Often, it starts with a bit of paper. Samsung's design gurus like to cut out the shape of their latest dreamed-up device and wear it for a while. If they like it, they'll move onto a 3D printed version. If they continue to believe in it, they take it to Samsung headquarters - where they'll probably be told it's unworkable, at least for now. That's usually a sign they're onto something good.
We're inside Samsung Design America, San Francisco, where drones fly indoors, shelves of design books pave the walls, and photos are strictly forbidden. But you won't find blueprints for the Galaxy S8 or the next Samsung 4K TV here. This is a testing ground for Samsung's more outlandish, less realistic ideas. Folding screens, modular phones, smartwatch displays that wrap around your wrist; and rumors of Samsung's more "out there" concepts can probably be traced back to this very building.
Read this: Jacquard, Pilot and more - up close with the Designs of the Year
"If we don't hear the word 'impossible' when talking to engineer teams, then we're probably not working on the right thing," says Sameer Bhalla, one of Samsung's design directors, in a roundtable discussion.
Bhalla is responsible for most of Samsung's wearable devices to date, as is Sean Bornheimer, two people whose job it is to be forever frustrated by the pace of technological progress. Take the Gear Circle, Samsung's Bluetooth workout necklace launched in 2014. What the design team had in fact proposed at the time, down to the very look you see on them today, was the IconX. "But once we got to headquarters where the rubber meets the road in terms of commercializing it, they said it was impossible," says Bornheimer.
Drones fly indoors, shelves of design books pave the walls, and photos are strictly forbidden
So the IconX turned into, and went to market as, the Gear Circle, and as the design team suspected, users complained about the shape and the fit. Two years later, the technology was ready and - at last - Samsung's first true hearable was born. "It's a very rocky road," says Bornheimer, "and you have to go through a lot of potholes and a lot of trials and tribulations until you can get to your original vision."
But the Gear Fit wasn't really ready, and Bornheimer and Bhalla aren't afraid to admit they got a few things wrong. "When this first launched, the software didn't work, the content wasn't ready for it. The design got some nice plaudits but the product itself wasn't ready. And it took us two or three generations to turn this into a mature, commercial product."Some of the more interesting challenges have happened on the wrist. The first Gear Fit was a wearable born from the team's desire to simply have a screen that curved around the wrist. "The display didn't exist, none of the technology existed - a curved battery didn't exist - but it was just an idea," says Bornheimer. "What got this idea to move forward was us building the designs and people putting them on their wrists and seeing how beautiful it looked compared to traditional smartwatches at that time which were really, really bulky and techy looking." This was, after all, the era of the Galaxy Gear and Sony Smartwatch.
Just being able to present information in portrait, not only landscape, was something they say they learned from the Gear Fit and applied it to the Gear Fit2. And, credit where it's due, despite not really taking off, the curved OLED display was cutting edge for a wearable in 2014. "At the time it was great," says Bhalla. "Now it looks a little dated".
The untethered wearable conundrum
Samsung once patented a watch that projected its UI onto the hand
The problems Samsung has wrestled with in wearables will be familiar to most big tech companies, but cellular connectivity, which Samsung jumped on early, opened the door to a host of especially interesting challenges and temptations. "We wanted to learn: what would people do with a wearable that has LTE in it?" says Bornheimer. "Customers were starting to ask for LTE in wearables, but they couldn't really explain why."
For example, would you watch TV on a smartwatch? Probably not, but the idea was certainly tossed around in the early stages of the the company's first 3G smartwatch. "One hypothesis came up," says Bhalla, "Why don't we put Netflix on this product? And we had to sit down and say, hey, is this the place where you're going to watch House of Cards? Is this the right interface?
One hypothesis came up... why don't we put Netflix on it?
"So we struggled with... what is the right use case? And the leading ones that seemed appealing and super, super cool, because Netflix at this time was brand new, this was 2015, Netflix was just starting in the digital game, so it sounded cool, of course... But of course it doesn't make sense. You're not going to watch movies on your wrist."
