In this episode, we dive deep into the emerging trends with Brian Sletten, a forward-thinking software Engineer. Recorded live in Chicago, IL.
Michael : You’re listening to No Fluff Just Stuff podcast. This is your host, Michael Carducci, and we are live in Chicago, Illinois. I am joined, this week, with one of my favorite people on the tour here, Mr. Brian Sletten. I think this is a really great talk for him, because Brian always described himself as a forward-leaning software engineer. The topic this week came up, as we opened this up to anything people might be interested in, is emerging tech. I know this is an area, Brian, that you spend a lot of time thinking about, and there are actually a couple of questions that came up in the course of just finding out where people’s interests lies in this topic that really are right in your wheelhouse. Brian, why don’t you say “Hello” to everybody here, and if they haven’t met you before, tell us just a little bit about yourself.
Brian: My name’s Brian Sletten. I’m an independent software consultant. I live in Northern California, and my career has largely been living about 10 to 15 years ahead of the markets, which is a fun, but not very lucrative place to be, or helping big companies stop from stepping on themselves.
Michael : I think this is definitely a great question to be diving into because it is something that is usually on the forefront of the minds of engineers. I know some of these questions around what technologies are emerging or around the corner? It is kind of a perennial question that comes up on the tour, would you agree with that?
Brian: Mm-hmm (affirmative)-
Michael : There’s a lot of reasons for this, and I know I can only speak for myself, when one of the reasons I care, at least a little bit about a lot of these things, is that it gets old having to constantly learn new things, especially just as you’re finally getting fluent with whatever the flavor-of-the-week is. It’s no longer the flavor-of-the-week. It would just be really nice to kind of have an idea of where I should be focusing my energy now so that it just becomes in vogue by the time I’m becoming fluent with it, instead of always seeming to play catch-up. That’s definitely one aspect of it. That’s one dimension of it, but there’s a lot of things to be thinking about in terms of things that are emerging, areas of our industry that are kind of going away areas, that are going to be in demand. From a career trajectory part of it as well, and I don’t know if you have any thoughts on that?
Brian: Sure. One of the nice things about having a long career, in this industry, is that you get pretty good at identifying trends, and identifying things that you should spend your time on, things to avoid. As an example, somebody who sort of lived through CORBA, was able to look at web services and say, “This is CORBA with angled brackets, right?” It had certain solutions to problems, but those problems were not universal, per se. That was an approach that I was able to successfully avoid getting sucked into too much, or to understand why something like the web, as a way of sharing information, was maybe preferable to more complicated, tool-heavy, approach of web services. It’s not something that you can know, going into this, without having experienced some of these things, but you learn to identify patterns, you learn to identify architectural risks and opportunities. That can help drive interest, or lack of interest, in particular technologies.
The consequences of these things as well, like what has the consequence of making somebody’s lives better? Improving society, making developer’s lives easier, I don’t know. What is actually helping improve things?
Michael : I think that’s an important point of it, is that we look at a lot of new technology almost in a vacuum, without asking the questions like, “How is this solving existing problems? How is this making our lives any better?” Looking at the “Why” of some of these things is definitely valuable. I know you have kind of a process. You have kind of a way of evaluating new technologies because being in this field, speaking, training, constantly looking ahead, a lot of things are coming up constantly. I believe you’ve mentioned in the past, you’ve kind of got a framework for dealing with these things. Do you mind sharing that with us?
Brian: Sure. Again, there’s no rocket surgery around it. It’s just a question of, we have finite amounts of time, we have finite energy and resources to ingest these technologies. As I’m going through my day, I tend to try to follow interesting people on Twitter and various social networks. I try to include people from different technology bases to different industrial bases, different parts of the world, just so I’m not living in a filter bubble.
Michael : That’s easy to do in social media.
Brian: Right. When something gets mentioned, I sort of refer to my “snooze model” where the first time I see something, I may go look it up briefly just to kind of contextualize it, but I’m not going to spend a whole lot of time on it until I start seeing other references to it in media. It doesn’t have to, necessarily, become full-blown articles, because usually by the time it’s gotten to be big enough that people are writing articles about it, it may, or may not, have proven itself interesting at that point. It’s more just making sure to follow interesting people and look at what they’re interested in, and then using a frequency metric to figure out. Whether this is something you should spend your time getting into, so when you can’t have your own pattern recognition based on experience following interesting people, and leveraging theirs is a useful way of figuring out what to spend your time on as well.
