Pragmatic 9: The Internet Makes It Even Better

19 January, 2014


John and Ben discuss the practical risks associated with connecting every single device we own to the internet. What happens when ore functionality is tied to a remote server and the company changes or closes? Timely in light of Nest Labs acquisition.

Transcript available
[MUSIC] >> This is Pragmatic, a weekly discussion show contemplating the practical application of technology. Exploring the real-world trade-offs, we look at how great ideas are transformed into products and services that can change our lives. Nothing is as simple as it seems. I'm Ben Alexander and my co-host is John Gigi. How you doing, John? - I'm doing very well. And how you doing, Ben? - I'm doing great. - Awesome. Well, just gonna open up as I do most of the episodes with some thank yous. Some quick thank yous for Twitter shout outs to some people that said some nice things about the show. Jim Biancolo, Dave Ligart, I think it is. Mayank Yadev, Hannes Egler, I think that is. And also to Adam Wright, who's apparently binge listened to episodes one to six in the space of one week. Not sure if that's wise, but he's still enjoying it. So that's good. Adam, drink lots of water. It'll help you get through. Yeah, that's intense. Anyway, OK. Glad he's enjoying it. So, sorry if I've missed anyone. It's been a busy week because I've actually started a new job this week with a different company. I've been flat out, but I think that's everyone on the list. If I've missed you, apologies. I always say thank you back on Twitter as well. Thank you for those shout outs. Also special thank yous for articles written about the show. Thank you to David Hutcherson for writing a nice article on his site about the show. Thank you very much for that. And also to Derek Martin, he referenced the show in a 1080p article. There's links for both of those in the show notes if you're curious. We've also had four amazing iTunes reviews in the last four days, and they're all from a different country. So we have N... I don't even... These are just usernames, but NJR0 from the UK, Rush Ring Leader from the US, which is kind of cool. Mars C-O-K-L from Canada and Shoggy92 from Switzerland. - Sounds like a radio station. - It does a little bit, yeah. In any case, thank you so much for the iTunes reviews, especially really appreciate that. And also just stoked that everyone is enjoying the show so much, so thank you again. We also had a poll at the end of last episode and we've had quite, I think quite a reasonable response. I'd like to, first of all, upfront thank everybody who took the time to participate in the poll. Really do appreciate it. We were trying to gauge, based on some of the feedback we had early on about the usage of Imperial and Metric in the show, based on the response, as far as I can see, there is certainly enough interest in mentioning both sets of units for us to continue our conversations in both Imperial and Metric. So, the problem is, I think that in the end, we sort of think in our own measurement system that we had mostly where we grew up, what we grew up in. And you tend to think in those measurements. And it's kind of like learning other languages. You've mastered a language, I think, when you start to think in that language and you look at something and you know straight away what the word is in that other language. And I think it's much the same with measurement. and I just want to make sure that everyone is covered. So thank you very much for the input on that one. That was it was interesting. And I think it's sort of highlighted to me, at least, that that there's a great there's a diversity of listeners in the show that I want to make sure I account for and include everybody. So the next question that we had was about doing a live show. And we had, I think, enough interest that we should do a live show as a trial. And we're going to do that next week for episode 10. We'll be doing that one live and the details for that will come out during this week from Ben. Yeah. And you'll see that through Twitter and so on. Ben, did you want to talk briefly about that? Yeah, I was just going to say, well, you know, we'll tweet before we do it and we'll put a link up on the site. You can get over to it and yeah, it's pretty easy. Excellent. Okay. So, for those people that are keen, keep an eye on the site, keep an eye on the Twitter feeds and you'll see that during the week. Okay. So, again, that was the poll. So, thank you very much for everyone for participating. Really appreciate it so much. And that's the poll. One more thing before we get stuck into the topic for this week. And that is we have made a decision to restructure the shows. Some of you will have already noticed this, others may not. The episodes are still numbered sequentially, but will from 1, 2, 3, 4 and so on, and they'll be associated with a specific topic. So one episode will be one topic, whatever that may be. And there may be a few little side strands in there as we discuss and delve into different topics, but the idea is that if we're going to talk about electricity and power generation, that'll be one topic topic like the battery problem for example. If we're going to talk about iPhone hardware that'll be one episode, one topic. And whenever there's follow-up to any episode at all it will be numbered along with the episode that it's relating to. So if there's follow-up to episode 2, the battery problem, it'll be listed as episode 2a. If there's additional follow-up beyond that it'll be then 2b, 2C, 2D and so on. The idea is that if you're interested in, if you've listened to the Battery Problem and you're interested in follow-up on that, then you can then choose to download those follow-up episodes and listen to them. If you're not interested, then there's no need to download them, you don't have to listen to them if you don't want to, because I appreciate that some people may not be interested in, you know, for example, in Episode 8 we talked about noise and hearing issues, well maybe some people aren't interested in that, so if there's follow-up to episode 