At MIT Technology Review, Tom Simonite writes about Facebook’s efforts to make its automated assistant M answer pretty much any request that comes its way, not matter how obscure. And for a very small group of beta testers, the bot actually works, delivering results so good you’d swear you were talking to a human being. Because you are.
M is so smart because it cheats. It works like Siri
in that when you tap out a message to M, algorithms
try to figure out what you want. When they can’t, though,
M doesn’t fall back on searching the Web or saying
“I’m sorry, I don’t understand the question.” Instead,
a human being invisibly takes over, responding to your
request as if the algorithms were still at the helm.
(Facebook declined to say how many of those workers
it has, or to make M available to try.)
That design is too expensive to scale to the 1.2 billion
people who use Facebook Messenger, so Facebook offered
M to a few thousand users in 2015 as a kind of semi-public
R&D project. Entwining human workers and algorithms
was intended to reveal how people would react to an
omniscient virtual assistant, and to provide data that
would let the algorithms learn to take over the work
of their human “trainers.”
This is the way I’ve been prototyping chatbots, too: start with simple human-to-human interactions.
I’m a big fan of this kind of prototype that put people where the pipes will eventually go. In a way, Uber is a similar prototype for self-driving cars: until the robots get the go-ahead to drive on their own, we’ll put a human in the driver’s seat and automate the rest of the experience (calling a car, giving directions, paying the tab).
When you’re trying out new interactions for untested or emerging technologies, the best MVP is often no tech at all. Powering a bot with people instead of artificial intelligence gets you early info about what people want, how they respond, and the kind of language to use. It proves out the demand of the service, hints at the shape it should take, and offers training data to give to the bots down the road.
Eventually the AI steps in. At Facebook, they’re still trying to use all that data to train the bots well enough so that they can take over. Simonite shares some of the techniques the M team is using, with mixed results. Even though machine-learning breakthroughs are coming fast and furious, the holy grail of broad and instant natural-language understanding is still tantalizingly out of reach. “Sometimes we say this is three years, or five years,“ M’s leader Laurent Landowski told Simonite. ”But maybe it’s 10 years or more.”
Two years after it launched, a platform that aspired
to build a more stable path forward for journalism
appears to be declining in relevance. At the same time
that Instant Articles were being designed, Facebook
was beginning work on the projects that would ultimately
undermine it. Starting in 2015, the company’s algorithms
began favoring video
over other content types, diminishing
the reach of Instant Articles in the feed. The following
year, Facebook’s News Feed deprioritized article links
in favor of posts from friends and family. The
arrival this month of ephemeral stories
on top of the News
Feed further de-emphasized the links on which many
publishers have come to depend.
In discussions with Facebook executives, former employees,
publishers, and industry observers, a portrait emerges
of a product that never lived up to the expectations
of the social media giant, or media companies. After
scrambling to rebuild their workflows around Instant
Articles, large publishers were left with a system
that failed to grow audiences or revenues.
Building a business on top of someone else’s platform offers little control or visibility—and ties your fortunes to their priorities, not your own. Newton writes that many publishers are instead throwing in with Google’s AMP platform, which feels like a frying-pan-to-fire maneuver.
David Weinberger considers what it means that machines now construct their own models for understanding data, quite divorced from our own (more simplistic) models. “The nature of computer-based justification is not at all like human justification. It is alien,” Weinberger writes. "But ‘alien’ doesn’t mean ‘wrong.’ When it comes to understanding how things are, the machines may be closer to the truth than we humans ever could be.”