Michael : Absolutely. I know there are a number of specific questions that we got from the room, from the audience here in Chicago, and one of the first ones that came up was blockchain. It’s one of those things we hear a lot about. I actually spend some of my time at conferences that are geared towards more management types in IT, CTO, CIOs, things like that. I found myself in a few of those, and it seems like blockchain is coming up all the time, and everybody’s talking about it, nobody really seems to say anything. I see that in a lot of other places as well, but it’s this fascinating technology, and people are kind of convinced that there are these other applications for it outside of cryptocurrencies, but nobody seems to be doing anything with it. Nobody seems to know what to do with it. I’d be really interested in your thoughts, because I know that’s an area of your expertise. In fact, I believe you’re about to go do a blockchain training event in London, is that right?
Brian: Correct. Blockchain, actually, a lot of technology and the ability to identify them, becomes easier if you feel the need for them before you actually come across them. Blockchain and Bitcoin were one of those. Of course, if I knew what the hell I was talking about, I’d be sitting on a beach retired by now, given the current Bitcoin price, but I’ve been tracking it from the early days, and just never really got into the process of doing the whole mining thing before it got too late. I’ll take it back to a friend of mine from college, a guy named Adrian. I used to have discussions about something that he had, a plan, which was to publish hashes of documentation as advertisements in newspapers.
The idea was: if you had an idea, you don’t want to share the idea prematurely, but in terms of subsequent potential legal battles, you want to demonstrate ownership of the idea prior to its being produced from somebody else. His idea was to take a document, hash it, and then hash the hashes into an opaque page of numbers, or an opaque part of a page of numbers, publish it in the news print. That way, if you ever had to go back in time, to say, “Look, this is the first time I presented evidence of ownership of the idea,” you’d be able to regenerate the hashes from the published page at a point in time. We had those kinds of conversations, and also, one of my bosses had worked at a company called CyberCash, and so the whole concept of digital currencies had popped up on my radar around that time. It was really the combination of getting into distributed computing, looking at cryptocurrencies and understanding the proof nature of what a distributed ledger would do, that got me to resonate with blockchain as a solution.
Obviously, everyone’s fairly comfortable with the ideas of what Bitcoin does, and how it leverages the blockchain, but the question was: What about non-currency uses of those? I would challenge you to say your point where you said nobody’s actually doing this. There’s actually dozens, and dozens, and dozens of new kinds of opportunities like people who are being given some kind of cryptocurrency, or value, around storage on your hard drive. I mean, I built a system 18 years ago where we sold idle time on unused computers around the world, like it’s SETI at home. The problem was, we had no real great way of compensating them because the financial system didn’t work in our favor. Credit card payments, and bank transaction fees, and things like that were too high, but in terms of them being able to allow us to use their resources for some value transfer, something like a blockchain solution would have been nice. We’re now starting to see if you let people use a certain amount of your hard drive, you can get some kind of value transfer.
Obviously, Airbnb and Lyft, and things like that, are models where selling access to resources, so they’re more efficiently utilized as opposed to you buy a car and it sits idle 80% of the time. We’re starting to see more and more people build systems like that around resource utilization.
Michael : This is where the proof of work type of thing comes in. In fact, if I understand correctly that is?
Michael : I think I heard this from you. There’s a start-up somewhere in Europe that is selling heating appliances that are distributed computing nodes?
Brian: Yes. There’s a company called Qarnot in Paris who are combining the externalities of high computation being energy expenditure and heat generation, which is a problem for data centers with the fact that people need heat to heat their homes. They’re now installing nodes in homes and apartments in Paris, and then scheduling activity on those. Obviously, they will have models that factor in the temperature outside, and whether something needs heat or not, or what not. In terms of scheduling algorithms, being able to do these sorts of things with a value transfer that is not tied against bank infrastructure transaction fees or time periods, those sorts of things start to make a lot more sense.
As an example, you can imagine companies that maintain histories of cars, are trusted. They’re trusted to tell the truth about the fact that the car has been in an accident or something, but you could imagine combining the emergence of LTE, Wi-Fi capabilities in modern cars with collision-detection systems such that if a car were in an accident, it could immediately publish into some kind of blockchain system, the event, and therefore, you wouldn’t need to go to a third party to discover that event. You could search the blockchain for this particular VIN to see if it ever registered itself in that capacity. That’s the kind of case where leveraging the trust, trustless, model of the blockchain where we can trust trusted transactions amongst untrusted participants, to capture and maintain long-lived data stores in time. Coincide that with an incentive program around people’s participation, and you start to see, if we move away from a purely currency model, we can start to imagine this case where more and more evidence of when transactions happened, who did them? Who did them first? That’s real economic value that can be used to generate new business models, it will have impact in terms of investment bankers and other sorts of people, do asset allocation for IPOs and things like that.
