research and practice in user-centered design has brought us to realise that it’s essential to understand what people want to do — and then give them the most effective way to do it. And that means stripping away what is irrelevant.
But less is/and more applies not only to specific kinds of things (Rams is an industrial designer and van der Rohe is an architect), not just to digital technologies, not to specific types of digital technologies, or to just a particular product — and not even to the design of physical things.
Anything (and everything) is, and can (and should) be designed.
And less is/and more is incredibly important.
In this story, which is part of our HaileyburyX publication on technology, learning and edtech, we are going to talk about less is/and more as an aesthetic — a way of thinking, an orientation, or a lens through which to see things — and about how it manifests itself in and can shape thinking about design, and how it can help us look at issues in the space of technology and learning.
There are lots of people working in the vast enterprise that is ‘design’ in the broadest sense of specific types of design practice — and also in the theory and philosophy of design, along with those who do systems design, or practice design thinking — who have far more sophisticated understandings than we do. We are not claiming authority.
Rather, we are going to talk about some ways in which less is/and more has shaped our thinking, about how it intersects with online learning, technology, and pedagogy and how it appears in some of the projects we are doing and how we do them.
But first some background.
Less is/and more, as little design as possible and ‘good design’
Dieter Rams is a German industrial designer who was responsible for the design of Braun’s consumer products for many years.
In a 1976 talk in New York about his work he said:
“A designer who wants to achieve good design must not regard himself as an artist who, according to taste and aesthetics, is merely dressing-up products with a last minute garment. The designer must be the gestaltingenieur or creative engineer.”
Rams condensed his approach into 10 principles for good design.
According to Rams, good design:
- Is innovative
- Makes a product useful
- Is aesthetic
- Makes a product understandable
- Is unobtrusive
- Is honest
- Is long-lasting
- Is thorough down to the last detail
- Is environmentally friendly
- Involves as little design as possible
It’s principle 10 we are especially interested in, but as is probably obvious, the principles are all related.
Principle 3, is aesthetic, says that a well-designed product (and we can use ‘product’ in any way we like) should be, amongst other things, beautiful, balanced, calming and convey a sense of restraint in its design.
Principle 4, makes a product understandable, suggests that good design makes a product understandable to the people who use it. It suggests that — to pick a relevant educational example such as an LMS — the tools to build content make sense; they appear in the appropriate context when people need them; they help the user to do what they want to do; that they make sense in the overall structure of the product; and to use them doesn’t require people to spend an excessive amount of time trying to work them out — or being trained before they can work them out.
Or principle 5, is honest, says that products should do what they say — not look more innovative or more powerful than they are or make promises that they can’t deliver on. A product that says it will help you “stay connected and organised. Accomplish more together across work, school, and life” should really do those things (this is the strapline for Microsoft Teams and you might want to reflect how far it delivers on those promises).
And finally, principle 10, Good design is as little design as possible, is Rams’ “less but better” — another way to say less is/and more. It suggests that good design is about stripping away all but the essentials.
And although it might seem like a contradiction — do less to get better — it is how to deliver on the other nine.
Of course, many people believe that they are a ‘good designer’.
But, as Rams suggests, ‘good design’ is the result of discipline — applying principles, along with thinking carefully about the real purpose of the product as a whole.
His term gestaltingenieur or “creative engineer” (or ‘design engineer’ — the word gestalt implies being concerned with the entirety of something) suggests that good design isn’t applying a template, simple rule-following or saying I like it. It’s a disciplined creative process.
Good design is not about ‘taste’ and definitely not about, as Rams says, “dressing up”:
I am troubled by the devaluing of the word ‘design’…design is not simply an adjective to place in front of a product’s name to somehow artificially enhance its value. Ever fewer people appear to understand that design is a serious profession; and for our future welfare we need more companies to take that profession seriously.”
To take an example we can all relate to, a graphic for an Instagram post — which is a designed thing — is often littered with different typefaces, graphics, styles or layouts. If one applied Rams’ principles — say principle 8, is thorough down to the last detail — one might pay attention to each and every detail when making choices of typeface, images or layout. You can also apply any of the other principles. And then principle 10 might make one consider whether this is the simplest way we could do this: can we do less and more.