The complexity of this alien logic often makes it completely opaque to humans—even those who program it. If we can’t understand the basis of machine-delivered “truths,” Weinberger suggests, they become categorically different from what we’ve always considered to be “knowledge”:
Clearly our computers have surpassed us in their power to discriminate, find patterns, and draw conclusions. That’s one reason we use them. Rather than reducing phenomena to fit a relatively simple model, we can now let our computers make models as big as they need to. But this also seems to mean that what we know depends upon the output of machines the functioning of which we cannot follow, explain, or understand. … If knowing has always entailed being able to explain and justify our true beliefs — Plato’s notion, which has persisted for over two thousand years — what are we to make of a new type of knowledge, in which that task of justification is not just difficult or daunting but impossible? …
One reaction to this could be to back off from relying upon computer models that are unintelligible to us so that knowledge continues to work the way that it has since Plato. This would mean foreswearing some types of knowledge. We foreswear some types of knowledge already: The courts forbid some evidence because allowing it would give police an incentive for gathering it illegally. Likewise, most research institutions require proposed projects to go through an institutional review board to forestall otherwise worthy programs that might harm the wellbeing of their test subjects.
This is super-intriguing: what are the circumstances where the stakes are so high that we simply can’t allow ourselves to trust the conclusions of our machines, not matter how confident we may be in the algorithm? When it comes to “forbidden” areas of machine-learning models, Weinberger points out credit agencies are already forbidden from tying certain predictive models to credit scores. If the machines decide that certain races, religions or ethnicities are prone to lower or higher credit scores, for example, credit agencies are legally forbidden from acting on that info.
The reason this is a dangerous area is because the machines’ conclusions are only as valuable as the training data we feed to them. And that training data depends on the perspective (and bias) of the folks who collect it:
For example, a system that was trained to evaluate the risks posed by individuals up for bail let hardened white criminals out while keeping in jail African Americans with less of a criminal record. The system was learning from the biases of the humans whose decisions were part of the data. The system the CIA uses to identify targets for drone strikes initially suggested a well-known Al Jazeera journalist because the system was trained on a tiny set of known terrorists. Human oversight is obviously still required, especially when we’re talking about drone strikes instead of categorizing cucumbers.
We’re still in the early days of what this oversight and machine-human partnership might look like, but we’re going to have to learn fast. Machine learning has suddenly become inexpensive and accessible to a whole range of organizations and uses, and we see it everywhere. This revolution has revealed the complexity of everyday systems at the same time that it’s let us cut right through them through the capacity and speed of modern computing—even if we don’t understand how we got there.
Where once we saw simple laws operating on relatively predictable data, we are now becoming acutely aware of the overwhelming complexity of even the simplest of situations. Where once the regularity of the movement of the heavenly bodies was our paradigm, and life’s constant unpredictable events were anomalies — mere “accidents,” a fine Aristotelian concept that differentiates them from a thing’s “essential” properties — now the contingency of all that happens is becoming our paradigmatic example.
This is bringing us to locate knowledge outside of our heads. We can only know what we know because we are deeply in league with alien tools of our own devising. Our mental stuff is not enough.
Frank Chimero mulls the beauty of the plain and the normal in design. I like the implicit humility Frank suggests in designs that root their beauty in the quiet satisfaction of their function—not “an overly accentuated, hyper-specific identity”:
I am for a design that’s like vanilla ice cream: simple
and sweet, plain without being austere. It should be
a base for more indulgent experiences on the occasions
they are needed, like adding chocolate chips and cookie
dough. Yet these special occassions are rare. A good
vanilla ice cream is usually enough. I don’t wish to
be dogmatic—every approach has its place, but sometimes
plainness needs defending in a world starved for attention
and wildly focused on individuality. Here is a reminder:
the surest way forward is usually a plain approach
done with close attention to detail. You can refine
the normal into the sophisticated by pursuing clarity
and consistency. Attentiveness turns the normal artful.
Examples include automated web designs from The Grid CMS and Wix, as well as the machine-generated page layouts at Vox and Flipboard. There are also bot-built logos, type pairings, image generators, content-aware photo croppers, and more.
