Connected products, systems and the little chip with a big brain
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 170 | |
| Author | ||
| License | CC Attribution - NonCommercial - ShareAlike 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
| Identifiers | 10.5446/50603 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
NDC Oslo 201420 / 170
2
3
5
6
8
11
12
16
21
22
23
27
31
35
37
42
43
45
47
48
49
50
52
54
56
57
58
61
65
66
67
71
74
77
80
81
83
84
85
87
88
89
90
91
94
95
96
97
98
100
102
107
108
112
114
115
116
118
120
121
122
123
126
127
128
130
133
135
137
138
139
140
141
142
143
144
145
147
148
149
150
153
155
156
157
158
159
160
161
162
163
166
169
170
00:00
SoftwareInternet der DingeHypermediaBusiness modelMultiplication signBitHybrid computerQuicksortSelf-organizationNeuroinformatikWordSoftware developerCodeModel theoryComputer animation
01:02
Product (business)SoftwareMultiplication signIntegrated development environmentQuicksortCategory of beingHypermediaSoftwareOffice suiteDirection (geometry)Business modelHybrid computerLevel (video gaming)Video gamePhysicalismComputer scienceProduktraumComputer animation
02:50
ExistenceProduct (business)QuicksortSpacetimeInternetworkingBitWordData managementNeuroinformatikConfidence intervalLevel (video gaming)Public key certificateBasis <Mathematik>PhysicalismTrailSpectrum (functional analysis)View (database)Ubiquitous computingGraph (mathematics)DistanceConnected spacePurchasingTouchscreenCoefficient of determination
04:56
QuicksortHypermediaProduktraumPersonal area networkContent (media)Universe (mathematics)Information technology consultingSocial classSquare numberDistanceForcing (mathematics)Digital photographyData conversionComputer iconPRINCE2Game theoryStructural loadPhysical systemWebsiteLetterpress printingPay televisionPortable communications deviceSpacetimeInformationBitBookmark (World Wide Web)Optical disc drive
06:24
HypermediaOptical disc driveFlow separationImage resolution2 (number)Latent heatLetterpress printingInheritance (object-oriented programming)Topological vector spaceMobile appMultiplication signMoment (mathematics)InformationImmersion (album)FacebookQuicksortPoint (geometry)Forcing (mathematics)Frame problemUniverse (mathematics)BitCurveWärmestrahlungPRINCE2Revision controlMereologyEntire functionComputer animation
07:52
Internet der DingeInclusion mapQuicksortBitFile formatWeb browserMessage passingWeb 2.0Speech synthesisMereologyInternetworkingElectronic visual displayWebsiteReading (process)Service (economics)Physical systemSpacetimeMultiplication signFactory (trading post)ProduktraumMedical imagingWärmestrahlungCASE <Informatik>Gastropod shellStudent's t-testProcess (computing)Hacker (term)LiquidCuboidSoftwareOrder (biology)Server (computing)Sign (mathematics)Structural loadBookmark (World Wide Web)Image resolutionWordTerm (mathematics)Degree (graph theory)Component-based software engineeringPressureComputer hardwareTouchscreenMiddlewareOptical disc driveData storage deviceData miningVirtual machineBit rateLine (geometry)NumberConsistencyGraphics tabletCommunications protocolPRINCE2Robotics40 (number)DivisorProduct (business)QuantificationSound effectModel theoryFamilyTime travelGraph coloringSpherical capFerry CorstenPrototypeMaterialization (paranormal)Matrix (mathematics)Operator (mathematics)
14:49
Operator (mathematics)TelecommunicationProcess (computing)BitFamilyFactory (trading post)Component-based software engineeringRoboticsVirtual machineLiquidData storage deviceRow (database)Normal (geometry)MereologyQuicksort
15:49
Ultimatum gameSystem programmingDirected setGame controllerBitSoftware testingProduct (business)Process (computing)TelecommunicationPlastikkarteSystem callShooting methodThermal radiationTerm (mathematics)ProduktraumInteractive televisionSurfaceService (economics)Materialization (paranormal)NeuroinformatikMixed realityPhysical systemOrder (biology)QuicksortVirtual machineSpacetimeCache (computing)Information privacyToken ringComponent-based software engineeringExistential quantificationMachine visionCAN busWindowDigital photographyReal numberRepresentation (politics)Graphical user interfaceMenu (computing)Uniqueness quantificationDifferent (Kate Ryan album)SoftwareFormal languageExtreme programmingLengthData miningFacebookIntegrated development environmentTwitterLevel (video gaming)Multiplication signModel theoryApproximationCategory of being12 (number)Server (computing)POKEAbstractionWeb browserFunctional (mathematics)Natural numberDialectField (computer science)AnalogyComputer virusoutputUser interfaceFigurate numberBEEPSoftware as a serviceRow (database)State of matterForcing (mathematics)Sound effectMathematicsReverse engineeringPrice indexMetropolitan area networkOpen sourceGoogolCoefficient of variationPoint (geometry)Alpha (investment)MereologyLogical constantStreaming mediaComputer animation
24:13
Directed setSoftwareBookmark (World Wide Web)BitSpacetimeLetterpress printingCategory of beingInternetworkingExpected valueHypermediaCuboidTerm (mathematics)QuicksortNeuroinformatikRow (database)Order (biology)PlastikkarteMathematicsMachine visionView (database)Representation (politics)Thread (computing)PhysicalismType theoryPhysical systemLevel (video gaming)Multiplication signOperator (mathematics)FlickrInformationSoftwareAbsolute valueProcess (computing)Degree (graph theory)Interactive televisionArray data structureExtension (kinesiology)Radio-frequency identificationLibrary (computing)WordComputer hardwareComponent-based software engineeringNetwork topologyPattern recognitionJoystickUser interfaceComputer scienceGraphics tabletVermaschtes NetzMassCloud computingException handlingEngineering physicsKeyboard shortcutInformation securityMoment (mathematics)Surjective functionCycle (graph theory)Computing platformSpeech synthesisArmBiostatisticsMereologyAuthenticationComputer iconNumberOptical disc driveSlide ruleMathematical analysisGraph (mathematics)WindowOffice suiteOnline helpVideo gameWeb 2.0System callArithmetic meanVector potentialAlgorithmGroup actionGoogolPoint (geometry)ProduktraumPerspective (visual)Line (geometry)
32:01
Optical disc driveMobile appExpected valueOrder (biology)Model theoryInformation securityPrime idealBitProcess (computing)QuicksortService (economics)Level (video gaming)Type theoryNoise (electronics)CASE <Informatik>Graphics tabletMoment (mathematics)FamilySingle sign-onTunisShared memoryContent (media)VideoconferencingObject (grammar)Computer animation
34:05
System programmingProduktraumSoftwareSoftwareSpacetimeOptical disc driveShared memoryRoboticsProduktraumPoint (geometry)Multiplication signProduct (business)Business modelPhysical systemComputer animationJSON
35:04
System callPoint (geometry)Complete metric spaceNumberBitShift operatorOperator (mathematics)Multiplication signVisualization (computer graphics)Electronic mailing listMathematicsProgramming paradigmSlide ruleQuicksortEngineering physicsElectric generatorMassCuboidProcess (computing)Computer animationLecture/Conference
36:42
Semiconductor memoryOperator (mathematics)Data storage deviceExtension (kinesiology)Local ringInstance (computer science)Object (grammar)LogicTerm (mathematics)Program slicingDistribution (mathematics)SoftwareBitMusical ensembleService (economics)Virtual machineLibrary (computing)NeuroinformatikLevel (video gaming)Portable communications deviceFactory (trading post)File formatProduct (business)Compact CassetteHypermediaProduktraumComa BerenicesGraphics tabletState of matterSystem callIntegrated development environmentEquivalence relationMoving averageConnected spaceSeries (mathematics)Web pageWorkstation <Musikinstrument>
40:26
Personal digital assistantProduktraumFood energyControl flowInclusion mapPoint (geometry)TouchscreenKey (cryptography)Uniqueness quantificationVirtual machineElectronic visual displayLiquidMobile appTerm (mathematics)Electronic mailing listInternetworkingPlastikkarteArithmetic meanNeuroinformatikWater vaporMultiplication signQuicksortOrder (biology)Product (business)Functional (mathematics)ProduktraumInjektivitätSpacetimeView (database)Greatest elementIntegrated development environmentMoment (mathematics)SoftwareConnected spaceVideoconferencingData storage deviceHard disk driveMechanism designBlogRemote administrationVideo gameBoolean algebraHydraulic motorPhysical systemModel theoryPoint cloudSystem callWavePrototypeWritingCommercial Orbital Transportation ServicesComputer animation
45:36
