Where is this IoT bus taking us?
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 | 84 | |
| Author | ||
| License | CC Attribution 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 purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/32437 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
FrOSCon 201647 / 84
2
4
5
7
11
16
25
26
27
28
29
30
31
32
34
36
37
40
41
43
46
47
48
49
50
52
55
56
58
59
62
63
64
65
66
67
68
69
70
71
72
73
75
76
77
79
82
83
84
00:00
Bit rateComputerProjective planeData storage deviceWordExecution unitBitInternetworkingPoint (geometry)Channel capacityMultiplication signInternet der DingeForm factor (electronics)Computer programmingSummierbarkeitPlastikkarteKernel (computing)WebcamPower (physics)Medical imagingFundamental theorem of algebraXMLUMLLecture/Conference
03:08
Data storage deviceMaxima and minimaInternetworkingPairwise comparisonNumberElectronic visual displayRight angleEvent horizonProcess (computing)
04:22
Point cloudComputer virusKälteerzeugungMultiplication signDifferent (Kate Ryan album)MereologyReading (process)Topological vector spaceInformationInternetworkingVariety (linguistics)Food energyCuboidWell-formed formulaTraffic reportingBitInternet der DingeOvalDependent and independent variablesSoftware testingGame theoryPower (physics)Web browserLine (geometry)Figurate numberGodInheritance (object-oriented programming)Electronic visual displayLattice (order)MicrocontrollerDigitizingDefault (computer science)Information securityRight anglePasswordPerturbation theoryHookingPoint (geometry)Hidden Markov modelBus (computing)Cellular automatonPlastikkarte
12:47
Cellular automatonFood energyExecution unitRight angleComputer programmingVideo gameDivisorPoint (geometry)VideoconferencingPhysical systemYouTubeDifferent (Kate Ryan album)FreezingSmartphoneSoftwareMultiplication signBus (computing)Computer networkUniverse (mathematics)UsabilityComputer fileElectronic mailing listDialectInternetworkingProduct (business)Reverse engineeringProxy serverLink (knot theory)Latent heatRandomizationLevel (video gaming)Software bugInternet der Dinge1 (number)Mobile appEntire functionSynchronizationVirtuelles privates NetzwerkSlide rulePoint cloudComputer animation
21:12
CountingSound effectBitAuthenticationSoftwareFood energyInformationPhysical systemSpeech synthesisPoint cloudLecture/Conference
22:45
Self-organizationComputer hardwareGame theoryInternetworkingInternet der DingeSoftwareFood energyGame controllerFamilyControl systemLibrary (computing)Greatest elementMultiplication signRight angleVideo gameComputer animationEngineering drawing
25:33
Food energyPlanningMereologyGame controllerBuildingRight angleMathematicsMultiplication signError messageLecture/Conference
26:15
Information securityOpen sourceKernel (computing)Open setWeb browserPower (physics)Firewall (computing)Vulnerability (computing)GUI widgetSoftwareView (database)Computer hardwareWordMoment (mathematics)Point (geometry)GodComputerInternetworkingAuthenticationComputing platformSound effectBuildingMicrocontrollerEntire functionFrequencyKeyboard shortcutMultiplication signGradientData storage deviceGame controllerRevision controlEndliche ModelltheorieControl flowWeightSoftware testingFunctional (mathematics)Marginal distributionReal numberWebsiteExecution unitControl systemCloud computingCommunications protocolMassAlphabet (computer science)NumberMathematicsMereologyNumbering schemeService (economics)Different (Kate Ryan album)Latent heatAndroid (robot)Goodness of fitProjective planeType theoryFirmwareComputer animation
35:52
Moore's lawAreaPower (physics)Software testingComputerComputer programmingPoint (geometry)Channel capacityHypermediaRow (database)Video gameComputerTerm (mathematics)Lecture/Conference
37:46
Flash memoryOpen sourceSoftwareSerial portInternet der DingeComputer hardwareVoltmeterSoftware testingFood energyPower (physics)MathematicsParallel portModemQuicksortGoodness of fitStudent's t-testContent (media)Error messageComputer animation
39:28
Power (physics)Goodness of fitSource codePatch (Unix)SoftwareComputer hardwareInformation securitySpeichermodellVideoconferencingComputerTouch typingCodeGraphics tabletYouTubeWhiteboardData storage deviceAreaComputer programmingMedical imagingPersonal identification numberSoftware bugComputer fontLie groupMatching (graph theory)Gamma functionRight angleResultantStatement (computer science)Online helpTexture mappingUbiquitous computingPerspective (visual)Representation (politics)FluidException handlingLecture/Conference
44:56
Digital photographyComputer animation
45:48
System callComputer hardwareAbsolute valueCASE <Informatik>Multiplication signStack (abstract data type)Inclusion mapRoutingField programmable gate arrayComputer configurationProduct (business)Cycle (graph theory)Instance (computer science)Virtual LANSoftwareBitMathematicsOpen setPairwise comparisonPoint (geometry)Information securityFirmwareOutlierOpen sourceReverse engineeringSystem on a chipNormal (geometry)Endliche ModelltheorieWhiteboardAndroid (robot)InternetworkingAxiom of choiceBefehlsprozessorHacker (term)BuildingShared memoryComputing platformMoment (mathematics)AreaExclusive orAssociative propertyArmMusical ensembleLink (knot theory)ResultantSpacetimeFunctional (mathematics)Suite (music)Likelihood functionSpeichermodell
55:17
Computer animation
Transcript: English(auto-generated)
00:07
So I guess we'll go ahead and get started here. And we're going to start all the way back at the beginning with the ENIAC. And this was not exactly an internet of thing. It was just a thing at the time, because it was the first general purpose computing
00:22
device. And as you can see, it was huge. It was big. I mean, filled an entire room, took like two or three people to actually program it, and ran at the whopping speed of 0 point very negligible megahertz. And these were great.