8, they may see 8a come up in future and decide that they're not interested in listening to that and that's fine, that's no problem, not every topic will appeal to everybody and that's okay. They're also indexed by date, so in other words they'll come out with date as the feed progresses, so that gives you the ability to understand at what point in time that follow-up was actually recorded, so you'll know that based on the date in the feed that it'll all be released around about about the same time as the main episode for that week. So what Ben and I are doing is we'll record the follow-up segment on the same time that we're doing the main topic, but they'll be released separately. So you can tell that 2C was released three or four weeks after the actual topic aired. There's some discussion we thought about whether or not we would put them in a separate feed or we'd have them in the same feed. And we decided to go with the same feed simply because it would get, I think it would be too much trouble to ask someone to have to like to follow two feeds. So, you know, if you want follow-up, which is still part of the show, then, you know, you would then have to follow a follow-up feed as opposed to the main topic feed. But at this point in time, obviously this is an experiment. We haven't done anything like this before. We just want to try something different and we'll reevaluate in a few episodes time, see how it's working for people. And if there is enough demand to split it up, then we might do that. But the idea is we try and keep episodes to about 60 minutes, maybe 75 minutes in length. And most follow-up topics are, you know, on the order of 20 to 25 minutes. And that gives you, the listener, the opportunity to choose, pick and choose that which you want to listen to. If you want to listen to the lot, download the lot and listen to the lot, that's fine. You can listen to it just like you would a show that is cut more in a more standard way, whereby you do follow-up and then you do a main topic. And then sometimes they have people like to call it an after dark or after episode. We just wanted to get it more granular because, yeah, this isn't necessarily like an easy listening show. You kind of need to pay attention. And, you know, we get that. If we put everything out, is it like a two hour episode could be that's a little bit of an ask. So. And some of the early episodes were two hours because there's a lot of information in them, and sometimes that may still happen, but in the end, by separating out the follow-up, it gives you the listener the ability to just pick and choose what you want to listen to. And that's the flexibility we wanted to put that power and control in your hands so that you can choose. Anyhow, that's what we're doing. And we're obviously very, very keen to hear your feedback on whether you think that works or whether you think it doesn't. Happy to really would like to hear your feedback on that. So it is a little bit of extra work for Ben and I appreciate the extra work that he's putting in to do that. So we'll see how it goes. And without further ado, I think that's enough on that. Mike, into today's topic. - What do we got? - There's an incident recently with a little company called Nest. And I don't want to talk about thermostats exactly. I want to talk about what's a growing trend and some of the concerns. and I think people need to adjust expectations for. And that is the whole idea that the internet makes it even better. Like you've got a product, whatever it might be, you know, a light switch, power switch, a light bulb, you know, thermostat and so on. And the internet will make it better because it's connected and you can do really cool things with connected things. So there will come a time where everything has a connectivity thing and there'll be Wi-Fi, it'll be low energy Bluetooth, whatever it might be. And then there'll be a server somewhere where that data can be aggregated, linked with other data, and then you can do cool stuff with it, automated stuff with it. You can sort of take that to the next level, whatever that may be. And that all sounds lovely and futuristic and wonderful, but there are some big problems with it. And one of the things that worries me when I see things like Nest, and they're not the only example, is that what happens when the internet piece of it goes away for some reason? So I wanted to talk about that. So when you've got a standalone device, if it's a software-based device, if there's some kind of software, whether it's firmware and it's hard programmed into it or whether or not it's it's got some level of software that you can use for interacting with it. If that's independent, that means when I say independent, I mean, it can function without connection to the Internet, it'll work perfectly fine without a network connected. It doesn't require programming. You know, the old fashioned idea of you want to update the firmware on a printer, let's say, you know, you had to have a USB disk or not a USB. I mean, back when I remember, there were like special memory card you would have to put into these things. They weren't USB. They were, you know, it might be a compact flash card, let's say. And it would have to be formatted in the right formatter. And that would have to have a boot ROM loaded onto it. You'd go up to the copier and you'd plug this thing in. You'd put it into a special maintenance mode and that would update the software, right? Like the serial port on the back of my TV. Yeah, that sort of thing. Exactly. And there were all these antiquated ways that they used to do it. Of course, before then, there was no software update at all. And it was just, well, this is what your machine's got. And if you want a different EEPROM on it, we're going to physically extract the EEPROM you've got and put a new EEPROM in with a pin extractor and like a PLCC extractor. And there you go. Right. That was it. The whole idea of updating firmware on devices through different, through soft means has been a revolution in hardware. Well, hang on, I shouldn't say hardware. Any device that has hardware