DialectVirtual machineTelecommunicationPhysical systemEquivalence relationTouchscreenSingle-precision floating-point formatComputer animation
46:48
Virtual machineWhiteboardService (economics)TelecommunicationWater vaporGoodness of fitVirtual machineSpecial unitary groupMassCircleDialectRight angleUniverse (mathematics)Square numberProduct (business)Greatest elementProduktraumTablet computerInterface (computing)Shape (magazine)Cycle (graph theory)Hydraulic motorGame controllerEntire functionGodScaling (geometry)AreaComputer programTerm (mathematics)BitMicrocontrollerDenial-of-service attackComputer animation
49:16
Mobile appWeightData transmissionService (economics)Virtual machineComponent-based software engineeringMarkup languageGoodness of fitExecution unitRight anglePhysical systemPasswordComputer animation
50:55
Parity (mathematics)Component-based software engineeringBus (computing)Right angleUniqueness quantificationPoint (geometry)PurchasingComputer animation
52:03
Computer-generated imageryVideoconferencingExpressionProduct (business)Hill differential equationProjective planeDigital photographyVideoconferencingBlock (periodic table)PrototypeFocus (optics)ResultantMereologyWorkstation <Musikinstrument>Point (geometry)NeuroinformatikPhysical systemMachine visionBitQuicksortMusical ensembleMathematicsGoogolRight angleMathematical singularityPlanningHoaxArmTelecommunicationCuboidSurfaceProcess (computing)Software developerGame theoryHill differential equationInternetworkingWeb pageProduktraumGreatest elementWordOptical character recognitionComputer hardwareWhiteboardSoftwareCycle (graph theory)Shooting methodProduct (business)Object (grammar)BlogCloud computingMoore's lawOrder (biology)InfinityClient (computing)Core dumpPoint cloudSound effectBelegleserTerm (mathematics)Video projectorVirtual machineForm (programming)Moment (mathematics)RoboticsOpen sourceCoprocessor2 (number)Revision controlReal numberPort scannerServer (computing)Task (computing)Reading (process)FreewareSpring (hydrology)Set-top boxAnalogyE (mathematical constant)Computer animation
59:07
InformationSelf-organizationCuboidBitComputer animation
Transcript: English(auto-generated)
00:02
Can you hear me? That's working, that's definitely not my voice I can hear. Hello, thank you for coming, I've never been to a developers conference before, so I was a little bit scared, so you're going to have to forgive me,
00:23
the last time I wrote a piece of computer software or code, as some people talk about it, was about four and a half years ago, and even then it was pretty straightforward, so I did struggle a little bit to try and convince the organisers that what they really needed was my CTO, who could talk to you about actual software,
00:44
but they insisted that I was what they wanted, so I'm going to just talk a little bit about my company and what we do, and some more general themes around this word that's come about, called the Internet of Things. So my company is called Berg, we are based in North London,
01:03
we used to be a design agency, we founded in about 2005, we had a kind of hybrid model, so we did work with technology and media companies, as well as making our own software and products, and when you're an agency, our business model was basically, every sort of three months, you have to look in the bank
01:20
to make sure there's enough money to pay everyone, and then you go out and you speak to some very rich companies to try and make them marginally richer in exchange for money, which you charge for the time of the human beings you have in your company, and that's one business model. And then we decided that we were so clever, we were going to make our own product, so we took venture capital,
01:41
and for those of you who don't know what venture capital is, there's probably no one, but just bear with me, because it was fairly new to me, basically there are a bunch of extraordinarily rich guys, and they know even richer guys. The guys up here are the kind of people who manage the property portfolio for the Vatican, or the whole of the New York Police Department pension fund, or whatever,
02:03
and they look at their sort of $4 trillion, and they think, I tell you what, we'll put most of this into safe investments, and we'll just take a few hundred billion and spend it on high-risk nerd stuff and hope it turns into Facebook, and we're one of those. So the downside of this is you can't have an industrial chic studio in North London anymore,
02:22
you have to go somewhere really shitty where the rent is cheap, sort of like nylon carpets and polystyrene ceiling tiles. And so now we live here. And really the best way I can characterize this office is this, everything from the aesthetic to the kind of heart and soul that goes into keeping the building alive.
02:41
So yeah, it's spiritually enriching as an environment. But in many ways, I'm enjoying our new direction. So one of the things we've always found quite difficult to do is to define what we do, because we're mostly a design company, but in the room with me, as someone who was trained as a designer,
03:00
also is a computer scientist, and a good one at that, and people who talk about ground bounce and all the weird things that happen at the physics level in chips and radio interference in certain spectrums and certification of kind of small pieces of plastic with copper tracks running through them. So there are also some sort of scary minds sitting quite nearby.
03:23
But to talk about connected products, I thought I'd just show this graph because there's a lot of words, and historically, certainly in academia, people have been talking about a space where consumers would purchase internet-connected products for some decades now. And you end up with words like ubiquitous computing. That was big sometime in the 90s.
03:42
And now we've got this thing, Internet of Things, and they've really got far more syllables than I think they deserve. But in my opinion, you've really got three things going on. One is you've got electricity, which is weirdly the hardest one to come by, in my view. It's really rare, electricity. It's hard to make. It's hard to keep.
04:01
And it's kind of why I think people quite like to put the internet in a fridge because you know you've got a fairly confident power supply. And in my experience of working with large manufacturers, it's many of the reasons why your phone doesn't do as much as it could. Computationally, I think your phone could do a lot more than it does, but power management is one of the biggest problems with it.
04:21
So electricity is a bit of a dog, really. It's the one that people don't think about very much. Then you have some kind of connectivity by wires or radio. There are some other kind of fancy-pants methods, but really, you kind of want some bits and bytes going between one thing and another over distance, and either that's going to be wires or radio. And then you need some humans, some sort of bags of meat with their capacitive fingers
04:42
and their wallets spending things and poking at capacitive screens and stuff. And somewhere in the middle of that, there's connected products. And that's broadly where I find the actual practical work that I do on a day-to-day basis happening. So we made a product. One of the things we did after six or seven years of consulting with manufacturers
05:03
and technology companies and media companies, which is where this kind of world, for us at least, seems to have converged, we decided that we had to kind of try and actually make one ourselves. If we wanted to consult and tell people, yeah, you should be making a connected device or use this kind of system or think about that or manufacturers like this and those sorts of things,
05:21
that we should actually have a stake in the game ourselves. So we scraped up a load of money and we made a product called Little Printer. And it is a domestic product. It's a thermal printer that sits in your home. And you use a website to subscribe to subscription content, so news headlines, Foursquare check-ins, social media. Some things are really practical like travel data.
05:40
Some things are really social like Instagram photos. Some things are just weird like people do comics and kind of odd stuff like that. And you can also use it like a tiny kind of portable printer for your phone. So you just take photos and it sort of prints out like a kind of magical fax, like a kind of long distance Polaroid. Anyway, it also prints its own face. And that's my favorite bit about it.