00:41
This is a device that actually defined every computing device that has come since then. And even 20 years later, after the ENIAC, this is 5 megabytes of storage. And they're loading this onto a large cargo jet at the time.
01:02
Actually, I'm sorry. It's not even a cargo jet. It's just a large cargo airplane. That's 5 megabytes of data. How many people, when you snap a picture, generate more than 5 megabytes of data? The entire room should be raising their hand. So you can't even fit something that your cell phone camera is generating onto one of these.
01:23
I mean, every time I snap a picture out of one of my real camera, it's like 26 megs. I would need five or six of these things just to store the bloody image. And this is the internet. Quite literally, in March of 1977, you can draw the entire internet on a piece of paper.
01:44
So this is 20 years even after the picture I just showed you. And it's not very big. But everybody understood it at that point. And everything that is on that, I mean, this is literally the dawn of internet of things.
02:03
This is good old ARPANET. This is the prelude to all of the internet. And this is where we work. This is what has started the fundamental creation of internet of things. And this is where we are today. We've got cell phones that have more computational capacity
02:21
than every computer that existed until the late 1980s. And that's in your pocket, let alone just the GPU itself to make the little animations in Pokemon Go dance around and you throw Pokemon balls at them. It's mind boggling how much power is in a cell phone these days.
02:41
Storage. So I showed you a picture of five megabytes. The smallest chunk of storage, removable storage, that you can get your hands on easily these days is 200 gigabytes in a micro SD card form factor. That's something that's a third the size of a postage stamp that's holding 200 gigabytes of storage.
03:05
And the internet has grown just a little bit. This is an attempt to map the internet in 2011. It's gotten bigger since then. And the problem is is that this is all technology that was there five minutes ago.
03:22
Everything is changing so rapidly these days that even trying to describe what is currently out there is outdated. Because storage, when I double checked this earlier today, 200 gigabytes of storage may have been the maximum you could have bought. Within the last five minutes, although it is a Saturday,
03:41
somebody could have announced an even bigger chunk of storage. The internet has grown so astronomically since 2011 in comparison to that picture that it can't even really be displayed. I mean, there are so many nodes now on the internet, particularly with the advent of IPv6, that it's hard to even fathom the number of devices that are getting connected to the internet
04:02
and the number of cell phones that are out there. I mean, how many people have a drawer full of cell phones that they've just thrown away because they still have data on them and they don't wanna like do anything with it. I'm sure everybody in this room has a drawer full of them somewhere. I know I do. I think I've got two now actually.
04:21
But that's where we're at right now, kind of. But this is where we're headed because really this is the IoT bus. And we wanna know what's coming. What's around the corner? Why is this all being done? And what's all broken with it? And what can we do about that? And that's the interesting part here.
04:41
So, an oven. An oven's a really simple device. You plug it in to either a gas line or an electrical main. You turn it on and it makes things hot. And you turn it off before your house burns down. This is a really great idea. So what's the one thing we should do that, let's hook it up to the internet. Because that's a great idea.
05:00
Because then I can turn it on and off remotely with my phone. Because that's not a bad idea at all. Coffee maker. Because the first thing you do when you get up in the morning is you know, you pull your phone out and you're like, I want coffee. Honestly, I can't actually see anybody doing that. I'm not sure that the phone works before the coffee's been had.
05:21
So you've got kind of a chicken and egg problem. I need to turn the coffee maker on. But I, crock pots. Again, a device that makes things get hot. Controlled by your cell phone. I mean,
05:42
and kind of venturing off. I mean, IoT really is pervading everything. So you know, I've talked about three little kitchen gadgets. First response in the last six months, I believe, came out with a Bluetooth enabled pregnancy tester. This is a, honest to God,
06:00
honest to God, it's a $20 pregnancy tester. This is literally throw away Bluetooth low energy device. It's a one time use. And if the device finds that you're pregnant, it actually pops up all this information on your cell phone,
06:21
giving you offers for diapers and where to go and do all these kinds of things. I mean, it's an absolutely brilliant and horrible idea at the same time. And I would love to have been in one of the marketing meetings where somebody goes, so we wanna add a Bluetooth low energy device with a microcontroller to this whole thing.
06:40
Because the amount of technology that is required to get from, to read out even the data off of the test strip is astronomically amazing. I mean, it's, this is medical, this is modern medical science
07:00
where technology is actually capable of doing on the spot reading of information and then sending it to your cell phone to send you coupons. And of course, you know, with this being Internet of Things, it connects to the cloud and it reports back to first response on a bunch of different things. And there's another device
07:21
that I'm not including in here for some very specific reasons. If you're very curious, go and take a look at some of the talks from DEF CON this year. It's another very personal device that was found to be, while it was being used, was actually reporting data back to the parent company. This is incredibly private information
07:41
and it's being reported back up to the company and these things exist. We've got, you know, continuing in the medical line, we've got pacemakers. How many people want something that's connected directly to your heart being controllable via Bluetooth? That's a great idea, right?
