Don’t listen to Le Corbusier—or Jakob Nielsen
by amy
Cheerful software, above all, honors the truth about humanity:
Humans are not rational beings.
A human is a walking sack of squishy meat and liquids, awash in chemicals.
We laugh. We cry. Sometimes we laugh while crying. We love, and hate, and dream about tomorrow while paying no attention to today. We do ridiculous things in pursuit of love or happiness or self-esteem. We sabotage ourselves. We see faces in inanimate objects, clouds, rock formations, and unevenly toasted bread. Then we sell them on eBay.
We pray to giant humans up in the sky. We think that a fly could be our grandmother. We work for free because we’re bored. We create art, dance, and sing even if we are starving. We give to others when we have little, or we give none when we have a lot, even if we gain no clear survival benefit either way.
Rationality’s not all it’s cracked up to be
We can aspire to cold, pure logic—but our lives fall apart if we get there.
People who have had damage to the emotional centers of the brain can’t even decide which pair of shoes to wear. If you—with your intact brain—think harder about a choice, the chances go up that you will make one you regret.
This is what psychology and neuroscience tell us: the beauty and the rot is all mixed up. You can’t have a human without both.
But that doesn’t apply to designers. Does it?
Despite all this, it’s so appealing to think that the work we do can be reduced from infinite uncertainty to finite, scientific certainty.
We think that eye-tracking studies, or any other research, can go beyond telling us what they found and tell us what to do.
There’s nothing wrong with the scientific study of usability. Until you pretend that it’s prescriptive, instead of descriptive. That something good for humans can be reduced to mathematical certainty.
That there is a science for designing.
Then you get engineerism.
Engineerism is nothing new
This happened before. When this notion occurred architecture, bad things happened:
Home life today is being paralysed by the deplorable notion that we must have furniture. This notion should be rooted out and replaced by that of equipment.
– Le Corbusier, in response to Mme. Savoye’s request to fit an armchair or two into her famous but unlivable Corbusier-designed home
Le Corbusier thought sofas and armchairs were a terrible thing — the softer and cuddlier, the worse.
“What [modern man] wants is a monk’s cell, well lit and heated, with a corner from which he can look at the stars,” he wrote.
He designed what he called “machines for living,” buildings that served all of the functions that a human putatively required.
They were a beautifully austere, ‘scientific’ vision.
They were hated, and abandoned.
People don’t want to live in scientific certainties
Sofas are what people want.
And “machines for living” don’t honor the messiness that a human life represents.
When they have a choice, a person’s home is more than a place that meets their scientifically ascertained needs. It’s a reflection of who they are, and how they want to be. They fill it with things that support their future aspirations, and remind them of the past. Whether they love lofty open rooms and lots of glass, or exposed wooden beams and tiny windows.
Architecture is not just a way to house bodies and fulfill function, but to feel.
Software is no different.
Engineerism is taking over interaction design
The voices of interaction design pretend to be scientists. They take eye tracking studies, and scrolling studies—and never mention how well-done the content in question might be, or whether it was exciting and relevant to the test subject.
“Users don’t scroll.”
“Put content on the left side.”
They argue that the ultimate test of a software design is whether it is usable. Just like Le Corbusier and his army of engineers, they focus on the pretend science of utility.
This is the same as a machine for living in or a well-lit monk’s cell.
We don’t make software our homes, but we spend 7 – 8 hours a day there.
Software cries out for personality, for ornamentation, for delight. To reflect higher aspirations, and evoke emotion.
For designers who don’t think it’s silly to imagine software as a helping hand, rather than a tool to be used.
In short, software cries out for love.
Thanks to Alain de Botton’s fabulous book The Architecture of Happiness, for helping me cement a way to explain the things I felt about design and providing the Le Corbusier quotes. You should read Architecture. Start with this excerpt, which dovetails perfectly with my essay.
Comments
I very much like this article. I have to say, I HAVE spent countless hours on radically un-usable sites and have gone back to some of those so obviously, there’s something else to it.
Brava. Great points enjoyably conveyed.