Lots to see and learn here about how designers will collaborate with our robot overlords.
for example, the âinnovationâ known as Gas Station
TVâthat is, the televisions embedded in gasoline pumps
that blast advertising and other pseudo-programming
at the captive pumper. There is no escape: as the CEO
of Gas Station TV puts it, âWe like to say youâre tied
to that screen with an 8-foot rubber hose for about
five minutes.â It is an invention that singlehandedly
may have created a new case for the electric car.
Attention theft happens anywhere you find your time
and attention taken without consent. The most egregious
examples are found where, like at the gas station,
we are captive audiences. In that genre are things
like the new, targeted advertising screens found in
hospital waiting rooms (broadcasting things like âThe
Newborn Channelâ for expecting parents); the airlines
that play full-volume advertising from a screen right
in front of your face; the advertising-screens in office
elevators; or that universally unloved invention known
as âTaxi TV.â
What to do about ad screens that are imposed on us in these captive scenarios? Wu suggests towns and cities have managed this problem before:
In the 1940s cities banned noisy advertising trucks bearing loudspeakers; the case against advertising screens and sound-trucks is basically the same. It is a small thing cities and towns can do to make our age of bombardment a bit more bearable.
At MIT Technology Review, Will Knight writes about the unknowable logic of our most sophisticated algorithms. We are creating machines that we don’t fully understand. Deep Patient is one example, a system that analyzes hundreds of thousands of medical records looking for patterns:
Deep Patient is a bit puzzling. It appears to anticipate the onset of psychiatric disorders like schizophrenia surprisingly well. But since schizophrenia is notoriously difficult for physicians to predict, [project leader Joel] Dudley wondered how this was possible. He still doesn’t know. The new tool offers no clue as to how it does this. If something like Deep Patient is actually going to help doctors, it will ideally give them the rationale for its prediction, to reassure them that it is accurate and to justify, say, a change in the drugs someone is being prescribed. “We can build these models,” Dudley says ruefully, “but we don’t know how they work.”
As deep learning begins to drive decisions in some of the most intimate and impactful aspects of life and culture—policing, medicine, banking, military defense, even how our cars drive—what do we need to know about how they think?
As the technology advances, we might soon cross some
threshold beyond which using AI requires a leap of
faith. Sure, we humans can’t always truly explain our
thought processes either—but we find ways to intuitively
trust and gauge people. Will that also be possible
with machines that think and make decisions differently
from the way a human would? We’ve never before built
machines that operate in ways their creators don’t
understand. How well can we expect to communicate—and
get along with—intelligent machines that could be unpredictable
This is especially important when the machines come up with bad answers. How do we understand where they went wrong? Or to know how to help them learn from the mistake? Knight offers a few examples of how researchers are experimenting with this, and many come down to new ways of visualizing and presenting the logic flow.
This resonates strongly with a key belief I have: the design of data-driven interfaces has to get just as much attention as the underlying data science itself—perhaps even more. If we’re going to build systems smart enough to know when they’re not smart enough, we need to be especially clever about how those systems signal the confidence of their answers and how they arrived at them. That’s the stuff of truly useful human-machine partnerships, and it’s a design problem I find myself working on more and more these days.
One hitch: we humans aren’t always so great at explaining our thinking or biases, either. What makes us think that we can train machines to do it any better?
Just as many aspects of human behavior are impossible to explain in detail, perhaps it won’t be possible for AI to explain everything it does. “Even if somebody can give you a reasonable-sounding explanation [for his or her actions], it probably is incomplete, and the same could very well be true for AI,” says Clune, of the University of Wyoming. “It might just be part of the nature of intelligence that only part of it is exposed to rational explanation. Some of it is just instinctual, or subconscious, or inscrutable.”
Brad suggests that development teams then build implementation-specific versions of the components that match the recommended rendered output. So you might have a React layer, an Angular layer, and so on. But those implementation details are all carefully segregated from the recommended markup.