08:00
How's the security on Bluetooth? Anybody want to give me their Bluetooth password? Really? Takers, takers? And, you know, okay, you know, you've got, you know, medical devices, but, you know, home locks. Out of curiosity,
08:21
how many people have a digital lock on their houses at home? Anybody? One. Z-Wave? No? Oh. Oh, so is that the internet of parents?
08:41
But I mean, you know, technology is now literally opening doors for us. You know, you walk up. Oh, the power's out. I can't open, oh, hmm. And, you know, in the same vein as door locks,
09:02
we've got power outlets. You know, these devices that connect in between, you know, like your refrigerator and your wall outlet. And what's the worst possible thing that could go wrong with a power outlet being connected to the internet? I won't actually answer that question
09:20
because it's really, really horrible. Oh, okay. So the worst thing that could possibly happen is you hook this thing up to your refrigerator, you see that button on the front? Because this actually happened to me. If somebody bumps the refrigerator, turns it off.
09:44
At which point you're like, huh, why doesn't that work? You figure out what's going on, you turn it back on and then somebody bumps the fridge again and the whole thing turns off. At which point this part didn't happen. You get botulism poisoning because everything in the refrigerator goes bad. And you know, you die.
10:02
And speaking of refrigerators, the internet-connected refrigerator. How many people have ever even seen an internet-connected refrigerator? How many people want an internet-connected refrigerator? Honestly. Okay, one. Okay, why do you want it?
10:21
It's a nice phone? Oh, a nice toy. I'm sure that people, there's a lot of other people who thought it was a very nice toy too. A few years ago, every internet-connected fridge in the world was used as a spam relay. That's a really nice toy.
10:46
But yeah, I mean, it'll display your calendar, it'll tell you when the next soccer game is and it will tell you exactly what that little blue pill that we all desperately or so desperately need is and where we can buy it on the internet
11:01
because it's totally not fake if it's on the internet. And I mean, this is just the default security permissions because there was a web browser and people would go browse the internet from the refrigerator. Doorbells! You know, you push the doorbell,
11:20
it lets you know who's there and you can, you know, this is kind of like the door locks. And TVs, oh, TVs. How many people have a smart TV? I'm sorry. How many of those have cameras on them? Anybody?
11:41
One brave soul who's going to admit that he's probably been infected with a virus on his TV and is streaming himself sitting on the couch watching TV to everybody. Oh, you've got a sticker on it? Oh, you are better than most people then. So I'm gonna digress on the TVs a little bit because this is gonna be kind of a digression
12:01
for everything here. Internet-connected things are scary. They're really, really scary because once they're connected to the internet, they have a tendency for a variety of reasons to want to dial back home somewhere or give you access to, well, the internet. I mean, this is kind of the internet of things. And the internet is a really dangerous place.
12:22
It is a surprisingly dangerous place and there are, you know, people setting up spam relays on refrigerators. And, you know, there's a lot of new evidence that's coming out about smart locks being trivially openable from the internet. Ovens and crock pots accidentally burning things down.
12:42
I mean, this is, you know, one of the things that we've got to, you know, really start thinking about as we, you know, build these devices. Because IoT isn't going away and it's gonna get more pervasive. Like, you know, thermostats. How many people have heard about the recent issue with,
13:04
I believe it's the Nest, I think it was specifically the Nest, just completely wigging out and turning off heat in the middle of winter? Because, you know, a couple of people have heard about this. If you haven't, it's actually a really interesting article. Go look it up. But the short answer is there was a software bug
13:22
and it just turned off heat in thousands of ohms in the middle of winter. Because that won't lead to burst pipes and people freezing to death and, you know, pets freezing because they were left in a house that they believed was going to be heated. I mean, there's definitely some interesting aspects to that
13:42
or to all of that. And again, so we've just talked about thermostats, which, you know, aren't that dangerous. Let's talk about a smoke detector that's connected to the internet. What could possibly go wrong? I mean, everybody's been woken up by that. It's, I swear, it's 3 a.m. every single time this happens.
14:04
It's 3 a.m. and the battery alarm starts going off on your smoke detector, right? So what happens if, you know, somebody can just trigger that all the time? Like every 20 minutes, you know? Or, you know, silently not tell you
14:22
when there's actually a fire. Or, you know, carbon monoxide or something, you know? And thermostats, I mean, not even thermostats, but thermometers connected to the internet. This one, in fact, you plug into your phone.
14:42
You can see the headphone jack on the left there. And you plug it in, you get your temperature, and why does this need to be connected to a cell phone? Do I really need to upload my temperature being 98 point something or other in the magical Fahrenheit system
15:01
that apparently is magical and nobody understands over here? Sorry, I'm from America. Oh, I'm sorry, Merica. Oh, what happens when you plug the other side? I don't know, actually. I would not recommend trying.
15:22
Oh, you can take the temperature of your smartphone? Maybe. I don't think it'll work that way, though. I think you'll just short something out and everything will explode. Which, you know, gets hot at that point.