Along these lines, you might also enjoy Tom Wolfe’s clever screed “From Bauhaus to Our House.” I sure did. It rails against many of the most dehumanizing Modernist architectural excesses and the twisted philosophies that drove them and that still infect the architectural establishment.
One typo: “occurred architecture” should be “occurred in architecture.”
Great thoughts here. I love the notion of ‘software as a helping hand’.
It is our responsibility as designers to demonstrate that art is an infinite incubator of possibilities. There is definitely a difference between useful and enjoyable.
Have you ever read Le Corbusier’s grand manifesto, “The Modulor”? It was a comprehensive scheme of architecture scaled to the measurements of the human body. The idea was that every object in an architectural environment should be built to proportions that were ideal for human bodies to occupy. And of course Corbu’s homes were notorious, everyone would constantly knock their knees and elbows on walls and corners, the homes were almost impossible to live in.
But Corbu, despite the quote you gave, did design furniture. Have you ever sat in a Corbu chair or sofa? You immediately know the proportions are all wrong for a human body, the arms are either too high or too low, the depth and height are all wrong. But it is an idealized form of a sofa and chair, beautiful lines and proportions. Sitting on Corbu furniture is an experience, the fit is just wrong enough, it immediately says that your body doesn’t really belong there.
Although I appreciate the exuberance and passion of your manifesto, I have to tell you that I am always moved to laughter when I read what a cold, heartless bastard Corbu was. He is the favorite modern demon of Romantics everywhere. What is often missed, usually because the context in which Le Corbusier worked is misunderstood – is that he was essentially a Humanist (and a Romantic in his own right!) – and was very much someone who appreciated all of what makes us human. He has been misread by people with their own agendas (ie, Tom Wolfe) so often that he is now the hobby horse for anyone who wishes to question the premises and intents of the Modern Movement. If you’d take the time to study him more deeply, you might come to very different conclusions than you have from reading the likes of De Botton – and if you think Alain De Botton’s got something important or original to say about about architectural design, you are sadly mistaken.
And to Charles, who thinks Corbu’s furniture is poorly sized, I agree – but please remember that the average human in the 1930′s was much smaller than we are today! I am the proud owner of a Corbu Chaise Longue, and it fits my smaller than normal body just fine, thank you. There is no place better to read a good book.
I applaud your desire to take on the likes of Jakob Nielsen and those who want to apply fast and hard rules to interactive experience and the design of software (or anything). I just think using Le Corbusier as an example of what is wrong with design is misplaced. If you knew his work as I do, you might think differently about the man and his work.
Best of luck! I look forward to more of your work.
I believe it should say “prescriptive”, not “proscriptive”.
I once took an Aesthetics class back in college and we studied architecture for about half the class. At the time I thought I would be an architect, but I ended up as a user experience designer.
It’s nice to hear this perspective and I think it’s been a bit of a wake up call to me to return to my thoughts on architecture.
I may have to pick up Maya Lin or Gaston Bachelard again
I think you’re getting it backwards. Le Corbusier’s problem was not that his design process was too scientific; his problem was that his design process was too subjective. Le Corbusier did exactly what you seem to suggest designers should do: he put trust into his feelings and his own opinions instead of objective data. Hence, his designs were unusable. Even a small user test would have revealed the problems with his design.
Just in case this is not obvious, a “well-lit monk’s cell” is not a usable space, and would never be the result of an objective design process.
A sofa, on the other hand, is a usable space, and would score highly in user testing, even though you seem to imply that it would not.
I honestly don’t understand the conflation between a design process that tests designs with actual users and puts human needs and requirements before personal opinions, and a result that is somehow devoid of humanity. The two seem to be opposites.
Lukas, actually, that was exactly my point. Both Le Corbusier and Nielsen masquerade as scientists — as using cold, hard, rationalist logic — but neither meet the criteria.
It’s the personal idealism masquerading as science that’s the problem.
I added stabby single quotes to make my viewpoint regarding the ‘science’ more clear.