and software in it. It's been a massive, massive leap in the last 15 to 20 years as the network, as the Internet has spread further and further and more and more people are getting Wi-Fi in their houses with ADSL, with cable modems or with Fios, the more we do, the more connected we get, the lower that barrier becomes to the point at which now we have operating systems whereby they simply call back to a server somewhere. Oh, there's a software update available, updates it in the background. Oh, by the way, I just updated your software. Click restart and you're done. Right. The convenience of it is very, very attractive and very addictive in many ways. There's all sorts of advantages to it. But it also worries me because if the internet goes away, you still have to have a product that's going to work. You can't- If you put yourself in a position whereby you have a perfectly independent device, if the internet disappears or maybe you won't get firmware updates, okay, that's okay, probably. You know, but if it means you can't access certain features of it that you bought this device for, then that's a bit of a problem. Because if that's what drew you in. So let's take an example of a light switch. Any old run-of-the-mill light switch, not automated, just a plain old simple light switch. I buy a light switch. I have an expectation that that thing is going to work for maybe 20, 30 years, maybe more than that. I've seen light switches in houses last 50, 60 years. Old Bakelite ones, for example. They're still going. They're chronically faded and look a bit strange and rustic, but they're still working. So if you buy an internet connected device that has remote access features, the question is, should you expect the same usability in 20 or 30 years time? So as some sort of evidence to build my case, I'm going to refer in particular to one, which I think is a case in point to an extent. It's not a home automation device because let's face it, you know, internet connected home automation devices are very new entrance. And when I say new, I mean the last two to three years, like really new. It's something that's become, it's starting to accelerate. We've talked about this previously on the app. So we did about automation, but the point is that there was a device called the Peak. And I'm not sure if you remember this thing. It was never released in Australia, but I was aware of it from following what was going on in North America at the time, coverage on the different blogs and so on. And it was introduced in 2008. Right. I remember this. Yeah. Remember this? And it's basically, the concept was data plan. Essentially, you pay for it up front, and it was a lifetime cost. Right. So it was extremely, well, cheap, let's be honest, cheap hardware. Yeah. And they had special deals. They, the company, Peak had special deals with, I think it was T-Mobile and a few others. And they would provide connectivity and Peak would provide the servers that would do all the aggregation and compression and all that sort of stuff. Kind of like the way BlackBerry works a little bit. But the idea is that with BlackBerry, you can buy a BlackBerry server and you can put that on your premises in your location and that will do all the data compression and all the other BS with email features and interaction with exchange. So it sends a very minimal amount of data out to the devices and that was the whole foundation. These were designed for 2G, right? I mean this was the, yes, that was the big limitation. So yeah, that's right. So the idea though is that you still had a centralized server. Well Peak had the same kind of architecture based to my understanding. And if I'm incorrect, then please correct me. But I believe that in the end, they were in charge of their servers. Their servers would pull all sorts of different data and so on, present that for users. The genius cloud. Yeah, look, I'm not. OK, so let's just... You're getting ahead of me a little bit there. Such a... Yeah, I don't know about that name. Anyway, now, their devices did not sell well. They just didn't. And the reasons that they didn't have been hotly debated. But frankly, when you've got something that looks like a pager with a keyboard on it, and they did, they looked a bit like maybe a fat credit card with little keys on it. Yeah, it's small. Yeah. It's like, well, I've got this iPhone. They launched in 2008. Yeah. And they launched in 2008. That's the problem. Yeah. And I mean, had this thing been launched 10 years earlier, it would have been a completely different story, I'm sure. Whether or not they still were around, I don't know. But now I'm getting ahead of myself. So in 2012, they essentially cut off their service. They discontinued all support for all the peak hardware that was out there. They said, look, we're done. We're going to move to this thing we're calling Genius Cloud. And, you know, we've got money and we don't really care. and hardly anyone's using it anyway, so shrug, whatever. Very dismissive, very, how should I say, little unprofessional, maybe. Certainly not particularly concerned about their users and their users' well-being. And there are people that love these things. You just got to read through some of the forums, let's flip through some of the forums of these things. And there are some people that really loved them. And there's also people that love Blackberries, but, you know, shrug. But the point is, having never used one, I couldn't comment. but when you bought it, the idea was that you had lifetime use of it. Now, where the peak is different, I mean, I realize, let's be honest, that mobile devices are generally considered disposable. You know, if they have a much, much smaller lifespan, I was saying, think of a light switch early on, right? Right. Well, a light switch sits in the wall. It doesn't move. It's permanently attached, doesn't get dropped. You know, hopefully you don't hit it with a sledgehammer or any other blunt instrument. You know, you're not headbutting it, you know, barring serious catastrophe, it's not going to have any issues. Whereas a mobile phone can be dropped. It