15:40
God, I love these slides that for no other reason than I get to just joke about random products that exist. Okay, and the device that everybody will recognize. This is one of the greatest IoT devices in the world. The Linksys WRT54GL. Do you know how many things the WRT54GL
16:01
has been plugged into and hooked up to and made do things? Everything. You know, thermostats, thermometers, Roombas. Never seen a WRT54GL running around on a Roomba, usually with a knife. It's an amazing thing, go look for the videos on YouTube.
16:22
I wish I could actually show them, but I didn't think to add that. But these are one of the most, I mean, the WRT54GL is just indicative of a device that everybody has. And, you know, everybody in this room probably understands how it works. How many people know how a network works, actually?
16:42
I'm, okay, three quarters of people. Three quarters of the audience raised their hand. How many people think that everybody else in the world understands how this works? No one? No one's gonna take that?
17:02
Oh, yeah, it's, and you know, and I put this up here for a very specific reason. And I've got a few more devices I'm gonna show in a minute, but every one of these devices wants to be on the internet. But nobody knows how to configure that except for about the, you know, three quarters of the people in this room.
17:21
Which none of you get hit by a bus because the world would lose a great many people. So all of these devices figure out really neat ways to dial back home. Usually by setting up a reverse proxy, they dial back out and then they connect to some magical central server
17:40
so that your cell phone, when you're standing here in Germany, can talk to your device that's sitting in Portland, Oregon. Because, you know, not doing that over a VPN, you know, one, VPNs are hard, nobody can set it up on that, except for the three quarters of the people in this room. And two, well, it's just usability issues.
18:02
You know, most normal users are, you know, they don't understand their networks very well. And, you know, and some of that's, you know, our own fault for making networks hard and some of that's just because networks are hard. And companies want to make, you know, everybody, you know, wants to be able to control their thermostat from their phone.
18:22
They want to be able to control their IP, you know, look at their IP cameras or the random outlets or their coffee maker, assuming they're not first thing in the morning. You know, and they want to be able to check, oh, did I leave the stove on? How many people would have loved to have had a, did I leave the stove on app for their phone when they're going away?
18:41
Yeah, a couple of people. I mean, you know, the convenience factor for a lot of these devices really is there. The internet is actually bringing something to these devices that didn't exist before and is really important and really powerful, except for the coffee maker. I still, I again, don't understand.
19:02
You know, the smart outlets, I mean, a lot of these smart outlets even include devices to monitor energy usage. And if they reach some magical level, they'll turn things off. I mean, I know people who are hooking these up to soldering irons to prevent themselves from accidentally burning down their shops and their labs.
19:21
I know people who are hooking them up to all kinds of different things for either safety purposes or monitoring purposes, or even just, you know, oh, well, I'm not in that room. I'm just gonna turn it off. Because everybody carries their cell phone around with them all the time, right? This is the most wonderfully trackable device
19:41
in the universe. Constantly broadcasting Wi-Fi, constantly broadcasting cell signal, and for most people, constantly broadcasting Bluetooth low energy, or at least Bluetooth. You know, if you had the right sensors around your house, you could pinpoint yourself within inches inside your house, and you could turn devices on magically
20:01
as you wander from room to room. Because, you know, that's both a brilliant idea and one of the scariest things I could even remotely think of. And we've got Fitbits. How many people have a Fitbit? Is there only like three of us in the entire room?
20:23
Okay, maybe this is more an American thing, but almost everybody has one of these devices. And the really neat thing is, is I could be an absolute jerk right now and run the sync program on my cell phone, and every Fitbit in this room will get synced to the cloud. And I've never seen you before in my life,
20:42
and there was another gentleman over here, and I've never seen you before in my life. And yet my cell phone will pull data off of a device that you are carrying on you, filter all of that through it, which means that I also have all of your data. How awesome is that?
21:12
Yeah? And I mean, and some of the Fitbits even have rudimentary ability to track
21:22
kind of where you're at. Not quite GPS accuracy, but they can kind of give you, you know, you walk like this way. I mean, these are really complicated accelerometers. And you can do dead reckoning based on a really good accelerometer, which means that all of that data about where you've been, yeah, I have that now too.
21:40
I mean, for all I know, you know, I could pull up this data and you could have been at McDonald's an hour ago. Probably weren't, but I'm sure you were in a talk. But you know, and you know, a lot of the Bluetooth low energy devices that are out there actually do this. They do no authentication. They just broadcast information.
22:01
You know, specifically when the Syncbit or the Fitbit system just queries, are you a Fitbit? I would like your data now. And it gives me all of the data and then I upload it to the cloud. Unless I'm a middle party and I just collect your data. Now, to be fair, all I'm getting is a unique ID in your step,
22:20
generally your step count. Which, yeah, okay, fine. But I've got your unique ID, which means now I can track you. And if I have a big enough network of devices that can speak Bluetooth low energy, I can follow you everywhere. And yeah, but these are, you know, really great devices.
22:44
I mean, heck, I've got one on me. And one of the conference organizers alerted me to this earlier. I didn't even know about this thing. This is a Wi-Fi enabled pet feeder. So this will monitor when your pet eats, how much it eats, and then it will dispense
23:01
based on various criteria that's built into it. And it's connected to the internet. And you know, think about this for a minute. Hi, oh, you know, I'm out on vacation for a weekend and Fido ate, you know, some food. Great, let's feed him some more. Except what happens when it doesn't feed Fido?