There’s a tremendous number of blockchain examples that are falling outside of the cryptocurrency space. Don and Alex Tapscott’s blockchain revolution is a good, non-technical business oriented book for discovering some of those possibilities.
Michael : Is there anything, just kind of in closing to wrap up this point, is there anything that we, as software engineers should be doing maybe to develop some skills in this area? You mentioned that it was obviously a little too late to start mining Bitcoin. The best time to do that was probably about a decade ago.
Brian: If you invent a time machine and go back in time and start mining Bitcoin, then you could pay off the time machine investment.
Michael : Other than that, is there anything we should be doing now?
Brian: The trick now is to start to think about these business models and to think in terms more along the lines of smart contracts. Bitcoin and blockchain are of particular use, but we’re seeing the emergence of new platforms like Ethereum that allow a contract to be encoded in a machine-processable way that it’s also able to leverage the economic incentives, the trust models around these sorts of things. We’re only beginning to understand what’s going to come from that. Now, they’ve been hit by a series of technical flaws and what not, that have caused some pretty high-profile, and embarrassing, thefts from the network, but they’re proceeding at pace, and imagining entirely new kinds of disruptive industries on top of existing industries by combining these ideas together. I would say look at what Ethereum is doing, dig into what the smart contracts are there.
As an example, you could create your own DNS system where somebody could register a name, if they’re the first person to get there, those sorts of things, you could build on top of the back of the platform without having to set up all of the infrastructure yourself. If you can think of a business idea that could benefit from distributed, shared, agreement amongst untrusted people, things like that are just ripe for innovation, and disruption. Improved cost, improved time-to-market, improved transaction times, so we’re not stuck with the existing banking systems, and what not, who are going to live on float for holding our transfers for a day or two in the process. These sorts of things can be made much more rapid. I would say investigate Ethereum and this Solidity contract language as an example of what’s coming.
Michael : Just to kind of transition on to another one of the questions, if blockchain is very definitely one of the squares in technology bingo, I think IOT is another one of those, just in terms of the buzzword-i-ness around these technologies. That’s actually one of the questions that came up. We’re seeing more and more Internet-of-things. There’s a big security dimension of it that we start thinking about, especially as these things are getting rushed in the market. The other thing is, we’re going to get more and more traffic from these devices, and a lot of these devices are proving to be very, very, very chatty. What impact do you think this is going to have on our Internet-backbone and the infrastructure as it is? We get all these chatty devices, and we start getting billions of these things over a short period of time.
Brian: I think the good news is, we have decades of experience in terms of levels of scale and how to scale the Internet up. It’s right there in the name. It is the Internet-of-things. It’s not the arbitrary architecture that a couple of Yahoo’s threw together, kind of things. We’re leveraging our capacity to bring multiple participants onto this platform, and allow it to leverage the strengths that are there, from a scale perspective, from a technology-evolution perspective. That’s everything. Those are all the problems they don’t have to solve as these things come alive, as new, underdeveloped countries, people come online with phones and what not. I mean, we have similar sorts of problems in terms of mobile devices and technologies. It’s not just the doorbell sensors and what not, but they are generating quite a lot of activity.
That’s why the move towards IPv6, and other kinds of scalable infrastructure, faster Wi-Fi networks, those allow us to respond to new topologies without having to lay fiber cables and what not. We’re at a point now, where at my house, I just signed up for Gigabit at home, over coaxial, so advancements in technology now allow them to approach. I would not say, actually, provide gigabit per second throughput there. That’s, in part, to manage the fact that we now have dozens of devices and doorbells, and lights, and online backup systems, and thing after thing, after thing, contributing to our bandwidth needs, so that when I sit down to watch a Netflix show or something like that, I don’t have to compete for bandwidth. I have enough capacity to have an HD or these days, 4K stream-
Michael : With everything else.
Brian: With everything else that’s on the network. The market is responding to the demand for this. They’re putting more infrastructure in place, they’re developing new ways of getting more capacity out of the extant, infrastructure. Those things are going to service, I think, at least from a bandwidth-perspective for at least the immediate future.