The good news then is that ‘good design’ is in reach of anyone who understands the principles and can apply them. Good design can be turned from a mysterious art into something that is available to any reflective practitioner who consistently applies the principles in a disciplined and creative way.
This example also shows why it makes sense to apply design principles to something seemingly unimportant as an Insta post that disappears upwards in a feed and is gone, maybe never to be seen again.
Because if it is not important, why bother doing it? Surely the point of an Insta post is to influence — to make someone aware of something, to present an idea, to convey a concept, deliver news, build a connection or even bring a bit of delight. And if those things are important, why not treat them as such by creating a graphic that has ‘good design’. It may be one of many graphics, created for different platforms, at different times, for different audiences, which are in aggregate part of the broader aim of marketing, or communications or brand. Which is incredibly important.
And, if we are going to apply design discipline to something as small, as unobtrusive, as a graphic for an Insta post then surely we need to do the same for everything — a tweet, a process, a meeting, a course, an online form, a building. Everything. If you have decided to do something, it must be important and so it should be as well designed as possible — down to the last detail.
So we are going to illustrate how we are applying these design principles, especially principle 10 — less is/and more or as little design as possible — in our work. As we said at the top of this story, this is a lens through which to look at things.
And what that lens does is to help us to see things clearly.
HaileyburyX: as little design as possible
Our HaileyburyX project is building online courses.
Emerging from Haileybury as a project before the pandemic, HaileyburyX aims to take the work of brilliant Haileybury teachers not just our own student community but out into the world for any student, anywhere, anytime.
As part of thinking through what that really meant, it became immediately apparent that simply converting existing courses to online courses wasn’t going to work. For one thing, our existing courses were built for face-to-face, or (increasingly) blended, delivery and wouldn’t work online as well as they did in those contexts (if at all).
Many teachers have realised this and have talked just how hard it is to do that conversion, especially when many of the technologies that teachers have to work with are unfamiliar, are not suited to the task and when the entire context — the to-and-fro dance between on-campus/off-campus — will be the norm for some time.
But irrespective of that, developing courses is just hard: distilling knowledge, at the right level, within the constraints of a curriculum, using a variety of modes — and hopefully supporting individualisation and personalisation — demands deep knowledge and expertise. It’s demanding whether you are designing a homeschooling programme or courses as part of a formal curriculum, or at primary, secondary, tertiary levels, or for professional learning.
So what should we do? Try and ‘onlinify’ courses with minimal effort and hope they work? Cast around for suitable pre-built online courses that kind-of suited our needs? Develop entirely new content from scratch using existing tools and processes?
None of these seemed to us to an approach that would satisfy our aspirations or to deliver learning that would work online, would make the most of the assets and capabilities we have already have — or most importantly, deliver outstanding learning experiences.
So we did two things.
Based on our assessment that some new ways of working would be needed for this new kind of design, we asked teachers who work on HaileyburyX to use some of the values, principles, methods and tools from an approach called Agile, now the accepted way of building software.
We have talked about Agile extensively before in our Medium publication, but what Agile emphasises, amongst other things, is placing people at the centre of developing products — in our case not only the learners who we are developing courses for but the teachers who work together to develop those courses.
One of the values of Agile is “Individuals and interactions over processes and tools”. This means letting work happen by harnessing the problem-solving capabilities of those doing it, facilitating autonomy, local and collaborative decision-making, and giving teams the tools they need to do their job. And then getting out of the way while they do it. Agile shares many of Rams’ less is/and more principles.
Secondly, we wanted to provide a way of guiding those developing courses towards the essence of design principles including, but not only, those of Rams.
So we developed the Guardrails, a set of learning design-specific guidance for teams developing HaileyburyX courses.
The Guardrails try, in the simplest and most economical way possible, to give course teams enough support to do their work without constraining them.
The Guardrails don’t specify what to do, but suggest one key step (“develop a learner story”), a set of four questions to consider and a set of five activities that learners could (but not necessarily) undertake in any learning experience.
We worked hard to enact Rams’ less is/and more principles in the Guardrails because, like the courses we are building, the Guardrails are also a designed object — and they are also part of the entire project. They are not documentation, a manual or an interesting bit of reading.
We wanted to innovate everything in HaileyburyX. And that applies to something as seemingly unimportant as a slide deck (principle 1, is innovative); we wanted to ensure the Guardrails were useful (principle 2), aesthetically appealing (principle 3), explanatory (principle 4) and unobtrusive and not over-designed (principle 5).