23:21
For like a week. Because you're not home. I'm just saying, I'm not. I'm not advocating that by any stretch of the imagination. I think that's a horrible idea. And of course, from the keynote this morning,
23:41
everybody plays Pokemon Go, right? Everybody's on the blue team, right? What, no, you're not on Pokemon Go. Would you like an invite? No, I guess this isn't ingress like it used to be. But you know, we've now got games that are generating chunks of hardware to help you play the game.
24:02
And I mean, this isn't like a controller. This is just like a watch. That, you know, when you're wandering around and you get near a Pokemon that your phone then tells the watch that you're near a Pokemon and you press the button so you can catch it. That's all it does. It is literally a Bluetooth low energy button and a small vibrating motor.
24:24
This is the internet of things. These are the things we are building right now. And these are the things that are actually, these are obviously all things that are making it to market. Because there are things like Kickstarter and Indiegogo and everybody with half an idea,
24:42
either software or hardware, is putting these things out. And some of these are absolutely amazing. I don't want everybody to think that I'm throwing these up there to mock them. Well, I'm kind of mocking them. But some of these devices are genuinely amazing devices. I mean, the Fitbit has actually gotten more people to go and do exercise by gamifying exercise
25:04
than I have seen in, you know, ever in my entire life. I've never seen a device that's actually been successful at getting people to go and walk a hypothetical 10,000 steps. We can argue about the accuracy of the 10,000 steps later.
25:21
And, you know, there's people who are, you know, taking and, you know, even just the home, I'm gonna completely blank on the word, climate control systems. You know, people are now actually, you know, monitoring and watching what their houses and their buildings do better.
25:42
And they're making changes, either to save energy or to realize, oh, well, I really don't need the air conditioning on today, let's just turn it off. Because all of these devices are, you know, because this is a brilliant plan and I'm glad to be a part of it.
26:01
Really, IoT devices are both brilliant and horrible all at the same time. There's really no way to get around this. I have IoT devices sitting in my house right now. I've got a bunch of IP-based cameras. You know, and IP-based cameras are all made in China. You know, whether they claim they're not is irrelevant. They're all made in China.
26:21
And they all phone home. I've turned off everything for these, every magical checkbox I can find in the firmware. And they're all still trying to phone home. And why do I know this? Because I threw them all on a VLAN, firewalled the VLAN off so that it couldn't connect to the internet. And I watched all the attempted traffic fail.
26:41
And these are brand name cameras. They're all trying to phone home. They're all trying to tell the company that made them something. And I have no idea what it is. They're not supposed to be calling home, and yet they are. I've got a Z-Wave controller, which if you're not familiar with Z-Wave, it's one of the low power home automation
27:01
type control systems these days. That the entire device lives in my house runs on my network. It's great. It's all self-contained except for one small piece. It connects to the internet to do authentication. I can't authenticate to my local device without going to a cloud provider that the manufacturer provides, log into their website,
27:21
at which point it redirects me back to the unit in my house. What could possibly go wrong with that entire scheme? Oh, and by the way, you can't control anything unless you're authenticated. I've been having words with that particular manufacturer
27:40
over that for a while now. They're not, I don't think they like me. But there's one thing here, and most everybody in this room really isn't gonna find this particularly amazingly weird or different or even just odd. It just is. These are the things that are here to help
28:01
us actually solve all these problems because these devices, they're all being manufactured. I mean, how many people, I mean, you've all got cell phones in here. How many of you have Android? How many of you have Marshmallow on your device? So of the people who had their hands up, 80% of them lowered them for the people at home.
28:24
How many of you have, what was the one before Marshmallow? I'm gonna completely, Lollipop, thank you. How many of you have Lollipop on your phone? Not many. KitKat? How many of you have no idea
28:40
what version of Android is on your phone? What? Okay, how many of you have six, which is Marshmallow? How many of you have five X something, which is KitKat? Oh, Lollipop, sorry.
29:02
Version numbers are hard. Really, can somebody go and talk to Ubuntu about their naming scheme? Because I don't understand it anymore. I mean, you start with Wordy Warthog. I mean, I accept the Amish paid to me,
29:20
not that I think that they did, but now we're like on a completely different letter of the alphabet, and I don't, hey. How many of you are on four dot something on Android since I was asking questions about Android? Three? If any of you are on two, please just give me your cell phone. I will give you a new one.
29:44
Are you seriously on a two something? Seriously, a Moto E right now is $29 U.S., and you can run Marshmallow on it.
30:03
So what you're really saying is that you haven't gotten a security update in like four years, four or five years. Do you know how many kernel security vulnerabilities there have been since then? Do you connect to the internet?
30:24
Do you use the browser? How many people think that's a really bad idea? You're all right. No, and I'm ragging on it for no other reason that I wanna use it as a point here,
30:42
is that, you know, devices are now getting made. You know, and all of those devices that I showed, they're all made by companies, and they all have one real big goal. Make the thing, get it out there so that you buy it, and then move on to the next one. Because support's really, really hard in the long run. You know, car companies, cell phone manufacturers,
31:01
nobody really, I mean, even computer manufacturers in the general sense don't want to support things for more than a very small amount of time, because it's all cost to them. And you're buying devices at the lowest possible cost already, which means that their margins just basically don't exist.