Eye tracking is not a science. Usability testing is not a science. It pretends to be science. It can tell you things — yes. But it cannot tell you why.
I repeat: Nielsen’s faux science never tells you about how the test was performed, the goodness, usefulness, or relevance of the content to the test subjects.
That’s not science. It’s pseudo-science.
His partner Don Norman’s work is much more scientific, but he’s a quiet partner and doesn’t make loud, detrimental announcements about how the web should be.
Finally – I never said the designer should rely solely on HER feelings as to what was right, and nothing else. I said software should EVOKE feelings, and RESPECT the power & importance of feelings in people’s lives.
There is a huge difference between the two statements, don’t you think?
Laurence, I know he designed furniture; I’m a design nut, I couldn’t help but know it. And yes, they are ill-proportioned. I’m not tall but I find the cube chair to be awkwardly low.
I also know and agree that he was a Romantic… but he was Romantic in the sense that he wanted to reshape humanity with his vision, whether they wanted to be reshaped or not.
That’s the sign of somebody questionable at best, and dangerous at the worst.
Don’t make the mistake of thinking that my exposure to Le Corbusier is limited to Alain de Botton’s work. I’m quite familiar with the Modernists and Bauhaus movements in general (and have never read that Tom Wolfe book).
I know — and I disagree.
I agree, engineering is dull. Pseudo science is always embarrassing. Science is about questions, engineering is about details. Design science is really closer to engineering. Personally I think it’s related to computer programmers involved in the process that like to pretend they’re doing something hard. They influenced the squishy brains of designers who are right to be squishy. Now maths will always be harder, more creative, and much more beautiful
I think there are talented UX designers and graphic designers who should feel free to ignore Nielsen etc. when their instinct tells them to.
But: In the kingdom of the blind, the one-eyed man is king. Much of the web-design world is still the kingdom of the blind, sad to say, and in my experience the inhabitants would be well advised to follow one-eyed Jakob Nielsen instead of their own random ideas.
Amy,
Okay, I read the text again, I think I misunderstood. If I understand correctly, what you’re saying is:
1) Don’t blindly follow Nielsen’s “rules”, but
2) Do run usability tests with your own designs.
I totally agree with that.
Lukas,
That’s more like it. Plus:
3) Emotion, subjective enjoyment, are more important than ‘usability’
Freckle, for example, bucks the ‘usability’ gospel in many places. But our customers write us love letters.
I think there is an over emphasis on the science and function aspect of usability. which I think you article is underling and that we need that science and the art of software usablity maybe you should take a look at this video about what makes design intuitive http://interactiondesign.sva.edu/blog/entry/video_jared_spool_what_makes_design_intuitive/ if you think about it usability reinforces old ways of interaction and does not create new ways of interacting with software.
Sounds a bit defensive to me… There clearly are benefits to interaction design, and the human mind actually does function rationally in most scenarios. Trying to start a battle here won’t do any good for anyone, users included.
Amy,
You say: “Emotion, subjective enjoyment, are more important than ‘usability’”
In my experience, there is pretty much no difference between the two. People are happy when they feel in control. They feel in control when something is usable, when they can correctly predict what will happen when they do something, and when they can figure out how to reach their goal quickly and easily.
Only a usable application can be enjoyable. Of course, you can make a usable application more enjoyable by using tools outside of the realm of usability, but you can never make a non-usable application enjoyable. People will never enjoy your application if they are unable to print their document, and they will never enjoy your website if they can’t find the information they’re looking for.
Simply not true, Lukas!
Please read “Emotional Design” by Don Norman. I take great issue with the person he later went into business with (Nielsen), but Don’s book puts the lie to what you’re saying.
It’s also been out for years so there’s no excuse for being a professional designer, usability engineer, interaction expert, blah de blah whatever, without reading it.
He cites several studies that show that things that elicit an emotional reaction, or are even just plain pretty, see greater use and are even perceived to “work better”.
More importantly, you can look around you everywhere and find that people choose things that elicit emotional reactions (e.g. Ferraris, iPhones, the infamous Alessi juicer, metal-handled knives, fancy designer shoes/chairs) while having serious functional limitations, or are just flat-out difficult.