The design system itself doesn’t care how you build it as long as the end result comes out the right way. Of course, developers do care how it’s built, and one promise of design systems is to deliver efficiencies there. So organizations should make it a goal for teams to share those platform-specific implementations, Brad writes:
This architecture provides a clear path for getting the tech-agnostic, canonical design system into real working software that uses specific technologies. Because it doesn’t bet the farm on any one technology, the system is able to adapt to inevitable changes to tools, technologies, and trends (hence the placeholder for the “new hotness”). Moreover, product teams that share a tech stack can share efforts in maintaining the tech-specific version of the design system.
Overall, our results showed that, while real-world
social networks were positively associated with overall
well-being, the use of Facebook was negatively associated
with overall well-being. These results were particularly
strong for mental health; most measures of Facebook
use in one year predicted a decrease in mental health
in a later year. We found consistently that both liking
others’ content and clicking links significantly predicted
a subsequent reduction in self-reported physical health,
mental health, and life satisfaction.
Our models included measures of real-world networks
and adjusted for baseline Facebook use. When we accounted
for a person’s level of initial well-being, initial
real-world networks, and initial level of Facebook
use, increased use of Facebook was still associated
with a likelihood of diminished future well-being.
This provides some evidence that the association between
Facebook use and compromised well-being is a dynamic
WPO Stats is a super-useful collection of stats from Tammy Everts and Tim Kadlec to demonstrate the business value of faster websites. If you need support for making the business case for your performance project, here’s your go-to library.
BBC has seen that they lose an additional 10% of users
for every additional second it takes for their site
to load. [source]
AliExpress reduced load time by 36% and saw a 10.5%
increase in orders and a 27% increase in conversion
for new customers. [source]
For every 100ms decrease in homepage load speed, Mobify’s
customer base saw a 1.11% lift in session based conversion,
amounting to an average annual revenue increase of
I missed this a few weeks back. At Search Engine Land, Danny Sullivan reported that Google is empowering its 10,000 human reviewers to start flagging offensive content, an effort to get a handle on hate speech in search results. The gambit: with a little human help from these “quality raters,” the algorithm can learn to identify what I call hostile information zones.
The results that quality raters flag is used as âtraining
dataâ for Googleâs human coders who write search algorithms,
as well as for its machine learning systems. Basically,
content of this nature is used to help Google figure
out how to automatically identify upsetting or offensive
content in general.â¦
Google told Search Engine Land that has already been testing these new guidelines with a subset of its quality raters and used that data as part of a ranking change back in December. That was aimed at reducing offensive content that was appearing for searches such as âdid the Holocaust happen.â
The results for that particular search have certainly improved. In part, the ranking change helped. In part, all the new content that appeared in response to outrage over those search results had an impact.
âWe will see how some of this works out. Iâll be honest. Weâre learning as we go,â [Google engineer Paul Haahr] said.
Federated Learning allows for smarter models, lower
latency, and less power consumption, all while ensuring
privacy. And this approach has another immediate benefit:
in addition to providing an update to the shared model,
the improved model on your phone can also be used immediately,
powering experiences personalized by the way you use
We’re currently testing Federated Learning in Gboard
on Android, the Google Keyboard. When Gboard shows
a suggested query, your phone locally stores information
about the current context and whether you clicked the
suggestion. Federated Learning processes that history
on-device to suggest improvements to the next iteration
of Gboard’s query suggestion model.
Old way: beam everything you do on your Google keyboard (!!) back to the mothership. New way: keep it all local, and beam back only an encrypted summary of relevant learnings. “Your device downloads the current model, improves it by learning from data on your phone, and then summarizes the changes as a small focused update.” To do this, Google has smartphones running a minature version of TensorFlow, the open-source software library for machine learning .
One knock against predictive interfaces is how much you have to give up about yourself to get the benefits. If this new model works as promised, new systems may be just as helpful, without the central service absorbing your nitty-gritty details to learn how.
But there are a few problems with pattern libraries.