31:21
And so their want to support these devices for a long period of time is basically zero. And this presents an obvious problem, because then you end up with cell phones that are running Android 2.something when we're on Android 6.something. I don't even know what the absolute latest is. But the number of things that have changed or been fixed
31:43
or, you know, been found to be a problem in those four intervening versions, that's at least four years, I think it's more like five with two something, is just, it's crazy talk. And yet, you know, there are still people, unfortunately, running around for whatever reason.
32:03
I have no idea what your reason for it is, other than it's probably a really nice phone and it probably has like a real keyboard. I, yeah. Yeah, the real key, oh, man, I miss real keyboards on cell phones. Yeah, my typing has not gotten better without the keyboard, tempting.
32:24
Except then I'd be running a cell phone with 2.something. But here's one of the biggest things that's going to help all of us. And this is not a surprise to any of us in this room. I mean, you're at an open source convention or a conference.
32:40
We all get this. Open source, if everything was open source, we could at least go and fix it. Because at some point, we all want to just go and fix things. We want to make it better, we want to change things. But open source only works so much if you've got access to the underlying hardware. I mean, we've all got cell phones. How many of you can actually get even root on your cell phone that you believe you can get?
33:05
That's a much smaller handful of the people who said that they had Android phones, or any phone in general. I mean, it makes it really hard to get access to these things. Because, you know, again, the companies don't want to open these protocols. They don't want to open up their hardware. Except that that's one of the things that can save us
33:23
from the impending mess that we're all looking at. Open source hardware. How many of you have even heard of open source hardware? Oh, good. Good, I like all of you people. I don't have to explain this. Open source hardware is just one of those things that it makes everything easier. It gives you a really great starting point
33:40
to build your own devices from. And honestly, you know, once a company stops wanting to support something, why wouldn't they want to make it open source hardware? Because then, you know, people are going to go off, they're going to use it, they're going to maintain it. Now, admittedly, that also means that you're not going to buy the latest and greatest widget that they put out. So, okay.
34:01
And there's a couple of projects here that use open source software and hardware. Specifically the Minnow Board Foundation, or the Minnow Board Foundation, which I happen to be a member of. And BeagleBoard, which is another open hardware platform. These are devices that are out there that are, you know, people are taking and using in all kinds of different ways. And you've got companies
34:21
that are making open source hardware, and using open source software, and putting everything out there so that people can build stuff. Adafruit and Seed and Sparkfun. And in fact, the device that's up there, the red PCB, is actually a small microcontroller called the ESP8266. Want to guess what it's really great for? Internet of things!
34:42
It's got a Wi-Fi chip built into it. Honest to God, it's got a full Wi-Fi stack, including access point mode. That entire board, at least in US dollars, was $14.99 when I looked at it yesterday. $14.99 for an entire,
35:01
just the module itself, not in that specific breakout board. You can get the chip that runs all of that for $5. I mean, when, I'm sorry? And even less if you go with the, this might not be an official ESP8266 from Alibaba.
35:20
That's true, they do all come from China. And I'm, yeah, I'm not gonna get into that. I'm not gonna get into that. I'm sorry? Well, I mean, this gets back to, yeah, I was gonna say, this gets back to the pregnancy test device. They doubled the cost of the pregnancy test from $10 to $20 to add the extra functionality.
35:43
But this is why they can do that, is the hardware is so bloody cheap. We are literally living in a golden age of computational performance. Yeah, it only does IPv4. Nobody cares about that IPv6 thing except, you know, everybody that should care about it.
36:04
But yeah, I mean, we are literally living in a golden age of computational performance and cost. I mean, if you go back to when computers first exist, you know, were first created the ENIAC, this is a computer that would have taken up probably more than this room in floor space.
36:22
And the amount of power and cost to run it was just astronomical. And nowadays we have more computational power in my watch. You know, it's just a, you know, little Android smartwatch. There's more computational power in that than existed in the ENIAC. In fact, there's probably more computational power
36:42
in my Fitbit than existed in the ENIAC. And these are devices that we're buying for a pittance for their computational capacity. I mean, the ESP8266 is, it's probably faster than the 386, you know, a lot of us grew up programming on.
37:00
And yet I can buy that for as little as $5. And I can stick it into pregnancy tests and literally throw it away because it is now so cheap. I mean, think about that for a minute. Computational power has gotten so cheap, we can literally use it once and throw it away.
37:21
I mean, how is this not defined as a golden age for everything we're doing? And one of the most frightening things I could ever say is we are, you know, we are at the point where computational capacity is that inexpensive.
37:42
And here's why all of those things actually help. No, this doesn't work. Please don't try this at home. I believe you'll actually short something really badly. And for those of you too young to realize what this all is, that's a parallel port, connected to,
38:01
or a parallel port, no modem adapter that's connected to a serial, a parallel to serial port adapter that's connected to a serial port to PS2 adapter that's connected to a PS2 to USB adapter that's connected to a USB flash drive. No, this really doesn't work. I'm not even sure that power flows properly through that.
38:21
And I think the power that comes out of a parallel port is 12 volt. So that's really not gonna end well for the flash drive. But you know, and the reason open source software and hardware is really gonna change the, is not only changing the world as it is, because a lot of the devices that, you know, that are in the internet of things are coming out of,
38:43
or at least vaguely related to the open source hardware that exists, is because people are taking all of these devices and they're doing things differently than what the manufacturer actually intended. I mean, how many people, honest to goodness, believe that that Bluetooth low energy device
39:01
that's in the pregnancy test was actually intended for a pregnancy test device? How many people would have even thought of that? I mean, it's absolutely brilliant. You know, let's take a Bluetooth low energy device and this, and let's, yes. This is great, then we'll give them coupons.