Michael : So your contention, basically, is that we’re going to be able to keep up with it for the foreseeable future? You don’t think there’s going to be a big, hockey-stick kind of spike that’s going to catch us off guard?
Brian: Certainly from a bandwidth-perspective, I think we’re going to be okay. The deeper implications are what the devices do, and that data, and the fact that all of these companies are generally not considering what it means to send this kind of data over networks. If they don’t have that capacity to do TLS transport layer encryption, then the data that they’re collecting is being sent over networks. There’s a perimeter mentality at home, and at work, that people have about … Well, the bad guys are outside, and the good guys are on the inside. What with phishing attacks, and rogue devices and things like that, that’s a clearly decades-outdated mentality. If they don’t start using encryption to protect the data in transit, as well as the ability to protect it at rest, as they store this kind of information, the big issue is going to be from a security perspective. These people are trying to create new markets, they’re trying to disrupt existing markets, and security is often a sadly trailing consideration. There’s a joke. I apologize, I can’t remember who I saw it from, but they that the S of IOT stands for “security”.
Michael : There is no S.
Brian: There is no S in IOT, that’s the problem.
Michael : Well, this is why we have refrigerators and Internet-connected toasters attacking websites, and holding people hostage. It’s interesting. There’s definitely that security thing, and I’m always reminded of the big target-data breach that the attack, the entry vector, to get into the network, was the Internet-connected HVAC system, that it was the thermostats, essentially, that were the attack vector.
Brian: It was the thermostats and the outsourcing of the management of the environments to third parties who needed permission to get into the various stores to test and control those things. Ultimately, the problem comes down to usage models, right? We have access control lists and simple cases for trying to map principals to capabilities, but the moment those contexts change, or the models get too complex, those things fall apart. We need richer authorization models that will deal with the fact that this data is being collected on a network that also has an Alexa thing, that’s listening to what we’re saying, right? It’s the combination of these things that’s really going to be exploited.
Michael : Speaking of Alexa and Amazon, and listening in all the time, now we’re giving Amazon keys to our front door.
Brian: No, we’re not. I’m not.
Michael : All right, who’s “we” right? There’s one other aspect of IOT that I don’t think has really come up in this discussion that I think is relevant as well, in terms of immersion technologies that the sheer volume of data, we’ve talked about it going over the wire, but the sheer volume of data that needs to be centralized, I see that as an unsustainable thing. This is ultimately going to be kind of a return to edge computing. The example that I’ve heard come up a few times form yourself and speakers here, is the self-driving cars. These are going to be generating gigabytes of a telemetry data every single day. You multiply this by the number of cars, by the number of days, this is just not feasible to store all of this centrally in some data center somewhere. We’re going to see a lot more of this processing and storage taking place in the devices themselves, so that’s kind of the other aspect that we’re probably going to see more of, just given the volumes of data that are going to need to be processed.
Brian: This is a fascinating point for me, because it tracks very nicely with my career. Trying to build this Internet-distributed computing platform 18 years ago. We knew at the time that there were only certain kinds of problems that would fit well on that model, and that data-centric, data-heavy things would not work well in that space, particularly, again 18 years ago, in terms of the bandwidth that was available. We did some experiments with things like Monte Carlo-based digital rendering, which produced fantastic results, but the data-to-compute threshold was too high to be effective at the time. What we’re seeing now is cloud-computing is one part of the advancement, multi-core systems are part of the advancement, GPUs are part of the advancement. Internet-of-things devices, and sensors being forward-deployed, on the edge, are part of the solution. Our architectures are changing and our programming models are going to have to factor that into this.
I encourage everyone listening to this, in the audience and beyond, to read an article that Herb Sutter wrote called: “Welcome to the Jungle”. In that, he cites another article that he wrote a couple of years before called: “The Free Lunch is Over”, but between those two, he does a fantastic job of contextualizing the advancements that are happening in our industry about meeting the needs, the computational needs, of different architectures, of different business models, of different data models, and basically putting the compute where it needs to be. We do not want our breaks in our car making cloud calls to decide whether to stop the car or not, right?