can get wet, go for a swim in the toilet. I don't know, everything goes for a swim in a toilet. Anyway, point is, you could drop orange juice on it, like happened to my wife's 3GS. There's all sorts of things that can happen to it because it's physically disconnected, you know, so therefore, it's going to get beaten up and trashed. And hence, it's going to have a shorter lifespan. You take a light switch out of the wall, carry it in your pocket and see how long it lasts. Guarantee you it won't be 20 years. Mind you, I don't know why you do that. So, okay, I'm out in the park, I feel like turning a light switch off. Anyway, I'm sorry, that's just silly. Point is, what was my point? Yes, of course. So, the point is that although the Peak was a mobile device and had a shorter life expectation. There were plenty of peaks out there that were still working just fine when they pulled the plug. And there's an expectation when they say lifetime, and that was part of the whole why there were so many people outraged about it, is that they put money down up front, no ongoing cost. It was like, there you go, it's lifetime for this device and you've got it. Because this thing apparently like it's sipped data, it was all highly compressed or whatever. but their business plan required servers to support them. And the server support was based on the concept of ongoing revenue, but there was no ongoing revenue or rather their hardware wasn't selling enough to create that. So, when I was thinking about this, I thought a little bit about Apple. When you think about it, Apple really doesn't, the majority of its business model is based on selling hardware hardware. And now that we've got internet connectivity and connected devices and the quote, unquote, the cloud, you know, now that you've got this, all that takes servers, which takes, you know, electricity to run, air conditioning, technicians, things break, you know, you put a server in, it's only good for five years, let's say, then you've got to upgrade it. Ongoing costs. And how is that funded? The sale of new hardware. What does that rely on? And that relies on good ongoing hardware sales. So the way Apple mitigate their risk is they have different product lines. So they've got iPods, iPhones, iPads, Macs, the occasional display, not a retina desktop display yet, for example, but they have all these different products. And the point is the majority of the stuff that they do, the vast majority is funded by the sale of their hardware. So where Peak failed is their hardware simply didn't sell enough. There weren't enough people that were interested in it, product was too late to market, and they simply could not continue to fund or chose not to continue to fund what was essentially a product that was pure loss. Like it was just generating loss for them. So they pull the pill on the servers and then they focus on cloud stuff And God knows what the hell that is. And it's out of the consumer space now. So- It was $300. Yeah. So it was- I mean, it was not cheap, cheap. No, but it was lifetime, right? Right. And that's the thing. What's that mean? Whose lifetime? Yeah, yeah, exactly. Pick a short-lived animal, the lifetime of that animal. I don't know. It's disingenuous and it's a bit scummy, but look, whatever. The point is that the peak fell apart for the fact that their business model was not sustained by subscription. They did not have ongoing income or revenue and that you can get away with that if you've got enough scale in your hardware and your hardware is essentially wearing out in such a rate and is good enough that people come back and buy the next generation every year, two years, three years, four years, like Apple has. But if you don't have that, the model doesn't work. Okay. So now let's talk a little bit about Nest. So that was found in 2010 and recently it was sold to Google for quite a chunk of change. Certainly wouldn't mind seeing those figures in my bank account, but never mind, that won't happen. So their model was to sell a piece of hardware, well, starting out with a thermostat and later on they did a smoke detector, I think it was. And again, same kind of idea. It's like a one-off purchase, updates free over the air and everything. It's got Wi-Fi connectivity and air connectivity. You could download an app for your iPhone and you can access settings on the device from outside the home. You could modify the temperature and change bits and bobs and so on, or remotely. And that was, you know, wonderful and cool. But the thing that worries me is that, and this is sort of just a technical aside, if you've got an ADSL modem, you've no doubt you've got a firewall with essentially with where you get one IP address allocated to you on the public facing side onto the internet wireless network, internet WAN, and then everything within your house is on a subnet. and it's all routed through your router. But what that means is that there's no direct path to get to a computer. So if I put a request out to the net, I go, I want to go to site blah. Well, that request starts at my computer. It goes to the router, the router then figures out where it's going, but it contains a trace back such that when you get the data it can go to the public IP address and then back to your computer again. So the trick, because then that means that no one can find your computer, all they can do is find your router. And that's the whole idea of a firewall. Well, you know, anyway, that's one way of doing it. The point is that in order to overcome that, well, like for example, I use LogMeIn on one of my Macs. And it sits behind a firewall. How does that work? Well, you install the software on the Mac and it will go and register with LogMeIn. and it tells LogMeIn the IP address that it's sitting behind on the path to get to the machine. And that runs in the background all the time. And all these devices do something very similar, they report back to a server and that server keeps track of what is where. So when you open up your iOS device or your Android device and you want to control your Nest from out and about, that server, sorry, that