A lot of the challenges were in software and logistics of antenna placement, but as the first kid on the block there were other unforeseen headaches too. "At the time there was no infrastructure in place to enable one number on multiple devices, so you have one device on your wrist.
"It seems really obvious now, but honestly three, three and a half years ago, people were like, oh maybe I want a different number here because this is my business wearable, maybe I want separate numbers for these things, and when you start using them it's really clear - people want the same number."
Samsung therefore had to go work with carriers like AT&T in help enable the infrastructure to support one number across multiple devices. "That was really a big change," says Bornheim. "And I don't think people would adopt any of these LTE wearables at any great mass without having that software in place."
And the final result of all this was the Samsung Gear S, a wearable that, yes, had 3G, but also a design defined by all these constrains which somehow managed to be at once both beautiful and hideous. "It's quite large - that's not where the design started," says Bornheimer. "That's not the product we're most proud of from a design standpoint, but it really pushed the bar forward in terms of what a wearable could do and where wearables would go in the future.
"It was kind of big and kind of weird, but we learned a lot from that product and that helped push the company in the right direction to putting LTE in wearables." And the Gear S lives on, sort of, in Samsung's Sim Band reference wearable.
A modular future
The tension of technology and fashion is another neverending headache for Samsung's wearable designers, and both Bhalla and Bornheimer admit that making profiles better suited for women has been an area Samsung has not been quite so good at - but promise it's working to fix that. It's also thinking of other ways - beyond making tech smaller - that wearables can be more fashionable. "The other way is remove stuff out of it," says Bhalla. "If it's a wearable, maybe it doesn't need so much of a screen. Maybe it doesn't need a screen at all."
We'll probably have a reasonably compelling AR headset five years from now
But one thing at least Bornheimer doesn't have an interest in is hybrids, despite the company showing off a version of the S3 with a Swiss-made movement at Baselworld earlier in the year. "That's true fashion, and you've gotta be kind of honest to your brand. We could design the most beautiful thing in the world, but at the end of the day it's a Samsung mechanical watch. It's just not sincere to consumers."
For Bhalla, modularity is a particularly interesting concept they're toying with. "We have quite a few modular concepts both in the wearable space and for non-traditional phones," he says. But this too has its own bag of commercial and technological obstacles.
"We had an idea for a modularity between two products that would put together and start doing other stuff. We had an issue in our original concept that any time different pieces of technology connect, there's now a point of infraction for waterproofing, so you can't waterproof each individual device if you want them to touch."
Perhaps at the very least, the camera which featured on the original Samsung Galaxy Gear could even make a return one day. Samsung's not saying for certain, but Pranav Mistry, Samsung's global senior VP of research, told us he thinks it will eventually make a comeback once smartwatches can better stand alone (and also told us that back in the days of the Galaxy Gear, Samsung bent the strap camera literally one million times to test its durability).
And of course, there's health and fitness, but neither Bornheimer or Bhalla are convinced anything Samsung nor anyone else has is near what's yet compelling enough to make this a must-have, some of which is down to good old fashioned bureaucracy. "If you've ever picked up an FDA or medical grade personal healthcare product, they're using technology from five years ago," says Bornheimer. "They're big, they're bulky, they're using some kind of faded out screen, and thats because the product development cycle is so long to get something like that to market."
The team believe the new FDA fast-track path, which Samsung will be part of the pilot for, could go some way to fixing that bottleneck. "You need to take something that's a five- or seven-year product cycle and bring it into a one-year product cycle that will allow you to deliver those technological changes a lot faster. These sorts of bureaucratic or certification related changes are going to be a big part of making that a reality."
So what else beyond smartwatches, trackers and hearables? AR is naturally on Samsung's radar just as much as everyone else's, and there's a good chance that, locked safely away in a room or a computer hard drive, maybe just a few meters away in fact, the first pieces are starting to come together.
"We'll probably have a reasonably compelling AR headset five years from now, that's probably a reasonable timeline," says Bornheim. "What's it going to look like? What functions are going to be on it? What are people really going to use it for? How small is it gonna be? Nobody knows, really.
"We need to build and iterate many, many times - and probably make some mistakes".
How we test