Michael : Ideally. Going a little bit about the security, and things like that, it seems like there are challenges that we’re dealing with, but we also have a lot of new technologies that are purported to make a difference to what we’re doing as software engineers. I know one piece of that is, I think we’re getting a little bit better about static analysis, well, from a security standpoint, but just in terms of a code-quality standpoint, but there’s a lot of technologies that are purported to make our lives as software engineers better. In your opinion, do you really feel like these are … Is this making a difference? Is this making our lives better as software engineers? Or are we starting to over-engineer things? Are we getting more and more complex frameworks that are kind of going into existence that aren’t really even solving the problems that we really have?
Brian: I think it’s a little bit of both. A lot of times, there’s just flux. People engage RDD, Resume-driven development, in their technology selection, and that’s not generally a helpful approach, except for, perhaps the person who’s trying to pad their resume. There’s a really fascinating perspective on microservices. I think that sort of highlights what’s going on. Nobody sat down, historically, to create a microservice as architecture. When we look at Netflix, and SoundCloud, and Gilt, and these various organizations that we hold up as success stories. Their goal was not to say, “Hey, let’s go break up the monolith into these simple, little things.” It just happens that that became the bottleneck of their capacity to operate at a modern cadence. If you go back and look at the developments in the industry that have happened in the last 20 years, you can see this happening.
The first initiative was to solve the planning problem. Back in the early projects that I worked on in my career, we would have 18 to 20-month release cycles, and that’s just impossible to get right because people don’t know what they’re building. They’re not building the right thing, they’re not building the thing right. Agile was a response to that, to make sure that stakeholders were involved, so that we did build the thing right, and we built the right thing, and we were able to, as software developers, operate at a cadence, as individual contributions could be more predictable, more accumulative, and we knew, largely better when things would be delivered and when value would be added to the production systems.
Then the next bottleneck became when you and I, as individual contributors, had to integrate our code, that became the bottleneck. From a planning and daily activity scheduling perspective, we were fine, but as soon as we had to connect our stuff together, things fell apart and our cadence was disrupted. That’s where continuous integration emerged as a discipline where we can solve that problem by biting off little bits of integration here and there, such that our cadence was able to maintain through the integration process. Then the bottleneck became the deployment process, and our ops people have, historically, been able to have long periods of time to roll this stuff out. Now, the developers were able to respond to changing business environments and requirements, and develop this productivity cadence, and integrate continuously, deployment became the problem. That’s where DevOps emerged and Continuous Delivery emerged as a discipline.
It wasn’t until organizations had adopted all of those things that the idea of breaking up monoliths into microservices that could have their own release cycles, became more interesting.
Michael : This is why Neal Ford is fond of saying, “Your organization must be at least this big to adopt microservices.”
Brian: Absolutely. You’re just not prepared, organizationally, to be able to handle the responsibility of constantly deploying simple, reasonable bits of functionality unless you have adopted these other sorts of things.
Michael : What’s really interesting is you were talking about how microservices were a solution to a very specific problem that nobody really set out, and then initially to build microservices that was just the right tool for the job. When I look at a lot of refactoring efforts right now, people are breaking up their monolith into microservices for the pure reason that they feel the need to break their monolith into microservices.
Brian: To get back to your question, and to tie it into that problem, we have new technologies, we have new procedural approaches. We have new ways of doing things, but all of them serve certain purposes. If you serve the purpose of the architecture, of the process, of the approach, and you understand the value-add that you’re getting out of it, then I think these things do make an accumulative approach towards doing things better. If, along the way, you’re picking things because it’s trendy, or because you want it on your resume, or because everybody else is doing it and you’re not prepared for it, in that case, we’re not going to see the benefit. Overall, I think the adoption of developers, unit testing, and integration testing, and DevOps, and Continuous Delivery, and those sorts of activities, I think, have allowed organizations to better-align on business boundaries. That has made things like microservices viable, but if you don’t start from a mature perspective, then none of these new technologies are going to make a difference and you’re just going to fail differently.
Michael : We’re running a little light on time. I want to cover at least one more topic here that came up from the live audience here in Chicago, and just a new thing that we’re just seeing more and more, that’s beginning to pick up steam is all the VR, AR stuff. I certainly remember the first time I put on a Vibe headset, I was completely blown away by the experience. I was not expecting to be as immersive, and just as incredible as it was, I felt though, at the same time, that we haven’t even begun to scratch the surface of the capabilities of the technology. There’s a great company called Prime. They’re building VR code visualization tools, and I got to play with their technology in UberConf and actually do a code review with another person of my codebase and just walk through it, and look at things, and point at things, which was pretty incredible.