39:23
Or the connected fridge or the connected oven. I mean, how many of you have backed something on Kickstarter? Oh, come on, don't lie. Just because it didn't show up at your house doesn't mean that you didn't back it.
39:43
Okay, fine. But I mean, there's a lot of the, I mean, people are just coming up with ideas and these have all existed before. But I mean, things have gotten so cheap and so easy to do either because open hardware now exists and is making this more approachable or making the hardware more approachable
40:00
to software people or the hardware, or from the hardware perspective, the software is now more approachable because we've got open source software that's now pervasive and everywhere. But these are things that, you know, everybody's got, everybody's using. And they're going to take every piece of either software or hardware and they're going to use it in ways
40:20
that we have never even imagined. How many people, I mean, we're all software people here, right? No hardware people in the audience, right? It's really, gosh, I'm actually a software person that's turned into a hardware person, so it's kind of weird to be talking at software conferences now. Because I'm like, oh, I want to really talk to you about this weird transistor that I found.
40:41
And then everybody's like, what's a transistor? Where's my for loop? But how many people have seen their code used in ways that they would never have even thought of? One, no, that can't be it.
41:01
Do you guys all like work on really boring things? I'm hearing a lot of yeses. I'm sorry, there's some really neat, I've got some really neat storage stuff that I'm working on if you want to come help. It's at least probably, well, actually it's storage, which means by definition it's boring.
41:23
But I mean, go find, I mean, even software is being used in ways that nobody even remotely thought of. I mean, some of the hardware boards that I've been making lately, it takes me 12 different programs to get an image to it actually showing up on a piece of hardware.
41:42
Just like even an outline for a PCB. It takes me 12 programs. And I'm sure that 10 of those programs, actually probably 11 of those, never thought that they'd be used in PCB design. I mean, Inkscape, GIMP, ImageMagick.
42:03
How many of these sound like PCB design tools? Oh, okay, somebody's done hardware design. Okay. But you know, everything that we do and everything that we touch these days is going to be used in ways
42:20
that we're never going to understand or that we're never going to predict. Which leads to two obvious statements. The first is stop believing that you know what your users are going to do with your software and start listening to them. Because they're going to go and do things that make no sense to you. And they're going to ask for features that seem absolutely ludicrous. But it may just be that your tool is less bad
42:42
than everything else they can use. And that's why they're trying to solve their own problems with it. And two, just you know, be open to the possibility that you know, oh gosh darn it. And again, the magic of Linux and presentations,
43:03
we'll see if that, oh, that did come back. Because if things are open, we at least have the power to fix things. Because I'm sure nobody in this room has never run into a bug on a piece of software that they're using and immediately gone, screamed bloody murder, written a patch and submitted it.
43:22
No one, right? No one? Good. Because I'm sure I've received several of those at some point. But if, and if we're doing this in the open and we're doing this in hardware and we're doing this in software, we actually have the chance to fix things. Which means that all of those weird comments I had about IOT devices earlier,
43:42
even if they're not fixable when the device comes out, if we have the opportunity to go and fix them later, this means that we as an entire society are safer and better off. And again, everything is going to get used in ways that you've never even remotely thought of.
44:01
This is a conference badge, an electronic conference badge that was from a security conference two years ago. At the conference, someone turned it into a quadcopter. And honest to God, it flew. There's video on YouTube of it flying.
44:20
And I can't, and I honestly, while I was doing this presentation, I could not think of a better example of people are going to use things in ways that you had never even remotely thought of. This is a conference badge that has a bunch of blinking LEDs and some touch pad areas and some through hole pins that you can attach more things to it. And what did they do? They made a quadcopter and they flew it.
44:42
But I wanted people to start thinking about not what they're trying to do, but what their devices or their software could be used for. You know, yeah, don't ignore your major focus, but accept that the world is changing, it will always be changing,
45:01
and people are going to do things in ways that you could have never expected. And I'm running slightly fast, so hopefully you guys have a lot of questions, and not about the flaming octopus with a replica of K-9 sitting next to it. Well, you can ask questions about it too, I will happily answer them, because I built the K-9.
45:21
I did not build the giant flaming octopus. So, questions, comments? That photograph was taken at the San Francisco Maker Faire, I believe three years ago, two years ago, two and a half years ago.
45:40
Where do you get, I'm not gonna answer him. K-9, you can build the octopus, you have to go to San Francisco.