We also wanted to make sure that the Guardrails didn’t claim to do anything they didn’t — that they were honest (principle 6). The Guardrails say:
The Guardrails are not rules, a method or a system. They prompt the team to ask questions and consider issues.
The Guardrails don’t claim to be a process, or a manual, or a step-by-step recipe that when applied will churn out a course. And that’s because we want our courses to be unique and deliver unique learning experiences.
The aim is also that the Guardrails persist (principle 7) (and evolve as the project evolves), and they are designed with attention to detail (principle 8).
We did that by following principle 10 — as little design as possible: both visually (the Guardrails are 5 PowerPoint slides, mostly whitespace) and in terms of the text (the Guardrails are around 500 words). The Guardrails contain no diagrams, notes or references, complex examples or case studies.
It should be obvious why and how less is/and more is important in the Guardrails.
The Guardrails are an essential part of the HaleyburyX project (the gestalt, in Rams’ quote). They are as simple as possible because they are part of a project that is entirely motivated by less is/and more. Their aim is to provide enough of a framework to encourage creativity and innovation in course design.
The Guardrails, because they are not a process or method, also demonstrate less is/and more by shifting the work of course design into discussion and collaboration rather than the production of documents, plans and finished products. In line with our Agile Learning Design approach, teams produce minimum viable products — incomplete versions of courses that can be tested with real learners and iterated by the team.
There’s more to tell about this story but we hope to have shown how less is/and more is a way to encourage innovation, clarity, honesty, thoroughness — and especially good design through as little design as possible.
You can read more about our approach to learning design in HaileyburyX at
‘Good design’ and blended learning
Blended learning is a term that has no single definition.
Ask ten educators “what is blended learning?” and the chances are you will get ten different answers.
What is clear though, is that blended learning sits on a continuum between completely face-to-face learning and completely online learning. The degree of blend depends on the context.
(If you want to look at how we have approached looking at blended learning, you can download our draft ‘Blended Learning: a playbook for effective pedagogy’ — written with our Haileybury colleagues Carlie Gannon and Craig Nicholls — that we used to get the conversation flowing about blended learning at Haileybury).
It’s also important to recognise that ‘blended learning’ isn’t a new term at all.
From our point of view, teachers have always blended these things:
- instructional strategies
- methods of delivery
- forms of media
- learning activity development
And although blending classroom instruction isn’t new, blending with technology is less familiar — but the fundamental thing to bear in mind is that there is no one way to blend. It’s contextual.
As we said when talking about Agile Learning Design, the process of fitting blended learning to its context is an iterative process that involves a back and forth between the design and its consumers — in this case, learners. Assuming that we just do blended learning isn’t going to result in a good outcome.
So, following on from our discussion about design — that everything can, and should be designed — can less is/and more apply to blended learning?
We think so.
But first here’s a recap of Rams’ ten principles. Good design:
- Is innovative
- Makes a product useful
- Is aesthetic
- Makes a product understandable
- Is unobtrusive
- Is honest
- Is long-lasting
- Is thorough down to the last detail
- Is environmentally friendly
- Involves as little design as possible
Blended learning really is an exercise in less is/and more.
Less and/is more means not overloading learning experiences with more content, overlaying more methodologies or using more technologies just because they are there. We should apply the principles here just like everywhere else: we need to be conscious of our design choices, design for cognitive fit, think carefully about how we blend online and face-to-face components — and above all do as little design as possible (principle 10).
Principles 2 (useful), 4 (understandable) and principle 5 (unobtrusive) are incredibly important in designing blended learning because the focus is on the learners — providing them with a consistent experience, providing them with meaningful and intuitive learning opportunities and getting design out of the way to allow learners to spend more of their time doing the learning — not coping with the fact that their learning is blended.
And principle 8 (thorough down to the last detail) may be one of the key things to remember — and when combined with the kind of Agile approach we have described earlier, which focuses on iterative test-and-learn cycles — provides an opportunity to move away from the statement we are doing blended learning to the statement we are doing well-designed learning. As Julie Vargas, in her book Behavior Analysis for Effective Teaching, says:
“While technology has its place, we first and foremost want to make sure we’re prioritising effective pedagogy and not simply masking bad practice”
And the way to ensure that we don’t “mask bad practice” is through a less is/and more approach where we focus on real innovation, not just blending for the sake of blending. ‘Good design’ in blended learning isn’t applying a template or simply rule-following.