06:01
But it's one of the things that's allowed it to kind of catch on as a kind of design icon in this space. And it's been very warmly received. Anyway, we manufactured them. And there are thousands of them floating around in the world with people pressing the button on the top and printing out little bits of information and puzzles and things for their kids and all that kind of stuff.
06:21
And, you know, it exists. And it was, well, one of the things it does that's quite nice is at conferences like this, it prints out kind of itineraries. And so it's very timely. It's kind of funny because in the media universe, you seem to have these two quite convergent, quite separate sort of forces at play at the moment.
06:42
One is you've got this obsession with like resolution, immersion, super high kind of very, very high definition, high resolution kind of information in media. Things like The Hobbit being shot at like 96 frames a second, like native in the camera. Or Oculus Rift and this kind of super immersion just like drowning in pixels,
07:01
like being punched in the eyeballs with Photoshop. And you've just got sort of IMAX kind of Samsung with their kind of big curved 4K TVs. And on the other hand, you've got like four guys in a basement in New York making easily the worst photography app in history and then selling it to Facebook for a billion dollars.
07:21
And it's just a bit odd because there's this second media universe, which is where 140 characters is worth more than an entire newspaper. And it's because it's timely and personal and specific that it arrives from someone at that point is worth more than the greatest journalists of our age writing about war. So it's a kind of odd time in media that you have these two universes.
07:43
Anyway, Little Printer speaks very much to that kind of light, timely, low resolution universe where the quality of the print isn't what matters. It's that it arrives at that moment. And of course there's a business version, so now it turns out that actually loads of people who have to deal with thermal printers as part of their businesses were banging on the door saying,
08:01
can we have one without the stupid face please? And it's basically because you don't need any local infrastructure for it. It's got a web API like in the thing. So you turn it on, Wi-Fi, internet. There's no like, you know that thing when you get companies like Cisco and things like that or NCR, all these giant middleware solutions and someone's got to have like a server room just to run like a dot matrix display.
08:24
And here you just go to your website, sign in, tell the printer what you want to do, just point your APIs at it, and it prints it out. So it's basically like a tiny, very, very low resolution browser with a very slow refresh rate. That's kind of one way to think about it.
08:40
So, what is an IoT? So just to kind of get a sense, and this may be, again, speaking to people, I sometimes do this talk to like design students where I'm like, it's an API, write that down. So forgive me if some of this feels like it's preaching to people who know more about it than I do. But sometimes I think it's worth just describing a few things that kind of give you a smell of what the space is like.
09:02
This is a product called GlowCaps, and I quite like it. It's produced by a guy called David Rose, who's ex-MIT, and it's a very successful company. It's been sold to private equity since it launched, and it's a medicine bottle top. And it's probably, I think, the bill of materials or the cost of the system to any given medical consumer,
09:21
as in any patient, because it's issued by doctors in the U.S. It's given out with the medicine as part of the prescription. I think the cost of the system is something like $200. But the cost of a psychotic person not taking the drugs that keeps them sane is so high, legally and punitively, and in terms of insurance,
09:41
that it's worth giving out everyone an entire system to record whether or not they're taking their medicine. So when the bottle top gets opened and closed, it sends a message to the system to say that you've taken your medicine. If you don't take your medicine, it phones the police or your mother or gives you a badge on Foursquare or whatever the system is that it's been tuned to. The second one is Fuel, which has since been discontinued by Nike,
10:03
but no doubt their close relationship with Apple will mean that in some respect it continues. But it's a kind of quantized self-product that measures your activity, reflects it back to you so you can tell how unfit you are. And the last one, another favorite of mine, is the Kindle, which is another weird product because e-ink is like the lowest resolution
10:22
profoundly expensive technology. I mean, if you go to factories in China, there are three factories in China that produce e-ink displays, and the e-ink display on the Kindle is easily 90% of the bill of materials of that product. It costs way more than the 3G component. It's literally the most expensive part.
10:41
And the Kindle is sold broadly at cost, a little bit of profit maybe, but it's really, compared to an LCD display, the only advantage to it is that it's not backlit, so you can read it in reflective light, and it doesn't consume any power when you're not changing it, which are both very, very useful things. But it's incredibly expensive, so it's an odd product to choose, unless you're Jeff Bezos, in which case you've just convinced
11:03
millions of people to buy a shop. Like, when we look at the Kindle, it's like, hey, neat. Like, book reading stuff works in the light, and you only charge it once a month. He looks at it like he's just sold you a buy now button. And it's quite extraordinary. Anyway, so there are some examples of things in that space. Some things which are a bit less discussed,
11:21
which is a bit odd in IoT, because IoT is a bit of a funny word for things that haven't ever quite succeeded. You know, it's a sort of weird ghetto of the park of, like, student experiments and Arduino hacks. But there are actually quite large grown-up service businesses that use stuff which is hardware in the world. And if you look outside of that kind of odd consumer realm
11:41
of kind of big, clunky Samsung gear and all that kind of stuff, then you get to some interesting things. I mean, GoPro is kind of a wearable. It doesn't have a persistent connection to the Internet, but the images that you make on it, you use somewhere else. And it's a profoundly successful company. The second thing is there's a business in the UK called Just Eat, which is like, it's really very successful.
12:02
And what it does is it aggregates restaurant menus from around, from takeaways and delivery companies like pizza stores and Indian takeaways and Chinese takeaways. It presents them to you on a website with a consistent format. You search by your postcode and say, oh, want Chinese food? Yeah, have that one, that one, that one. You press go. It faxes the restaurant kitchen, and the chef gets that fax and presses okay.
12:23
You get a message back on your browser that says the chef has received the order, and then that receipt turns up at your home at delivery. And so it's just a kind of, it's like the Uber for takeaways, if you know what I mean. But it needs this printer in the kitchen because there's not going to be an iPad screwed to the wall in the restaurant kitchen. They're just not going to do that.
12:41
So weirdly, it's actually an essential part of a component of a business which is not really about things at all. It's just better if there is one. So this is Shenzhen, and it's a region in China where loads of things get made. I'm sure you've heard of it. Foxconn live here. So do most of the flextronics manufacturers. So it's basically where all of the kind of junk
13:02
with screens that we click on and poke at gets made. And it's very big and very weird, and it's odd how little I think one understands as a kind of Western consumer of what goes on in the production and fabrication of the things that we regard as the actual stuff of consumer electronics. And because we made a piece of consumer electronics,
13:21
we had to go there. And manufacture, for me, is easily one of the most miserable processes available. Like software's really hard, but manufacture is so miserable. Basically, sort of anecdotally, I think you could say with manufacturers, you spend a lot of money and time designing something as you'd like it.
13:42
So you think, okay, this is my product. And you've prototyped it. You've had the manufacturing engineers in. You've done all the smart thinking, and you put it there. And then, with manufacturing process starts, the industrialization thing goes to the factory. You commit to spending about 250,000 pounds for 10,000 of something exactly like this
14:03
but a little bit more shit. And then what you have is 10,000 of these. You still don't have any money. Then you've got to sell it. Really, you can see why you'd rather make angry birds and just kind of go, oh, upload. And then sort of everything else seems to take care of itself.
14:21
Anyhow, this is an injection molding machine making the back of a little printer. It's a 40-ton machine, which basically means you get a bunch of bits of colored plastic and you put it in a hopper. You heat them up to some hundreds of degrees. And then you put 40 tons of pressure behind it because plastic only gets to the consistency of cold chewing gum when it's really hot. It's not like liquid. You're not really injecting liquid.