Ever sat in the seat of a Ferrari? They’re actually damnably uncomfortable.
And yet…
I always worry that as designers we follow ‘the rules’ as laid down by Neilsen and the like, without any further thought. We remove everything that might get in the way of our users end goal, but by doing this we remove the life from our products. People, real people crave interesting interactions, thought provoking experiences. As a UX community we should stop listening to the scientists and start thinking for ourselves. Freeing ourselves to create products that our customers will not only use, but will love. This is the only way forward, this is what we’re here for. Great article, thanks.
Begin rant:
I’m getting bored of this rather shallow, strawman meme. Its been rehashed so many times in the UX field that I’m surprised we are back here again. That whole “evil, cold, scientists versus warm, fuzzy, humanist designer” thing. I mean, haven’t we done this one already?
As I read it, you’re claiming that designers should ignore boring, cold, rational, science-y “evidence” about whether or not people can actually use the products they design, in favour of whimsical, cuddly and cute designs that will “charm” (some) users. Basically, you imply we should ignore anyone who suggests that our awesome creativity might not actually serve the needs of the people who, y’know…. actually pay our wages? Nice.
You’re preaching form over function. That’s a design non-starter in my book. Ever seen anyone actually use Phillippe Starck’s rocket “objet” to make orange juice? Me neither. Looks charming, though. But doesn’t a beautiful, useless object fall into the category of “art” not design?
Truth is, if we lived in a world where all software was already designed in a way that was 100% usable and useful then maybe I might have some patience with your argument that what we need is more curlicues and digital cuddles. But since I’m typing this rant on the iPhone’s appalling keyboard, well, I’d trade a better input method for any number of warm fuzzy Jobs flourishes, I assure you. (In fact my irate tone would likely be a lot less irate if the tool I’m forced to use wasn’t so bad at doing what I paid for it to do - letting me type what i want to say without endless corrections and backspaces. How human does that make the iPhone’s design?)
Perhaps what you need is a better understanding of what user-centred design actually is. The simplification of ”usability” is a mistake. Nobody really just does “usability” any more – any more than anyone really uses eyetracking or scrolling stats to make decisions. If that’s as far as your knowledge of UX today goes, I’m not surprised you’re disappointed.
How about a different phrase? I like to talk about ”evidence-based design”. In other words, gathering evidence from users in a reasonably controlled, unbiased way, so as to help designers make better design decisions – rather than basing those decisions on opinion and ego. Would you disagree that listening and learning from users is a Good Thing? So why not include it in your design toolbox?
And the reason why it’s a good idea to employ specialists to help you listen to users (even people like Jakob Nielsen) is that they are unbiased. They will not allow you to screen out the evidence you don’t want to hear. (The notorious “ugly baby” problem). I find it interesting and very revealing that you choose to use the love letters you get about Freckle as evidence that there are no usability problems with your product. Hmmm. That’s known as ”selection bias” - a recognised phenomenon in the field of social science – that same cold, rational ”pseudoscience” which you seem to hold in such contempt).
So: by all means let’s use our creativity to create cheerful, warm, fun experiences. But let’s start by building products that actually work the way that users need them to. And let’s do this by getting out there and understanding real people in the real world, not sitting godlike in front of our Macbooks, bemoaning those silly users and clients who ‘just don’t get it.’
Rant over.
http://iancollingwood.com
Sorry, Ian, but you’re flat-out wrong.
In one sentence:
I never said anyone should ignore usability heuristics and go with their hearts!
I never said people should not observe people using their software in the wild.
I said software has to EVOKE emotion, not that software designers should RELY on emotion.
I said that mechanical “usability” should not not the height of a software designer’s aspirations.
I said that mechanical usability science is, in fact, not a science — which is true. It’s a load of observations pretending to be a science. Treating like a science is a disservice to everynoe.
Can you please stop blaming me for other people’s arguments, and actually read mine?
Thanks!