Of course, there are examples of good, and less good, design in blended learning.
Good design in blended learning means a well-thought-out rationale, teachers that are fluent in both face-to-face delivery and digital pedagogies and can move between the two seamlessly. And one of the ways you can see good design, or evidence of less is/and more at play is that well-designed blended learning provides opportunities for student voice, choice and action.
When blended learning doesn’t work it is usually because it isn’t blended learning at all. At best, it might be called ‘technology-rich instruction’ — which is not the same thing. It often happens when there is — out of the best intentions — adoption of technology without effective professional development and with no design orientation.
In other contexts, our shorthand for this has been more tech than ed — a recognition that some edtechs are more about the technology than the education. For Victorian teachers, a prime example might be the Ultranet system, which was really about efficiency rather than education.
Why is this important? Why is blended learning, and well-designed blended learning, essential? This piece from Forbes sums it up:
institutions that are unable to offer a blended methodology that seamlessly integrates face-to-face and online teaching will increasingly find themselves left behind, until they are simply out of the race.
Good design and assessments
By looking at the HaileyburyX Guardrails and at blended learning, we hope you can see how the less is/and more approach is a useful lens to see things and can deliver value.
But before wrapping up, we’ll touch on one area which, especially in the current disrupted circumstances, is in focus: assessment.
For us, assessment is all about design: all about being unobtrusive, all about honesty and all about detail. And it is an area where as little design as possible would seem to be worth pursuing.
While we (and you) could write entire articles, books, journal papers and reports about assessment, we wanted to connect it to the less is/and more approach to see how looking at it through that lens might be of value.
In the HaileyburyX approach, as expressed through the Guardrails, assessment should be authentic. Of course, free from the constraints of a curriculum, you might argue that we could pursue authenticity. But we think authenticity can be pursued whatever the context.
Here’s what the Guardrails say about assessment as part of the four questions:
WHAT IS OUR AUTHENTIC ASSESSMENT?
Assessments are designed to authentically assess the skills, knowledge or competencies that are part of the learner story.
While this looks like a statement too broad to be useful, it’s important to remember that it’s connected to the other questions (What is our learner story?, How will we organise content? and How will we combine channels?); to the Five Learning Tasks (Share, Acquire, Create, Experience and Reflect); to the Agile Learning Design approach which emphasises feedback and iteration; and to the entire HaileyburyX project. Assessment is integrated into the entire planning, design, development, delivery and redevelopment of a course. This is authentic assessment.
For example, we are developing a set of microcourses on Artificial Intelligence — the HaileyburyXAI courses.
Some of the courses look at topics like the ethical dimensions of AI, some of them look at basic general AI concepts, and some of them are learning experiences that involve students in doing some building of simple, and then more complex, AI using bots and other technologies.
How should we assess these courses? What does authentic assessment look like?
In our planning, we realised that we needed flexible ways to assess learners’ skills and competencies that authentically assessed their understanding of the perspectives, concepts and ideas around AI, and the skills and competencies in building AI — and ideally encourage learners to do more and learn more.
So in our microcourses you can head down the ‘non-technical’ Understanding AI pathway, looking at AI Ethics, Designing Conversations, General AI and AI Basics. Should you choose you can exit this pathway having never built anything.
But you also have the opportunity to take that knowledge and enter the set of microcourses that are about Building AI. This building is initially in the form of simple bots, but later by creating dialogues with a digital person — a complex AI and machine-learning driven interactive avatar.
If you choose to stay in the Understanding AI pathway, an authentic assessment may include diagnostic, formative or summative assessments: the assessment for the AI Ethics microcourse, which is taught as a dialogue between a digital person and a learner, may just be that — a dialogue; the assessment for General AI microcourse may be a recreation of a Turing test, or a reflection on it, for example.
In the Building AI pathway, learners will build bots and design conversations but must first learn about some of the ways that bot technology and software like Google Flow works. The assessment of that learning is through the successful building of the bots are also assessed through Learning Challenges — learners sharing what they have built through either predefined challenges (a bot that talks about music, for example) or ones that learners propose; and to complete these challenges learners have to create their own rubrics against which they will be assessed.