application has to talk to the server to get the path to your thermostat. So without the server component you've got nothing. I'm assuming of course you're not a geek. If you're a geek and it's all... you could set up port forwarding in your router, you could get a fixed IP address which costs more, or you could use dynamic DNS there's a whole bunch of different solutions, but you could get into your system if you set it up correctly. Whether or not the Nest in particular specifically would support that, I couldn't say. But the point I'm trying to make is that in these market, in the market today, people don't know how to do that. They don't want to do that. Yeah, unless you're a geek, you're not going to know how to do that. So what do you do? You have a centralized server to go back to so that you can coordinate the control of these things. It's not rocket science. It's just, you know, it's just, it's the price of doing business for simplicity. - Right. And that's okay until the server isn't there anymore. Now, I'm saying, okay, the Nest is bought by Google. I'm not writing their Nest's obituary. I'm just saying, and there's no reason to suspect, well, actually, maybe there are reasons to suspect. There is no evidence to suggest that Google are going to just shut it down at any point. However, if you look at Google's history, they do that quite a bit. They shut down web services all the time, but the Nest isn't a web service exactly. It's a hardware device that has additional features added by being internet connected, that remote control ability, the ability to set its, to adjust its learned behaviours and so on. This is coming from someone, I'll just admit, put my head up right now. I haven't used the Nest. I don't have a Nest. And I don't think you can get one yet. Oh, I probably could, but I have no use for one because I have one air conditioner in the house. And that's it. There's no heating. It's not central heating. And most houses in Australia don't have central heating. We don't have boilers and heaters and all that stuff. Most houses don't have gas to them. You know, you're on gas. And where I am, there's no pipe gas anywhere. It's all done through gas cylinders. It's just not prevalent. We don't need it because we're not cold enough. If we're a colder climate, that's a different story. We're not the key demographic. We're not the market that they're after. And that's fine. Most houses don't have a thermostat in them. They just don't. You've got an air conditioner. You set the temperature on the AC unit, and that's it. That's the amount of control you get. Mind you, a nest might be able to do something with that, potentially, I guess. maybe, but I didn't look into that. And it's sort of beyond the scope of this discussion anyway. The point of what I'm getting at is that if you buy a thermostat and any old run-of-the-mill thermostat, you set a temperature, you walk away. It's not intelligent, doesn't learn. There are thermostats out there, I believe, correct me if I'm wrong, that can do a little bit of learning. But what's interesting is that all the extra features that the Nest thing has got because of the remote access ability. And that remote access is what draws a lot of people to getting that, the Nest. The fact that you can monitor it remotely, you can adjust it remotely, you can tweak the settings from wherever you like. Now that's appealing. So what happens if that goes away? Well, you've got an expensive learning thermostat. Now, if I've got any of the details of that wrong, please correct me. That's fine, because I haven't used one. how I'm going based on everything that I've read and seen, I've looked at the videos and demos and so on and in many respects the specifics of it is not the point. The point is if you buy the Nest thermostat with the intention of using any of the internet connected features at all and suddenly they are removed, well we're back to the light switch problem. You're not going to get the sort of connectivity lifespan out of this thing for the feature set that you bought it for in the amount of time you expected. So in 10 years you may lose it or five years you may lose it. Thermostats still work fine for another 10, 15, 20 years but the main reason you bought it is gone after only 5 or 10 let's say. And I think that's pretty bad. You're having features subtracted from your product. So a little bit more just quickly about the nest we'll move on and that is more related to the fact that Honeywell sued them. This is a point of note. They also developed their own competing product. I think it was released in 2012 and yeah, it's apart from the fact that it's ugly as hell. It's hell. It's, yeah. Yeah, Honeywell. I actually do cross paths with Honeywell gear from time to time in my line of work. But, yeah. Not thermostats generally. Years ago I had to build a I made it in Flash. I made a software recreation of a Honeywell thermostat UI, some temp job I had and wow. It was just insane. Like, it was, gosh, I wonder if it's still out there. I have to find, I'll have to see if I can find the old folder because it was, the idea of anyone actually working their way through to use this thing just boggled my mind while I was making it. Like, just who is going to do this? We always had this, we had the old Honeywell little dial ones that were just simple, but I found it on the page right here, but yeah. - Well, you know, the whole saying, someone needs to build a better mousetrap or do they need to build a better mousetrap? I mean, mousetraps we got work pretty good. You know, it's like, I don't know. I don't want to get all pessimistic about it. However, you do sometimes have to ask the question, is it a better result? I think that in the end, it probably is in this particular case, how useful that is to your energy savings and flexibility longer term is gonna vary from individual to individual. And that's not a judgment on this at all. This is simply an illustration. So next illustration, we'll move on enough about Nest. And that is, if this, then that. And this is something, or IFTT. - Right. - Now we talked about