Yes, they allow you to keep all of the smallest elements
consistent. But they don’t have an opinion about how
they should be put together. They don’t know anything
about your product or the concepts behind it.
To return to our Lego analogy, simply having a limited
pattern library of bricks to choose from doesn’t preclude
me from building some really crazy shit.
Now think about those branded Lego kits you can buy.
Each piece is much more opinionated. It knows what
it’s going to get used for. There are still generic
pieces involved, but when you put them together in
a certain way they form something specific, like the
leg of an AT-AT Walker. This is a design system.
I love it. Design systems are more than a kit of parts. The best design systems have a strong point of view—a gravitational force that coerces disparate components into patterns and ultimately into a coherent whole. The design system brings order to the pattern library and what would otherwise appear to be a chaotic jumble of components.
Another metaphor: if components are words, then patterns are sentences, and the design system is the full story.
If this nested arrangement echoes Brad Frost’s Atomic Design methodology, that’s by design. Atomic Design champions design elements built from a common set of lesser design elements. In Atomic Design, UI “atoms” assemble “molecules” which assemble “organisms” which assemble templates which assemble pages.
But there’s a common misunderstanding about Atomic Design which Connolly in turn suggests is a limitation:
Atomic Design will tell you to
take some of your basic elements (label, input, button),
stick them together, and call it a molecule. Then you
can reuse that molecule again and again. Further, you
can stick some molecules together to form a reusable
The problem with every real-world example of a system
like this that I’ve encountered is that they remain
willfully unaware of the product being built.
Atomic Design does indeed promote reuse, assembling larger parts from smaller ones. However, many mistake this philosophy for linear process, that somehow Atomic Design demands that all design must first start by building its smallest pieces (e.g. “start with buttons and labels”) before proceeding to page- and site-level design. It’s an approach that would indeed be blind to the end-result project, placing design tactics ahead of design strategy. But that’s exactly opposite to how Brad himself approaches projects.
It’s never a linear path from small to large; it’s a constant roundtrip between the two scales.
Right from the start, when Brad was first developing his tools and methodologies in our designs of TechCrunch and Entertainment Weekly, our process constantly zoomed back and forth between page level and atomic level. It’s never a linear path from small to large; it’s a constant roundtrip between the two scales.
As Connolly writes, “Complex systems can be designed, but to do so you must first sketch the outline. Only then can you start filling in the detail.”
Well said, and I totally agree. Indeed, our Atomic Design projects always begin with the big-picture questions. What are the business goals for the project? What are the user needs? What’s the brand promise? When we get to individual pages, it’s about the user mindset when they arrive, and the jobs the page has to do for both user and company.
From there, we do sketching of the whole page, identifying the broad design patterns that the page needs to do its job. We start to imagine the components necessary to bring those patterns to life.
Only then do we start to work at the atomic level, building out those component atoms and molecules to construct the pattern organisms, and ultimately the page itself. As more high-level pages and components are designed, we zoom back down to revisit the atoms and molecules, making adjustments to make them more flexible and support a wider range of organisms and pages. The atoms and molecules might compose the design, but it’s the high-level design that creates the order, the overall system.
In the end, a pattern library emerges. Here’s the important bit: the design system is implicit in the process that led to the library’s construction, and it’s implicit in the design’s use of components. For a small team on a contained project, that implicit knowledge may well be enough, commonly shared in the heads of the designers who built it.
But implicit knowledge won’t do when you’re working at scale across many projects and many teams. The design system has to be documented. That’s where all the other artifacts of a fully articulated design system come in: design principles, style guide, voice and tone, UX guidelines, code repository, and so on.
I agree very much with Connolly that those pieces are required for the “full-stack design system.” My only caveat is to add that an Atomic Design process can get you there, too.
Atomic Design surfaces all of those aspects during the course of the design process. Responsible designers document them.