14:41
You're forcing cold chewing gum into a metal box in order to make a tiny plastic shell. And there it is. And two brothers run that factory. One does the tooling. One does the injecting. It's very strange to China in that there are lots of operations in manufacture which are like little family businesses, like corner stores or something. It's not like there aren't just these huge
15:01
glittering factories everywhere filled with robots. You have to make this stuff. Obviously, this is like the electronics bit, circuit boards. And this is where you get all the poisons that screw up all the rivers. It comes from this part. And you have to work a lot of copper to sit around in a row, fill it through other bits of copper. This is a room in Eastern Europe where we do all the electronics assembly
15:21
where the billions of tiny robots pick and place and populate all of your electronics. And then you get these brilliant machines where you've sort of semi-glued on all your electronic components. And then you just shoot this torrent of hot solder at your circuit board. And it kind of glues on to the feet of the electronics to bond permanently the components that you have there.
15:41
And that's literally a liquid jet of solder just firing up. And you get these fantastic processes that are completely opaque in normal circumstances. You also, if you live in the West, governments are particularly tetchy about producing really bad electronics with radio in it. So if you make a microwave that's a bit leaky, it doesn't go down very well for anyone or your hair falls out,
16:01
or if you make a radio that's like not great, then like aeroplanes start falling out of the sky because you ruined air traffic control. That's not good. So there's a very, very aggressive testing process where you have to take products to little rooms like this with these funny walls and Faraday cages and like shoot it with stun guns to make sure that it restarts properly or it doesn't release some kind of weird radiation, even if it's like just naughty Bluetooth.
16:22
So that's not a lot of fun either, and it's expensive. So the nice thing about those problems in manufacture is that although they're miserable and expensive and nothing brilliant can really happen, they're at least linear. Like when you've done them, they're done. You just turn the handle now. We want to make more little printers.
16:40
You make a phone call, credit card. They turn up in the post. Like it's just done. Whereas software always feels a little bit like you've never finished. You know, it's a bit, there's just something a little bit like, it feels a bit like manufacturer's bacteria that you can kind of cure. Software's like a virus that never really goes away.
17:01
It's like cold sores or something. It sort of comes back every now and then. So for me, it's a much more serious problem in spite of the fact it's more familiar to me and the people I work with. This is Bill Verplanck. Bill Verplanck's got one of those brilliant CVs like 12th employee at Apple. His first job when he left industrial design college in the 1950s was to go and humanize Martian environments
17:22
because back when NASA thought it was the future of technology. Do you remember when people cared about space instead of just pictures of their food? And like NASA thought they were going to like use all their money to colonize Mars and the government would just back them indefinitely. They were worried that the thing about astronauts is they're all basically soldiers. That's what an astronaut is. You basically get the smartest guy in the Air Force
17:41
and stick him in a space suit and that's how you get astronauts. But you can't populate Mars with just like smart soldiers. You have to send teachers and nurses and kids. And the problem with Mars is it's a giant red desert with a 17 hour day. So they're worried that like everyone would just, they would send everyone there and then they'll have this kind of horrible psychotic episodes and kill each other in some kind of horrible Columbine massacre.
18:01
And so they hired Bill to like make it nice, to make sure Mars was nice enough. Anyway, he invented the term interaction design with Bill Moggridge, IDO, sometime in the 80s. So he's quite an important figure. And he did this drawing. And this is a drawing of an interaction. So there is our human, bag of meat, capacitive finger, et cetera, doing things to that ball.
18:21
And that ball is your product, the abstract thing, the system that one is designing. And that thing has got like handles and buttons. And back then there really were only handles and buttons, sliders, dials. You had some fairly straightforward analog inputs. This was before computer vision and capacitive surfaces. But still, there's only so many things you can do to a computer to tell it
18:40
or do to a product to tell it something. And then that product manifests stuff. It gets cold or hot or it beeps or the engine turns on or it goes left or whatever it does. And that manifests back to you and you feel that with your eyes or your ears or your hand or whatever. And then in your brain you have a kind of map, a sort of map of the system. And if you think about driving a car, driving a manual car, you kind of,
19:00
you do sort of learn it like that. You have an understanding of things and you kind of know even though you've never done it that if you're doing 60 on the motorway and you slip it into reverse, nothing great is going to happen. So, you know, it sort of works as a sort of idea of how that happens. But for me, phones have completely ruined this paradigm in that this no longer has any real value left.
19:24
And that is because if you think about your phone when you're using it, you're really not really using your phone. You're kind of using my phone a little bit in the way that systems work. And as I'm on my phone, I don't really get a sense of the edges of the device anymore.
19:41
When I'm talking to Siri, I'm not completely convinced that I'm talking just to this phone, not to a server farm that Wolfram Alpha owns parked in some corner of a desert in Texas. You know, or when I'm asking Google search, it's not like the browser isn't answering how old Barack Obama is or where the moon is or who was in the dirty dozen.
20:01
It's like that's going somewhere. So the edges of the machine don't feel like, don't quite feel the same. And think about something like Twitter. It's really integrated into the phone now. Notifications appear on Facebook and photos. That's on the phone. I mean, that's as much of my understanding of that as a product and understanding of it as the kind of going into the menu to change the language to Finnish or whatever, you know.
20:24
Plus it affects computers and the edges of that system aren't clear. And some people get upset by it. So if you swear on Twitter, it appears somewhere. Privacy is weird. You never quite know where something's going to get seen. What happens if someone tags a picture with you in it on Facebook? Does that appear on someone else's phone? It's this constant maelstrom as the services
20:40
and the UI and the design of these things actually start to get into a kind of weird, brilliant mix. But it certainly isn't like driving a car. Where you put the buttons and the order in the menu isn't quite the same in terms of relevance. So that doesn't work. Plus, you've got brands in on this now. You've got companies and systems and not even humans on the end of these lines.
21:02
I got this from Oslo back in February 2009 where it said, you know, it's SMS. You can see how old the phone is. And then it's like this is normally where I get told off by my mother for not phoning her on a Sunday. You know, and here we are. It says, Welcome to London city to Oslo to check in for some SaaS flight.
21:21
You know, just say yes. So I said yes. It says, you are checked in. It's completely bizarre to me. What have I just done? It's just so, this is not UI design. There's something else going on here. So, products are part of systems,
21:40
so I don't want to run over time. Representing systems in hard. And for me, if you are working in the field of interaction design or design for software, any kind of design where human beings have to walk up and poke things or things happen to them, it doesn't matter what you're doing, essentially at some deep level, the nature of software in its abstraction is that it is invisible because it's at its deepest level.
22:02
It's maths, as you all know far better than I do. You know, one holds models of how these systems work and they quite often go to some extreme lengths like graphical user interfaces or windows or trash cans or whatever to make some sense of metaphors and representations to humans so that they can operate on the functions that you've provided meaningfully for them.
22:20
So, making representations of invisible systems is kind of what design in software is. So, I'm sort of channeling an old colleague of mine, Darryl Bishop here, who's a very seasoned interaction designer and worked for a long time in Apple. And we always talk about this thing about cache.