And to tie that all together, all learners use the gaming platform Discord (on which the bots are built) to share their work in a shared channel called the Exploratory, whether that’s working bots or dialogues with digital people.
Ideally, learners will choose to move from one pathway to the other, because they will be able to turn AI concepts into real working AI and share what they have done with their peers.
So is this authentic assessment? Does it demonstrate less is/and more? We think so.
That’s because it’s innovative — at least in this context; it makes the microcourses more useful because the assessment is tied directly to the learning; we have tried to make the assessment explain some of the content — to make it more understandable; it is unobtrusive — it is tied directly into the learning experience; it is honest — in the sense that it assesses real knowledge and skills; it has been thoroughly designed down to the last detail. And of course, it involves as little design as possible — both in terms of tech (we use the same platform that the learning takes place on, Discord, for assessment) — and in terms of the assessment design, some of which which is done by learners.
Why this is important is to show that authentic assessment — considered from the outset in the design as part of the learner story, the four questions and the five activities — enables us to use both technologies and the mode of assessment (or in this case several modes of assessment) that fit the context. No matter how much technology we use, that technology can’t make a badly designed assessment better — it just gives students a poorly designed assessment in a different way.
It’s also important to know what we want to assess. In the case of the HaileyburyX courses like HaileyburyXAI, the learner story tells us what the learning outcomes are and allows us to measure learning growth against those outcomes; and in each microcourse it allows us to separate low-level outcomes from high-level ones; and we have lots of choices of types of assessment — from speculative fiction, through a working bot, to a conversation with a digital human.
We also think that authentic assessment, designed using a less is/and more approach, also deals with one of the more thorny issues around assessment — dishonesty.
In HaileyburyXAI we are doing technology-based assessment but there is no real way to be dishonest; the assessment is set in unique pathways, building working things, and working things assessed against learner-created rubrics.
It makes no sense to cheat — or how you would do it or what benefit cheating would bring. We think that if we approach assessment from the perspective of the authenticity fostered by less is/and more the problem isn’t really a problem.
In any case, the problem may not be what it seems, since there is no reason to conclude that academic dishonesty will become more common just because the technology could facilitate it. As the authors of this paper (a limited study in the context of university students) conclude:
…academic dishonesty in a single online class is no more pervasive than in traditional classrooms. We attribute this finding to the way online courses are designed, which may reduce the need for cheating, and that panic cheating, a typical form of cheating found in traditional classes, is less likely to occur in online classes.
Wrap-up: less is/and more
We set out to talk about some ways in which less is/and more has shaped our thinking and about how it intersects with online learning, technology, and pedagogy.
We’ve done that by looking at Dieter Ram’s 10 principles for good design and how we have tried to apply them — especially principle 10, less is/and more or as little design as possible — to the HaileyburyX project, to the Guardrails that are part of that project, to blended learning and to the design of assessment.
As Dieter Rams said in his 1976 talk:
The complexity of systems and shortage of natural resources should compel a change of individual attitudes and attitudes as a society. We learn as individuals and we learn as a group. We are beginning to understand the changes that we are only just seeing. We must take notice with increasing soberness and, hopefully, with growing alertness and rationalism.
‘The complexity of systems’, ‘shortage of natural resources’, ‘a change of individual attitudes and attitudes as a society’, ‘learning as individuals and learning as a group’ could exactly describe what’s happening in education right now.
And Rams’ response— “take notice with increasing soberness and, hopefully, with growing alertness and rationalism” — is exactly what’s needed, whether that’s to take ‘alert, rational notice’ of online learning, blended learning, assessment, edtech, professional development or curriculum design.
And less is/and more, or as little design as possible, is important because, as Rams also says:
There are no discrete actions anymore. Everything interacts and is dependent on other things; we must think more thoroughly about what we are doing, how we are doing it and why we are doing it.
Thanks
Our thanks to Dr. Stephan Müller for help with the interpretation of the German word gestaltingenieur; to Toan Huynh and Jeff Plumb for their work in TechXAI and the Soul Machines crew for the digital people who are part of TechXAI; and to Haileybury’s digital learning leaders Carlie Gannon and Craig Nichols for their work on Blended Learning: a playbook for effective technology.