this really briefly and we may end up doing more follow-up on this on the automation episode we did. And for those that don't know what it is, it's essentially a service that on the internet allows you to connect together different events that happen on different websites and now also home automation equipment such that, for example, you could track your geo location and if this, then that could be set up such that when you're within next number of miles or kilometers from home, that it automatically turns on your air conditioning system or turns on the TV set or turns on the kettle or whatever the hell, right? That sort of thing, it has that functionality. It's actually much more complicated than that. There's lots of things you can do. Auto posting to Twitter and I think a few other different things. It's quite a fun little thing to play with, but it is a web-based service. It has a service somewhere. All the software you're doing or you're programming with is not yours. You are essentially relying completely on their system. their system. So, let's say you are just new into home automation and you buy something from a Belkin, which is like a Wemo, which, you know, can control a PowerPoint. Right. And that PowerPoint is going to fire up something in your house when you're with the next number of kilometres of home. Now, currently, that is a functionality that is not available to you without any- Well, let's say that the Belkin Wemo is designed to work specifically with if this, then that. And that's why the way, that's what you did when you bought it, that's the way you configured it. Now we picture a world where if this and that disappears. So essentially now your purchase of hardware is no longer capable of doing what you had originally intended it to do. Now it may well also be able to be connected up to another home automation system, but maybe it can't be automated the same way you wanted it to be. So maybe that thing has now become useless to you because it's tied to an internet service. If that internet service goes away. So I guess the problem I've got is that this is going to start happening more and more. Right. And I thought about, well, what about other services, things that people just take for granted and say, oh, you know, it's always going to be there. Well, here's some other things I had to think about. What about radio and television? Because when you think about it, I realise it's not the internet technically. However, if you've got a radio station or a television station, they have a transmitter up on a hill somewhere with big antennas pumping out kilowatts of radio waves. Our TV sets pick them up and/or our FM radios or AM radios potentially, and we listen to them. Well, some people still do. So if, let's say, one of the radio stations would go out of business, you know, which happens sometimes, does that mean that your radio is now useless? Well, no, because there's dozens of other stations out there, presumably. In most cases, I think that's true. And why- And is it difficult for someone else to come in there? Well, no, because if you've got all the hardware and all the hardware is available off the shelf, you set a frequency, you buy a licence, you start broadcasting, you're getting your revenue from advertisers mostly and sponsors of the radio station. Therefore, you know, so long as you have people listening and you can have evidence that people are listening, then you have revenue. So, these radios are generic devices, they'll listen into anything you like, depending on who puts it out there and your source of revenue that you've got, your business model is completely different. Now, if you compare and contrast that with something like Peak, where you've got a hardware device, and let's say the FM radio that you go and buy, yay, I can now do cool stuff and listen to the radio. And then the radio station you really like goes out of business. your radio is not useless because the next one will then pop up and the next radio station will pop up and you can listen to them. The question is, did anyone care about taking over Peaks stuff and running a server and supporting their stuff? Well, no, they didn't because it was proprietary for one thing, but more the point, what's the revenue stream? Where's the revenue coming from? Because they sold their hardware, that model was completely different. So the problem I've got is when I was thinking about it is radio and television are completely different because their model is based on ongoing revenue that's ad supported. It's the ongoing revenue that's the point. And the fact that the devices are generic, you don't need anything that's that proprietary. It's been around for so long, there may be different brands and so on, but the rules for broadcasting radio are well understood and the equipment is widely available. It's not rocket science and it's not a problem. Right. So what kills this is the idea of of the business model, and it would be OK, I think. You know, I'd be less concerned, let me put this way, I'd be less concerned about internet connected devices if there was an ongoing business model where they would get revenue to sustain their existence. Yeah, even if it's enough to pay for one person to maintain one server, if that's all it took. Obviously, victim of your own success, if you need to have 100 servers or 200 servers or 2000 servers, and you need more than one person, and you need to buy new servers every five years, you know, you got to make sure your income is going to, the company's revenue is going to be able to cover that. And the only way I can think to do that is through subscription, a subscription based model. Yeah. You know, unless of course your product is capable of delivering advertising. Not sure how you do that on thermostat. Unless of course your name's Google and that's what they're going to do with it. You know, I wonder what the temperature is. Well, first you got to watch this video ad for 10 seconds. About a BMW or something. If you go to IFTTT, I mean, yeah. How are they going to distribute advertisements? It's whenever they do need to