22:40
And there's unfortunately no one in the front row, so I just have to imagine that there's someone sitting there. But this is a pound coin. It's worth 10 of your Norwegian kroners, approximately. And it's mine. We all know that because I just got it out of my pocket. But if I put it here and there was someone standing there,
23:00
it suddenly feels a little bit less like mine and a lot more like theirs. You know, I'm just a little bit nervous just standing here and it's over there. I might forget it. It's not really mine, but it's definitely my money. Now, if that person picked up that coin and put it in their pocket, there's real grounds for dispute. And if they give me back a different pound coin,
23:21
have they stolen my pound? Now, if I left my shoe there and they picked it up, they'd just be holding my shoe. You know, that's all. So, there's something unique about cache in that cache is essentially a component of a system. It's a physical token. There's all sorts of things it tells you. It tells you it's British because it's got the queen and some unicorns
23:40
and all that imperialist stuff on it. And it tells you what it's worth. It's a pound. It's not 10 pence. It's not 50 pence. It's a pound. So, it has some indicators. Even its physical materiality tells you it's a pound because it's quantified and qualified by the state. But it doesn't tell you who owns it on its surface. So, it's a social system in as much as it is a technical one
24:00
or a formal one. And there are others, but I don't want to bang on. But all it is to say is that there are systems all around us that aren't really to do with software, and they exist. And car insurance, that's another really weird one. I sort of need it to drive a phone up some company of some that are supposed to be in a market, and I give them a little bit of money that's based on something to do with me
24:22
and some analysis of risk on the promise that they might give it back to me against the value of the damage to my car should something happen if they believe that is what happened. And you need that, but it isn't represented anywhere, but it is on my car. I mean, my car is covered. So, there's just this odd thing of belief and systems and sociality
24:41
that actually sometimes software doesn't formalize, but that we still live with. So, this is a very nice quote from Darl. To use something, we have to be able to perceive it. So, I don't mean see it necessarily. I'm not saying everything has to be graphical or everything has to be procedurally laid out or represented literally. But this is just a very nice example of some user interfaces.
25:00
For one, it's on the side of one of those giant kind of articulated cherry pickers, we call them. They're sort of platforms on long kind of pneumatic arms that mean people can go and chop the tops of trees off or clean high up windows. And you can see there next to the buggy that everything just has a little joystick at each joint in the articulation of the arms so that you kind of drive the car by just sort of moving around. There is a fabulous example
25:20
of interaction design as far as I'm concerned. This is so weird. And you guys are gonna know more about this than me, and it's not gonna be weird to you, but you just have to suspend yourself and pretend you haven't done computer science degrees. You don't know what pipes are or arrays are or scary maths or algorithms, right? Just pretend you're me and you went to art school twice and that's how bright you are, right?
25:42
So you go out and you buy a Macintosh like this one, just like this, or you go and buy a Windows PC and you hear in the office this rumor and someone's doing something and they're typing into this weird black window and stuff's happening somewhere else. And you know, what are you doing? And they say, they whisper the mythical magical word, Python.
26:00
And you're like, Python, what, what? And you now know this magical word. That's the only way to find out about it. There's nothing in the computer. You know if you want to launch Photoshop, you kind of look for this icon or text edit. There's like a little picture of a pad with a pencil on it. I mean, it's not brilliant, but at least it's there. And you click on it and a thing opens and you can kind of hit the keyboard and the letters appear and you kind of get it, right? You're using the software.
26:21
Same thing with Photoshop. You want to do a little drawing, you just scrub your mouse around and you're kind of, all right, this is Photoshop, I get it. But if you want to use Python, which is easily as powerful with the libraries and the extensibility as Photoshop, is in that computer, I mean I bought it, you have to type the word Python in a little black window, probably in the right way as well.
26:40
I don't know what else you have to do, but I think you just type Python and then press return. And then it says, type help, copyright, credits, or license for information. And I'm like, what is this? And it turns out it's an absolutely giant, amazing piece of software with absolutely no formal representation to me. It doesn't make any effort to perceive itself, to make itself perceivable to me whatsoever,
27:01
which for people who have spent a long time learning how to use command lines and type esoteric characters in a row in order to make very powerful software work, it probably seems fairly natural. But from my perspective, it's utterly bizarre that so much effort would go into something that then goes out of its way almost to remain entirely opaque. Anyway, systems.
27:22
So, on to another thing, just in terms of a little bit back to things, this is a thing called the Gartner Hype Cycle. It's brilliant. It's absolutely brilliant. And you must all go and Google for the Gartner Hype Cycle. It's one of my favorite pieces of kind of internet comedy. But basically, there's this big company in America called Gartner, and there's like massive numbers of like NBA nerds.
27:41
All of the kind of guys out of Stanford and Accenture and stuff and Harvard that have just thought about business and nothing else for a year and a half, go out and they like read all the kind of tech press and gather together all of the kind of knowledge about technology. And then they produce this graph once every sort of three or four months, I think, but mainly once a year. And this one is from July 2013.
28:01
And they plot the sort of technology that's kind of being chatted about at the moment onto what's called the Hype Cycle. And they have at this beginning part this thing called the Innovation Trigger. So this is the stuff that's happening in all the labs at Imperial and MIT, the sort of the first piece of e-ink or the first computer vision that can tell when someone's smiling, all of those sorts of things.
28:21
And then it moves up to the peak of inflated expectations. So you can see up there back in July 2013, consumer 3D printing. Everyone was talking about how we're all going to have a 3D printer. It's going to be one in every corner shop. It's going to be the new photocopier, that kind of thing. And you come down to the trough of disillusionment when things sort of slide off there.
28:40
You have things like cloud computing and mesh network sensors are very, very unfashionable at the time. They're just sort of down in the kind of doldrums. And slowly, over time, they pick up into the slope of enlightenment, where you've got sort of biometric authentication. And if you go further up, like speech recognition into productivity, which for a long time, people like speech recognition.
29:02
No one wants to talk to computers. And now it's like all that Google and Apple can bang on about, right? They're all racing to the end there. So what we find, though, is that the stuff that's most interesting is there. It's the stuff that's quite cheap now. You think about RFID, which has never really left the trough of disillusionment.
29:22
It's never quite managed to find like a massive sort of secure market for itself. But it's fascinating. The reason why the things here are fascinating is because they're usually disproportionately cheap, especially when there's a hardware component. So if you went to a lab and you tried to make one RFID chip, like one tag for the first time,
29:42
I mean, someone must have done it. The cost must have been incalculable. Absolutely. It's a computer that you power with the electricity in the air, and it turns on and emits a meaningful signal through this paper-thin like thread of copper back to another reader. It's exquisite.
30:00
Just the level of physical engineering. Forget the software and the kind of stuff. Just the physical manufacturing engineering to usefully and consistently replicate that process. Forget about plastic boxes that go around the back of things. This is a computer in something less than the size of a playing card. And now, if you wanted to buy 10,000 of them, they cost you less than a penny each.
30:21
But to make one is incalculable. So it's a very interesting space that we find to look at. So I wanted to talk a little bit about some of the properties of a connected thing. And one of the things that's unusual about it is that it is software that is somewhere. And normally, in my experience, software exists in an abstract idea
30:41
in that it is something that can be replicatable. It's quite often based in the web. When you use something like Flickr or Instagram, you don't really think of it as being somewhere. Where it is doesn't really matter. You feel like you have a view onto a system and you are making judgments and deriving value from that view. But this is the thing that you press
31:02
in order to cross the street. When you press that, what you're really doing is using a piece of software that you can only really use in that place. So that button only exists there. So you do that piece of UI and the street changes and probably some massive scary system goes into operation and all of the traffic signals in London
31:22
realign themselves by a few milliseconds or whatever. And the world changes slightly. And so it's really located. To give you a bit more of a concrete point, this is a sort of cheesy, old-fashioned view of how people might one day, once in America certainly, consume television.
31:42
Television was a rare device. There wasn't one in every room. So it sat in a room and people gathered around it. Lots of people are looking at one piece of media. So they're having a shared experience of that. It's happening in the corner of that room. It doesn't happen in other places except for other places where there are televisions. So the weird thing about that is this.
32:03
And I've always found this quite a puzzle in that, you know, there's this kind of... If I've got my Netflix on and I'm signed into my TV and my girlfriend comes and wants to watch TV, legally, really, she kind of can't
32:22
because it's my account. And if she does buy some pay-for content through the Amazon Prime thing or whatever service we're actually using or some sort of pay-as-you-go-as-pay-as-you-view sort of on-demand kind of video, sort of HBO Go type service, then it's me that it costs money from. Plus, a lot of Netflix's model is a recommendations engine,
32:42
just as Amazon have, which means suddenly I'm getting weird recommendations for films I maybe don't have an interest in, and the model sort of begins to break. I mean, it's most conspicuous to me when you've got iTunes where it's kind of like you're constantly having to sign in and prove who you are to iTunes in order to spend any money on it. Now, I can see why, because kids run up, like,
33:02
$6,000 kind of app bills on, you know, Smurf Kingdom or whatever they're buying, like a jar of rainbows or a bag of stars or whatever. But, you know, I kind of feel like I'm 38. I'm pretty sure I don't want to have to sign into my TV in order to... It's just an odd security thing.