46:03
Yep, okay, so the question is, I'm gonna repeat that, how does open hardware save us from the lack of security that may be coming out with existing devices? So, depending on how this all works, it can't, because a lot of companies are not going to open source their designs,
46:20
they're not going to give you the ability to do anything with their designs. And this is kind of why I pointed out the WRT54GL, is because Linksys never expected anybody to go and start modifying the firmware on the device, and yet people like us in this room were dedicated enough to go, there's Linux in this thing,
46:40
I want to be able to control it, I need to be able to change it. And they went and they opened the device and they started figuring this out. Open hardware makes this easier, because at least at some point you can publish all of this, and you can save everybody in this room an absolute ton of time and effort reverse engineering what's already been made. And I'm not saying that when a product comes out
47:00
it has to be immediately open source hardware. Maybe you run on a cycle that every two or three years as a device literally falls out of favor, out of use, you open source the design. Because then at that point, the poor gentleman who's got the Android 2. something phone, he could at least, if he's got open specs,
47:21
open specs to every piece of that hardware, he could actually try and get Marshmallow running on the device. But in all likelihood, because the device has been closed, the manufacturer doesn't want to support it anymore, and probably half the people who worked on that particular platform are gone from the company. There's no mind share left
47:40
to even remotely go and work on that device. So while I'm making an advocacy play here for open hardware because it gives everybody a chance, maybe not necessarily right at the beginning, but eventually to be able to go in and modify and change things, that that's beneficial. And if a company starts from a point of open hardware,
48:03
this actually encourages them to come back and give it back to the open hardware world. I mean, there are companies that are taking, and I can speak to this, that are taking the middle board and they're putting it into products right now. And they're changing things about the board, but they're also giving those changes back.
48:21
I mean, they're kind of required to because of our licensing, because we're CC by SA. So that magical SA bit means that they have to give it back, kind of like the GPL. But these are companies that are still making money on product. They're manufacturing things, and yet just because anybody can take this and make it themselves,
48:41
these companies are still making money, they're still seeing a profit there. So this kind of starts putting a nail in the coffin of, oh, well, if we open this up, we'll never make money. So I know that's kind of a roundabout way to answer your question, but it's, okay.
49:15
So I'm gonna paraphrase this slightly. What is my understanding of open hardware with respect to the chips themselves
49:22
and whether the chips, or whether I would consider chips that may have open data sheets that you can go and look at to be open or not? Is that, that's kind of the direction, because you're talking about, is there even the point of, is it better to go down the FPGA route where you may control more of the hardware,
49:42
or at least the hardware design? So I actually follow the Open Source Hardware Association's model pretty closely for my personal belief. And generally speaking, I see open hardware as something that I can go and I can modify and I can change on my own. Which is kind of what the GPL says about software anyway.
50:05
But that doesn't necessarily mean that things like the atom chip that's on the minnow board. Yes, most of the specs for that particular chip are open, they're known, but things like, frankly, the microcode is not. Does that make it, but does that make the Intel chip that's used on it
50:24
not suitable for open hardware? That ends up in a mostly religious argument over whether you believe that any closed piece means that you can't, or closes the entire design or not. And I would argue it does not,
50:40
because frankly, building the, I've got an open hardware design, and I can pick and choose what pieces I put on there. That also means that I have the ability to choose what pieces may or may not work for me. So let's say that a transistor doesn't have an open enough data sheet for my purposes. I can at least, I at least have the opportunity or the possibility of going and finding another one
51:01
that may have a more open data sheet. Now, admittedly, it's a transistors. There's three pieces to it. It's kind of hard to not understand what it is, but you can scale this to any piece of the design. Does that mean that you're going to be able to find everything that would work for you? Probably, maybe, maybe not.
51:21
It's a tough call. Oh, no. No, I would absolutely include things like FPGAs in the open hardware movement. You know, I mean, they're just another piece that may or may not be on your design.
51:42
I mean, you could replace the entire, you know, CPU on a lot of the, you know, the open hardware boards now with an FPGA, and then you control the entire, I'm gonna put in quotes, the SOC itself. But I'm not saying that the, I'm not saying that an FPGA is exclusive to open hardware,
52:02
and I'm not saying that the pre-baked ICs are exclusive. I mean, everything is in that space. Well, I think asking for IoT devices that don't go home is like asking for newspapers without commercials. I agree.
52:28
And again, and this kind of comes back to some of my commentary about the normal user base. You know, while I would like my devices really desperately not to call home, there's a lot of people who, I mean, I have multiple VLANs in my home network.
52:43
I'm really weird though, and I accept that I'm, like, normal people are over there, and I'm somewhere in the next county over in comparison to how my network is set up. And I agree. You know, I would like the option to really tell these companies I really don't want to call home.
53:01
Really, I'm going to go live in my own little microcosm. But that's, I am the outlier in this case. I am not the normal kid. I am not the normal situation. But it would be awfully nice if I had more direct access over the software stack or even the hardware stack that exists in these devices that I could go and disable that, modify it,
53:20
or just write my own IP stack for my cameras, for instance. Personally, I would pay substantially more. But as far as I can tell, in a lot of cases, these options don't even exist. So at this point, my choice is
53:42
I can't get the feature I actually want, which means I'm going to go and pay the cheapest amount that I can to accomplish my goals, and then take a giant ban hammer at my firewall and turn off their ability to go out to the internet. And yes, that screws up some functionality.
54:02
Lights, oh, yes, I absolutely missed. I don't know how I didn't put an internet-connected light bulb up there. Um, and yes, I completely forgot to add those. If you really want to see some of the most interesting hacks in the world,
54:21
go look up hacks for the internet-connected light bulbs. Those things are some of the most redonkulously badly-coded devices in the world. But they're really cool. But you could have blinking lights nationwide.
54:41
International-wide. Worldwide. Sadly, I think that there may be enough infected light bulbs out there that that could actually be done at this point. Anyone else?
55:00
Questions, comments? I mean, I'll be around, so if you've got anything more specific you want to chat with me about, by all means. Otherwise, I will give you back the, I believe like three minutes, or five minutes of your day.