monetize. It would have to be a subscription model. I mean, yeah, unless you're- Because I don't want ads getting- I mean, I have 21 recipes there that I'm running right now. I have 30 and 21 are turned on and none of those are recipes that would work well with ads. So if ad support is out, the question, the burning question is, would someone pay a subscription to a company for that flexibility? And I'm willing to bet the majority of people would say no. Yeah, you'd have a hard- It's because it's- It's a hard sell. It's a hard sell. It's a, and it's a geeky thing. And as soon as they did that, you'd have 20 projects on GitHub trying to recreate it. So you could run it on your own server. Probably. And you'd look and think, oh, what could I just use with Automator? What could I, you know? See, this is the issue that I've got, is that everyone is saying, and I say everyone, OK, not everyone, but There are people that have said, and this is relating to CES, so I don't- I just want to specifically call out an article that was written by Ben Bajaran on the 10th of January called The Last Days of the Internet of Everything. I don't know if you've read this particular article. It was with relation to CES. If you haven't- Yeah. Yeah. It is- People should read it. It's a good article. It is a good article. It's thought provoking and I mostly agree with it. What I want to quote out of it is a specific sentence. I'll just read out to you. In just a few years time, if something is developed and shown at CES that does not connect to the internet or another electronic device, we will doom it an instant failure. And the expectation, therefore, is that it is inevitable that if we have not already crossed this threshold, we are imminently going to cross the threshold whereby anything new presented at CES that is not internet connected in some fashion is going to be essentially written off as useless, irrespective of whether or not it is. I think that's in the context of CES, I agree with that. Beyond that, I'm not so sure because obviously, you know, consumer electronics, you know, up until not that long, up until the Nest, how many people saw a thermostat as being consumer electronics, right? So sort of things change. You know, 20 years ago, before home automation, did anyone see the light switch as consumer electronics? No, I don't think so. It was just electrical. So, lines, things merge, markets that were clear become less clear. And as it evolves, the temptation to add more of this this whole idea of, oh, the Internet will just make it better somehow, we can just do, we can do this remotely, we do that remotely. It doesn't matter if you actually, if that's any value to you or not. It doesn't matter if that's any good. We'll just throw it in there because, hey, wouldn't it be cool if. And the problem with that idea is the expectation that when I buy something, that it will continue to work and all those features will be available. And if there's anything that the internet has taught us in the last decade is that online services don't last forever. So, it all sounds so good in the ad copy, you know, it sounds so good at CES, it sounds- Even the first few years where everything's all nice and good and they're getting good sales and then people start to realise, well, there's really not that much benefit to this. Then what do they do? Declining hardware sales, they don't have the revenue to continue to support the software services. The software piece dies and then the product dies, or at least the reason most people bought it dies. And you end up with an expensive thermostat or an expensive light switch. Well, if we look at something like going back to the if this, then that. And I'm also thinking about things like, like, um, Dropbox to a certain degree. A lot of these, uh, you could, you know, maybe it's this term has come around, like infrastructure as a service, right? A lot of these things aren't, uh, aren't products, right? Like in the Steve jobs sense, but they are, are useful. Um, their infrastructure, right? I mean, I mean, they're utilities that maybe you could finance their operation through contracts with not with individual consumers, but with with the the companies themselves. So maybe there's a maybe there's some and I'm just making something in my head. you know, maybe there's a service that would be the substrate for any number of connected thermostats, not just not just, you know, not just a particular brand, but it's something that would enable some general functionality, maybe in that case, just providing providing the server component. So you could know where it was at. So you could you could reconnect. I mean, do you think do you see what I'm saying? Do you think that's something that is a potential future that will see people start to fill that role because it's a problem. I mean it's it and it's not going to go away without some sort of reaction Not sure how much of that I've seen in beyond enthusiast level Like you said everyone will make a git clone of it and they'll try and set up a server here and there and They'll try and keep a little bit of a community of the enthusiasts going but on a mass market scale I'm not aware of it actually really happening and being a sustained success I can give you an example of a hardware based situation. There's in automation, for example, one of the brands that I've dealt with is Texas Instruments and TI sold their PLC division many, many years ago to Siemens. However, they also had licensing for a company to make modules specifically to fit their PLC chassis. And this company took over support agreements by creating all new cards based on current technology that fit in the old form factor, it was still programmed by the same software, and they did it at a decent price. And they essentially took over the support role for a lot of the old TI stuff from a hardware and firmware point of view, as well as some new cards that they brought on board. And I guess that is a rarity. Most of the time, bigger companies gobble up the little ones and then after a few years, they spit out what they don't want to keep. Whatever's