33:21
It's a bit like in case someone comes to my house and steals TV. So it's like in the middle of the night, I kind of get up, and it's like 2 in the morning, and I'm like, oh, what's going on? And I hear noises downstairs, and I gently pad downstairs to my living room, and there's like three guys in balaclavas, like downloading the Hobbit, you know.
33:42
And that's like the... I think there might be the least of my problems at that moment. So there's just something odd about the model here of something which is a television, which is a profoundly shared object. It lives in a house. People have expectations of how it will behave to the public. It's not in my pocket. It's not my device. It exists for people. But it's inherited a model that's derived entirely
34:01
from single sign-on on phones, and there's something there which doesn't quite fit. I mean, you can see Apple beginning to address it in their WWDC thing the other day. You know, they announced this family-sharing thing, so someone can actually buy something and give permission to someone else. So you don't have this kind of odd thing, and it's mostly, I think, because iPads are usually shared devices.
34:22
They live in houses. And anyway, that's just a kind of interesting quirk on the way that software is being constructed in that space. So in the kind of work we... I mean, as a startup, we still do work with manufacturers, and sometimes manufacturers use our system with their products,
34:40
and that's one of our business models. So we are often asked, what's the point of putting connectivity into products? And it's a very robust question, you know. And so I thought I'd have a go at answering it. And this is Robot JFK. Ask not what your product can do for software. Ask what software can do for your product. So I'm a little bit concerned about the time.
35:01
I don't want to run out. This is... I don't have any sound, unfortunately. But this is from the keynote that Steve Jobs made in 2007, where he announced the iPhone. And I'm going to play it. You won't be able to hear it. But you can see what he's mentioned there on the slide
35:22
is visual voicemail. He's saying, wouldn't it be great if you didn't have to listen to the four voicemails that you've already heard to get to the new one, which is number five, every time you want to check it? You just want it to appear, like, in a list. This is sort of a little bit astounding to me, really.
35:43
He mentions it again, like he comes back to this idea of visual voicemail. Just think about that. You've just launched your iPhone. It's like a complete paradigm shift in the product market. No one can make this. You've invented manufacturing technology. Your operations guy, Tim Cook,
36:00
has sourced completely unique deals around the chips to get the bill of materials to the point it is. You've probably had four guys sitting in a basement, like PhDs, just grinding out the maths to make pinch-zooming work on something that's, like, basically less powerful than a Raspberry Pi. And you're still talking about visual voicemail. You're like, forget about Candy Crush Saga
36:21
or Rovio and Angry Birds or the fact it's got a physics engine sitting in it and it's going to be more powerful than an Xbox in six generations. He's just talking about this quite modest, simple utilitarian value. You don't have to listen to the four voicemails that you have before you get to number five. And I think for designers, that's a really important lesson
36:41
because it's just sensible. You know, the problem is that in the business logic of the way manufacturers sit with software is manufacturers are really historically extremely bad at addressing their consumers. You know, if you think about a company like Electrolux,
37:01
you know, you own many washing machine brands in Europe, they really don't know who you are and they're really not that interested. They'd much rather turn a handle in a factory and big lumps of white metal go out the door and end up in retailers. And the people that actually sell that stuff are like local marketing teams or the retailers themselves.
37:24
The idea that manufacturers brand objects at all is quite weird, that this is an AEG washing machine or a Miele washing machine. It's a bit like saying, I've got a Foxconn phone that's running Apple software. Like, why? It's just like, who cares who made it? And I think in a culture where we understand the value of products
37:42
to be associated with the services that they have, you know, that feels very different. You know, I mean, can you imagine? I mean, this is just extraordinary, the story of this. There was a time, and I'm looking out there, and I think there are some people, yeah, I think we're all going to remember Walkmans. No one can pretend they're too young to remember Walkmans.
38:00
So, you know, can you imagine you're Sony? You've probably got the largest distribution and operational empire on the planet. No one can put chips and plastic in a hopper and get it into every store on the planet faster and cheaper than you. You've got a premium brand. You even own a slice of Chinese banking. You've got the PlayStation brand, and there's this thing called Walkman. So you're easily the coolest thing on the street.
38:21
It's portable music. You own all of it. Panasonic, who? Nothing. Just forget it. There's nothing out there. You actually define the data formats for media storage and then license to other companies, and everyone just chooses to do it. Plus, you've got Sony Music and Estates that you own legally, like half the music library. You probably own everything Madonna has ever written and Britney Spears and Michael Jackson.
38:42
So within, like, what, three years, some tiny little pissant beige computer manufacturer on the West Coast has just ironed your brand off the map. It just doesn't exist anymore. I mean, they must be pushing their fingernails through their hands. It's incalculable how stupid that was to let that go.
39:03
I'm a designer, but speaking to business nerds, they all kind of look away when you talk about Sony Walkman. You know, that was a bad one. And it's extraordinary how poorly they just understand consumer behaviors and habits. And when they sell you something, all they want back from you, and they don't really want it, is you know that you get that little postcard for the warranty?
39:23
That's their equivalent to an Apple ID. That's their Gmail address, as far as they're concerned. That's all they want to know. They'll see you later. Go and buy the cassettes from the store. And so it's just grossly broken. This is recently. If you go to sony.com, this is what you get.
39:42
Do you want to be marketed to in American or Norwegian? That's the question that you've been presented. When you go to sony.com, why don't you go to google.com? You get one of the most powerful tools in living memory. Or if you go to apple.com, you get sold an iPad. And they do that well. So it's like a phenomenally weak way of actually understanding consumers.
40:02
There's just no way that manufacturers are going to be the people that own the logic of this space or occupy it in our public imaginations in terms of how connected products will work. The reality of a connected product is really just an extension of a piece of software. It happens to be manufactured by someone who isn't Apple or Amazon in that instance. Maybe you manufacture it. Maybe someone else manufactures it.
40:21
But the reality is it's software that reaches a place that was hard to do before. Yep, he mentions it again. I won't click through this. There's no sound. So just to kind of take back to kind of industrial design of where the value of this kind of stuff is, probably in the short term in terms of how connected products in domestic environments can advance. You know, once there was this company in Britain called Singer
40:42
and it made sewing machines. And they're very important. And this thing has a treadle on it. You kind of do that with it and it turns the spindle and that makes the sewing machine go. And at some point, a smart person went, hey, electricity and motors, that's handy. You know, you put that in there, you don't have to do the treadle anymore. And this must have been quite a serious undertaking for a company that basically made cast iron mechanical stuff
41:04
to turn into a company that injection molded a piece of what's really a piece of consumer electronics. And they did. And so in a way, I think what we're looking for are the kind of few humble steps that mean that connected products are marginally more useful than their unconnected ancestors.
41:22
So here's Whirlpool, another big consumer electronics brand, deals with this kind of space. And we'll just have a brief look at what they're doing in this environment at the moment. So this is for Whirlpool. Introducing Whirlpool smart appliances, you know smart means something like this is going on, right, as some computer in your business
41:40
beyond just the embedded software in a chip that means the water gets hot and cold. With Sixth Sense Live TM technology, beggar that trademark in, don't want anyone stealing Sixth Sense Live. You know, it's like this is what they're saying. And these are two of the products in the list. And these are things you can buy them.