Is your organization wrestling with inconsistent interfaces and duplicative design work? Big Medium helps big companies scale great design through design systems. Get in touch for a workshop, executive session, or design engagement.
At The New York Times, Penelope Green reports that sleep is big business—and the tech industry is rushing in to tweak our natural rhythms, with mixed results:
Mr. Mercier sent me his Dreem headset, a weighty crown
of rubber and wire that he warned would be a tad uncomfortable.
The finished product, about $400, he said, will be
much lighter and slimmer. But it wasn’t the heft of
the thing that had me pulling it off each night. It
skeeved me out that it was reading — and interfering
with — my brain waves, a process I would rather not
I was just as wary of the Re-Timer goggles, $299, which
make for a goofy/spooky selfie in a darkened room.
My eye sockets glowed a deep fluorescent green, and
terrified the cat.
The science and research confirm that there’s an epidemic of sleeplessness, which is costly in both health and productivity. Are tech gadgets the answer when tech gadgets are likely a big part of the problem? Our screens keep us awake; always-on information demands contribute to anxiety and stress; and social FOMO is constant.
As technologists, we often suggest that more technology is the solution to technology’s problems. In the case of sleep, perhaps a little less technology is what’s needed. Green quotes “sleep ambassador” Nancy Rothstein:
and your Apple Watch are not going to do it for you.
We’ve lost the simplicity of sleep. All this writing,
all these websites, all this stuff. I’m thinking, Just
sleep. I want to say: ‘Shh. Make it dark, quiet and
cool. Take a bath.’”
“Annotations Editor launched in 2008, before the world
went mobile,” writes YouTube product manager Muli Salem
in a blog post.
“With 60 percent of YouTube’s watchtime
now on mobile, why go through the work of creating
annotations that won’t even reach the majority of your
If it doesn’t work on mobile, it doesn’t work, period.
Self-driving tractors are becoming more common. A John
Deere spokesperson told me the company currently has
about 200,000 self-driving tractors on farms around
the world, from the US to Germany. And they’re just
one example of a major investment that the agriculture
sector is making in artificial intelligence and the
Internet of Things.
According to John Deere, between 60 and 70 percent of the crop acreage in North America today is farmed using GPS-driven tractors. (Source)
Farmers have been inhabiting the future for a long time—self-driving tractors have been on the go for 15 years. Industrial farms are big business; they feature wide-open spaces; and they operate on private property. All make of this makes farms ideal test beds for tech that includes autonomous vehicles, drones, artificial intelligence, and smart objects.
Liz Stinson, in Wired, previews Lightform, a “projection-mapping” device that can read a room and project images (or interfaces) onto any surface, no matter how irregular. In a nutshell, it’s augmented/mixed reality projected directly onto the environment:
Lightformâs technology sets the stage for more complex
and immersive forms of interaction. The company aims
to develop high-resolution augmented reality projections
that track objects and respond to human input in real
time. Its ultimate goal: Make projected light so functional
and ubiquitous that it replaces screens as we know
them in daily life life. âReally what weâre doing is
bringing computing out into the real world where we
live,â Sodhi says.
What I like about emerging technologies like this one is that the tech comes to you. Your surroundings simply become digital; no need to strap on a headset or peer through a screen.
Virtual, augmented and mixed reality products like
HoloLens and Daydream are often seen as being in the
vanguard of this evolution, but the level of immersion
required by these experiences is a somewhat misleading
guide to the future.
The larger concept at play here is the notion that
digital capabilities â through projection, augmentation
or other more subtle forms of ingress â will become
woven into the physical fabric of life. The dream of
ubiquitous computing will not come in boxes, but rather
will hover and shimmer in transient spaces around us.
“Woven into the physical fabric of life.” This is the exciting opportunity about the physical interface, whether embodied in IoT gadgets, projected UI, or augmented reality: it literally grafts onto the world around us, on our terms. It’s tech that promises to bend to our lives, rather than the reverse.