not making money. I'm not saying that's what's going to happen with the Nest necessarily, but I think there's cause for concern. I think that in the end, Where I see the pragmatic component of this is people need to go into this with their eyes open A product that lasts, its hardware lasts, software never fails, right, technically If you've got buggy software that's one thing, but software doesn't deteriorate in time Capacitors do, resistors do, anything that moves, especially, does bearings, knobs, button switches, all those things, they all wear out. Age does not do them any favours. However, if a key component of a device that you buy is controlled by a third party that you have no control over and they cease to exist, your product could become either crippled or completely useless, depending upon how integrated and how reliant it is on the web. And what I think people need to do is when you do buy a device that is internet connected and has a bunch of functionality that you want to get out of that internet connection, just go in with your eyes open. It's so easy with smartphones because smartphones have got, you know, two, three-year shelf life depending on how, you know, depending on how nice you or kind you are to them. But most people don't have the same cell phone for 10 years. They just don't. people do, but they're a very small in number. So you don't have that. You have an expectation when you buy it. If the software ceases to be updated and it's no longer supported and all the apps you want to get now for iOS 7, like you've got a 3GS, let's say, you can't get half the apps anymore. Well, you know what? Your phone's now three, four years old, whatever. There's an expectation you need to upgrade. Right. But you go and tell someone that about a light switch. Oh, your light switch is five years old. You must need a new one. I mean, what? Well, it's because your 3GS, I mean, aside from any degradation and the mechanical mechanicals, it still does what you bought it to do for the most part. Well, that's true. I mean, it's obsolete, but it doesn't degrade. That's true. And I guess that's a bad analogy. I guess perhaps the point of what I'm saying is you go in with your eyes open and realize that if you buy something today that is based on that requires internet services to some extent that you're going to rely on. You have to go in with the expectation that someday those services will not be there. And ask yourself whether or not that device is still useful to you when they're gone. And that's the way I think people should look at it. Well, it makes you think about what you're really buying. Because I mean, I can say, yeah, your 3GS doesn't degrade. But if you know you're going to be pushing forward, You know what kind of user you are, then, you know, the job that you want this phone to do for you is going to change. Then, OK, you know, you know, it really is degrading, right? It's not doing, not doing what you want, but it requires you to admit that what you want is to move forward. I just want to switch. That's easy, you know, I don't need an internet controlled switch. I guess the point also is that if a company has an ongoing subscription model for access to their internet services, that would give me a much better and I hate to say it, warm, fuzzy feeling, because at least then there is some possibility that they are going to have a sustainable business model. Right. Such that you will continue to be able to use their service in years to come. Of course, it's still no guarantee there won't be mismanagement. The managing director or CEO may blow all the money at a casino in Vegas, for all you know, and you can't account for, I mean, if there's a war, if things go insane, if there's a subprime mortgage crisis, I mean, you don't know what's going to happen. But the funny thing is, the guy that bought the thermostat that wasn't internet connected, Like they won't care because none of it will affect them. So I think it's something to be wary of. Go in with your eyes open. Having things that are interconnected sounds like such a great idea. Oh, you could do this cool stuff with it. Do you really need to do that? Well, maybe, but go in with your eyes open and realize that in five to ten years' time, you might be holding something that's a paperweight. Should we wrap it up? Wrap it up. Wrap it up. If you want to talk more about this, you can find John on Twitter @JohnFidji. That's J-O-H-N-C-H-I-D-G-E-Y. The same on You should check out John's site, If you want to send an email, you can send it to [email protected]. I'm Ben Alexander. You can reach me on Twitter @phiatluxfm. And you should follow @PragmaticShow on Twitter to see show announcements and other related materials where we're going to be posting the time for all the time and location for our live show next week and it's a good place to get in touch if you want to talk about that or always a feedback on the new format very keen for anyone's comments on that so please let us know I'd love to hear your feedback all right John anything else we good I think we're good all right good show you guys next week see ya (Music) [Music] [Music] (dramatic music) [BLANK_AUDIO]
Duration 52 minutes and 54 seconds Direct Download

Show Notes

Related Links:

Premium supporters have access to high-quality, early released episodes with a full back-catalogues of previous episodes


Ben Alexander

Ben Alexander

Ben created and runs and Fiat Lux

John Chidgey

John Chidgey

John is an Electrical, Instrumentation and Control Systems Engineer, software developer, podcaster, vocal actor and runs TechDistortion and the Engineered Network. John is a Chartered Professional Engineer in both Electrical Engineering and Information, Telecommunications and Electronics Engineering (ITEE) and a semi-regular conference speaker.

John has produced and appeared on many podcasts including Pragmatic and Causality and is available for hire for Vocal Acting or advertising. He has experience and interest in HMI Design, Alarm Management, Cyber-security and Root Cause Analysis.

You can find him on the Fediverse and on Twitter.