42:02
You can really go to the store and buy this. I haven't got one of these. I haven't spent a lot of time playing with them. But let's just speculate on a minute on the dishwasher here. The three unique selling points of this product with its Sixth Sense Live technology are remote control lock. Try and imagine this.
42:20
You've got your dishwasher out. You slot it in. The plumbers come in, put it in there. There's a hole in your kitchen. You've got it in there. It costs you some $400 more than exactly the same dishwasher that doesn't have Sixth Sense Live technology in it. You've got it in there. Now you've got to get it on your Wi-Fi. I mean, you've got a Samsung TV with a whole screen and a remote control with 600 buttons, and it's hard to get on the Wi-Fi. Here you've got a dishwasher with, like, six buttons on it.
42:42
Can you imagine trying to put in, like, a wet key with, like, seven keys and no screen? Like, who knows how that magic works? But it's somehow... Maybe it's Bluetooth, and there's some magical kind of pairing thing. Let's just assume it's Bluetooth, and you just go near it, and it kind of knows now that that's the phone it's slave to or something like that, and they've thought about that. And you get it in there. And this is so you can launch an app,
43:02
which you then have to find on the App Store. So you go to the App Store, find the app, download this model, $0.06 Live technology. It comes down so that you can go... Doink! ...to open the door with your phone. When do you want to open the door of your dishwasher when you're not at your dishwasher? I mean, it's completely ludicrous. Imagine the money.
43:21
Given what we've just seen about manufacture, it's miserable business. It costs a fortune. Billions of dollars sits on container ships for months before you even get to charge for the product. And all you're doing with it is opening the door with your phone. It's just ridiculous, just profoundly ridiculous. And, of course, we thought we could do better
43:42
because we know it all. So we made our own washing machine, which we literally built, called CloudWatch. And it's a prototype. We're not manufacturing washing machines. Not yet. That's going to be someone else's pain. But we basically did a few things where we took an existing washing machine, an electrolytes washing machine, turns out... And we kind of hacked it.
44:00
So the first thing we wanted to understand is how washing machines really work from a manufacturer's point of view. So we added some of our own connectivity in there, and we realized that, yeah, lo and behold, we can control it. So if you press the start button at the bottom there, helpful video person, the light goes on in the background. So we also thought about how the machine might change in order to accommodate the network. What sort of things do you need on a machine
44:21
in order to talk meaningfully to the person using it about what's happening? What are the new functions and new activities you can have associated with it that help it to be more useful than washing machines were before? Because I don't see that many people complaining about washing machines. I mean, we're not sitting here in dirty clothes, so they can't work that badly.
44:40
So one of the things that we realized was no one uses the 17 different presets on their first wheel. So we just made a little ink display where you can change the presets to the three things you mostly use and label them properly. Then you can see when the wash is going to finish. For those people who do washing, that's a big deal because you don't want your clothes sitting in stagnant, damp,
45:00
closed washing machine growing mold. And you can turn the notifications on and off. The thing we stumbled over, though, that was perhaps most profound are the last two buttons on it, which is that when you run out of liquid, you're at the machine. That's when you know you've run out of washing liquid or detergent. And we configured this machine
45:20
so when you press the detergent button, it orders you one from Amazon. Or it adds it to your Ricardo list. And this is the thing that's caught the attention. We put this on the internet. Many people kind of looked at it and wrote blogs about it and things like that. I'll come back to the button in a minute. This is John Lewis in London.
45:42
That's a John Lewis washing machine. John Lewis are a retailer. They're just a big, nice, posh department store. This is a John Lewis washing machine. Really, this is an Electrolux washing machine. It's actually a Zanussi washing machine, which is in the Electrolux brand. We're talking about... So in Electrolux, every single washing machine that doesn't have a screen has the same printed circuit board in it.
46:02
You know like the LEDs behind the buttons, they're in the same places for every single Zanussi, AEG, la-la-la Electrolux washing machine that is on the market. The dial is in the same place proportionally. If you were to measure them on every single different brand, they would all be in the same place because the circuit board is literally identical.
46:20
So all John Lewis have done is they've just asked Zanussi to supply them with a special machine and they've changed the fascia. They've just changed the plastic on the front, stuck the John Lewis brand on it, and now they're claiming it's their machine. And this is the way that OEMs and manufacturing works. It's all a big, lazy system of deals and shipping and stuff like that. But what's interesting about it is that they charge 50 pound more for this machine
46:41
than they do for the equivalent Zanussi model, which is literally the same engine. It's the same circuit board. It's just got different plastic. If you open up a washing machine, there's a circuit board. There's a circuit board at the front, which is called the control board, and there's another circuit board at the back. Sorry, there's a control at the front, which is called the interface board and the control board that sits at the back.
47:01
Interface board does the buttons and dials and the program stuff, and the control board at the back is the bit that makes sure your house doesn't go on fire or flood and all the motors get controlled and the tachyometer's reporting properly, and that's the bit that does the heavy lifting. So it's relatively easy to intervene in that board. This is the board that's in every single low-end
47:21
electrolyte washing machine on every brand. That board is called the EWM-2200. It stands for Electrolux Washing Machine 2200. There isn't this massive diversity of electronics. It's extremely simple, what's going on here. And that chip right in the middle there is relatively straightforward microcontroller. It's not running Linux.
47:40
It doesn't do anything scary. It's like the Raspberry Pi looks like God to it. It's a really, really novel, tiny little thing. There's nothing on here that's that expensive. It's just happening. This is where we come back to the button. If you speak to manufacturers like Samsung or Electrolux or companies that make kitchen goods or houseware goods, they have a metric about these machines,
48:02
which is the value of the FMCG that goes through them. One of the reasons why the fridge is so significant to a company like Samsung is that you put 3,000 pounds through it every year compared to only 400 pounds through your washing machine. So that's why they're interested in the fridge, because the value of the services associated with that product
48:22
are much, much higher than other products in the kitchen. Your toaster just makes bread. That's cheap as chips. So they're not as interested in it in terms of the service. This is this entire area of supermarkets, right? Everything you see on these shelves only exists because of machines. Like, of course, there were detergents before,
48:41
but dishwasher tablets, these little gels, all of that kind of stuff, the whole kind of fog and universe of those things, they're only about servicing those machines. So really, you buy a washing machine, it consumes power and electricity and water. We know that. But also, it consumes conditioner, little tablets, these scalers. These are vast, squirreling dollar industries. Those companies are rich, right?
49:00
Procter & Gamble aren't going home hungry. There's a lot of money in Unilever getting you to buy weird little liquid tabs with, like, three compartments and a little yellow gel and a blue gel and a kind of little yin-yang shape on it, all of that kind of stuff. And so if you think about it, here's an espresso machine. Espresso don't manufacture those machines.
49:22
Krupp's and Magimix manufacture them. They sell you these coffee pods. Those coffee pods are about 37 pence. Think about how much coffee is in a tiny aluminium packet. We're talking about this unimaginable luxury markup. It makes Apple look bad. We're talking about, like, seven, maybe 8,000% markup on that coffee
49:41
compared to what you get when you buy, like, beans or ground. Even Starbucks are like, whoa, that's good. You know, you go into Starbucks and you buy a coffee, buy two pence worth of coffee for £3.90. That's why there's so many coffee shops. It's biblically profitable. There's literally nothing like it. So if you've got...
50:00
An espresso don't make those machines. They design them. They own all the patents on it. They just give it away to someone else to make it, and they want to get it into your house as cheaply as possible because what they want to do is sell you 37 pence capsules. If you've got a washing machine on it and the washing machine has got a Buy Now button for the goods that are worth £400, £500, £3,000 a year inside it, then Unilever might as well give you a washing machine
50:22
with the button to their goods on it, and if you press that button, you're buying their unit. You're buying their parcel. You're never going to go to the shop and buy another one. You just go, ha, buy now. You know, just like you do on your iPhone. You know, people buy apps twice because they forgot their password. It's just easier, right? It's just that kind of stuff. So for me, some of the connectivity value
50:42
and the disruption in the way the businesses work are that the services and systems and economics of those machines may get radically disruptive. It's not Electrolux or Sony that are going to do that. It's retailers. It's Amazon. Yeah, Amazon have WhisperNet, which is like the component, which is the relationship and technology in the Kindle,
51:01
which allows you to download books when you're on holiday, right? If you go on holiday and you're walking on a bus and you think, oh, I want to get the new Dan Brown or new Harry Potter 20, you know, and off you go. You do that. The publisher pays for the data transfer, and it just comes magically through the air, right, through some weird deal-making with all the 3G companies in the world, Vodafone and Sprint in the U.S. and whoever it is,
51:23
and it just works. You're a consumer. You don't have to pay a bill. It's being dealt with in the background. So if Amazon took that technology and flipped it open so that anyone could put, essentially, a WhisperNet component in their consumer devices, then you'd end up with something quite radical. What if they just supplied that chip?
51:41
It's not hard. That chip's been around for a decade. I mean, it used to be powerful. It's probably powerful enough to run cruise missiles, but, you know, most things are these days. It's just not hard to get powerful chips, especially if all you've got to do is turn a drum and a few motors and sensors. So for a very, very small adjustment to the bill of materials of this device, it becomes a unique purchasing point for FMCG companies.
52:04
Last point. I'm going to be quick. I've only got eight and a half minutes. This is one that you're going to know more about than I do, so please forgive me for just continuing to belittle you with simple and ridiculous truisms that you probably wrote all the software for in actuality. But I think this is one of the most brilliant things
52:22
that's ever happened. And I think a few people realized that when you go onto Google and you kind of go, huh, pizza, mm, or, mm, Angelina Jolie's got really nice cheekbones, mm, or whatever you've written, right? And it kind of goes like that. And you go, no, not that one. You go penguin. And it goes, ah, penguin. You're not that penguin. Batman penguin, you know, all that kind of stuff.
52:40
And you just kind of ask it things. And look at this, 213 million results in a quarter of a second. I mean, you know more than I do, right? But that for me is comparable to the fucking moon landing or something in terms of engineering effort. How do you get that to happen? You know, it's really amazing.
53:01
It's at any moment. And anyone can do it right now sort of for free. And I think there's something happening around cloud computing which, you know, you won't need to hear from me, which is this idea that you can get almost infinite computing power associated with very, very modest computers.
53:21
So this in the bottom right of Ryan corner here, this is a company called Arm that you all know about. And they have a prototyping system called Embed, which is without the E's. So it's just letter M, B, D, if you know what I mean. And it's a bit like that kind of Raspberry Pi Arduino sort of thing. It's for prototyping that kind of hardware. This is the cheapest board they do that can meaningfully
53:41
handle audio. It costs about five pounds for the whole development board. So let's just say the chip's a pound. Very, very low, simple costs. You know, it's down in the gutter compared to iPhones and TV set-top boxes and stuff like that. And there you've got the cloud. So let's just say that's all of Google, right? What that means is essentially you can speak into that chip
54:03
literally just with your voice and say pizza, and it can read out to you like the actual answers, 213 million results. The challenge is only to make it perceivable to the person using it. But what's interesting is that, historically, you needed big computers to do big tasks, and now you don't.
54:20
There's something about objects like this, simple, cheap junk that you bought in the supermarket that's actually able to take advantage of the kind of computing power you just can't put in a room unless you're Amazon and Google. Just big, scary server farms. And we've done some projects recently that kind of take advantage of that, and it's very weird. You've got this tiny little idiot of a processor with, like, a little bit of UI on it,
54:41
and it's doing all of Google at you. It's very, very powerful. It feels extraordinary. Yeah. All right, I'm going to skip this one, I'm afraid. Oh, no, I'm all right. I've got five minutes. There's a project we did with Google. So we worked with big technology companies, mostly in the U.S., like Google and Intel, et cetera, et cetera, thinking about new forms in UI,
55:00
new kinds of things they can do with their systems. And we did a project called Lamps. Now, basically, the weird thing about Google is they're so extraordinarily rich. I've never come across anything like it in terms of how loose they are with their money. They just kind of go, we like you guys. What do you want to do? We're like, well, normally with clients, like, you do what you're told, so why don't you tell us what to do? And they're like, no, they just do what? What do you want to do?
55:21
And I'd heard they did this kind of book scanning thing where they're like, yeah, we're just going to put all the books in the whole of history and make it part of the Internet so we can scan it and do more AdWords. And we're like, okay, and how do you do that? And they're like, well, we just built a big, giant system of robots and interns, and basically, we just read all the books with a computer. And these machines are incredible, right?
55:41
So they've got these scanners that get books. And because books aren't really 2D, they're like, you know, they've got Bo in them and Flex. They have like a LIDAR, like a Kinect, a laser scanner that shoots at the book, that scans the 3D surface of the book on that page. And then it takes 16 photos at every focal plane across that depth.
56:00
And then it takes the 3D data and all 16 photos and chooses which part of the photo that's in focus given the actual depth of the book at that point and re-stitches together a magical 3D version of the book where all bits of it are in focus and then uses that as the source of the optical character recognition it uses to generate the text in the book. You're like, what? And then some poor intern goes, you know,
56:22
turns the page, clicks, does it again. That's amazing. Imagine if you had that as like a domestic product. Imagine if you just made that part of your desk. Imagine if on your work desk you had this thing that can scan anything, read anything, and maybe project back out. So we did a project where we got a bunch of like
56:40
Kinectis, LIDAR, laser scanners, high definition cameras and projectors and started trying to read, analyze, and project back out what it was seeing on the desk so that the things in the world sort of become magically computational. We made a short film. So we made these blocks, which is really a game about where the computing happens. This is real projection.
57:02
This isn't doing any real computer vision. It's just real projection in one of the early video prototypes. We did do real stuff, but we were also doing this kind of fake stuff. This is just a wooden block with some springs in it. But when you use it and you play with it, there's no electronics in it. No signal is being sent to the computer. When you press those buttons,
57:21
the computer just sees that you're pressing the button, changes what it's projecting on it, and starts playing music. So this weird dumb block, this just dumb completely analog lump of wood suddenly becomes like it's computational, even though the computing is happening in a server somewhere or in a computer locally or in this weird lamp that we've got set up.
57:40
Anyway, it's a very interesting project. But just again, it's to reinforce that point about when you free yourself from the idea that computing cycles are rare and you say you can have as many as you want, then the objects in the world start to change quite radically. We make everything. He has to make everything he is thinking about in order to express it. It's from a film called Close Encounters of the Third Kind,
58:02
and poor old Henry Jafus gets obsessed with an alien landing and starts continually building this mountain. If you get the reference, it makes sense because you're a sci-fi nerd. If you don't, apologies. It's a bit weird. Why do any of this is a really good question. This is usually something I say to designers
58:21
because as software people and developers, you're already doing it. One of the things I like, this is from a blog post on the O'Reilly radar. O'Reilly, a big technical publisher, no doubt you know them. Jobs says, We believe that it's technology married with the humanities that yields us the results that make our hearts sing. The post writer, Doug Hill,
58:42
says that basically he was saying that Apple products have soul and that people are attracted to those products because they can feel that soul both consciously and unconsciously. There's just something nice when software is doing something really good, when it's beautiful, that it's just like the most extraordinary stuff that's ever happened. That's why I continue to work in it
59:00
and why I think most people should because configuring your Samsung Wi-Fi is miserable and it doesn't have to be. Thanks.
59:23
Paper in a box. There's a green bit of paper or a red bit of paper or a yellow bit of paper depending upon how stupid you think I've been. You put them in a little plastic box and that tells the organizers how bad it was as an idea to bring me here.