We’re still only scratching the surface of this. Now, we’ve got the new iPhone hardware and their release of AR kit. Just seeing some of the incredible capabilities that AR has in store. I don’t know if you have any thoughts really on where this is going? What the future of that looks like, and also, how as a software engineer do you start getting into these fields? Because I do believe there’s going to be some explosion in these areas, that we just haven’t even found the killer app yet for any of these and it’s coming.
Brian: Recurrent theme of this session is if you’re in this industry long enough, you’ll see things come around. I worked on the first-world, first whole-Earth environment, think Google Earth, but 23 years ago. At the time, it required SGIs, and we all had SGIs at our desks, but we were dealing with terabytes of information and things like that. It was fun, but it was not a consumer-oriented model, and obviously, before long, these things were being handled by graphics cards in PCs. The visualization that was possible with these tools, was incredible, but it was just outside of the hands of the average user, right? It was just very deep-pocket organizations were able to benefit from them.
I think the thing that we’re seeing now is that the consumerization of VR and AR are going to drive incredible, fun, initiatives. If you see what Peter Jackson and the [Weeda 00:33:03] folks were doing with some proof of concepts around Lord of the Ring content in this experience, think about what this is going to do to tabletop gaming and what not.
Michael : No kidding.
Brian: To play games against people in a physical environment that transcends where they are in the world, and make it look like they’re part of your immediate world. These are all social things. These are all social aspects of game play and technology that I think are going to be well-supported by the AR and the VR. Particularly the libraries and the toolkits that are showing up in the operating systems in the phones now, are going to make it easier for developers so you don’t have to understand, necessarily, all the math, all the linear algebra that goes into these sorts of things. You can just simply place things in a scene and have some behavior associated with that. That’s going to be immensely useful from a scientific exploration, being able to, as a researcher, put yourself within a muscle structure, within a bone structure to visualize these things.
How do to remote surgery, right? Hooking in haptic feedback for these sorts of things, that are missing in a purely digital or virtual world. That’s why I’m more bullish about augmented reality in the short-term than virtual reality. Ultimately, the kinds of things that are possible are going to be just insanely cool, fun, mind-blowing and terrifying at the same time.
Michael : When I was at Java One, and Oracle, open world early October, that was one of the demos they had. Was they had an augmented reality hard-hat where you could walk around the space that we were in, and see where all the pipes and HVAC duct work is, and all the wires, and everything like that in the room. You could kind of see through the walls, which I thought was pretty need, and I could see some value around things like that, but I got excited about AR when I was watching the Apple WWDC, or whatever that event was. I was watching it. It was pretty ho-hum, I thought. I was like, “Here’s another incremental Apple announcement thing, and they’re doing this now, and they’re adding that feature now, finally.” Then I saw the demo of AR Kit and there was that, where they just projected this whole world, and this whole scenario onto a table. Then it came under attack, and it was the moment that a guy walked off of the edge of the table and fell, that I just saw that, “That is just ridiculously cool.” I started to get excited about that.
I can see why there’s a lot of excitement around these things, and making these accessible. What would you suggest, if you wanted to get into this, right now?
Brian: Obviously, the AR Kit and the equivalent that are showing up in the Android space, off the top of my head, I don’t know what they’re called, but those are fairly low-effort platforms to get into, so I would say use that as a playground. We’re going to see innumerable examples of these showing up in the app stores, play around with them. The biggest hurdle here, I think, is going to be vision and ideas, and appropriate uses of these things. Your example of being able to look through walls to sort of see where the studs are, where the connection points in plumbing are. That’s going to be one aspect of it, perverts creating ways to see through clothes is another aspect of it. This is, again, where technology starts to get into social, legal, privacy kinds of concerns. Nothing is without these kinds of consequences. I think we’re doing better overall, but I think we as an industry have to always ask the question, both what do we get out of this? And what new problems does it solve? Very often, they’re social, and governance issues, not technical issues. We just need to be cognizant of that in our own day-to-day activities as well.
Michael : Absolutely. That’s where we always seem to fall down. Foreseeing these problems, but fortunately, I think we’re getting better at iterating and adapting, as we find these things. We’re running out of time here, so how about one more big round of applause for Brian Sletten, everybody?
Brian: Thank you.
Michael : We are getting very close to the end of the 2017 No Fluff Just Stuff tour. We’ve got a couple of events left, the ArchConf and The Rich Web Experience taking place in December. Definitely check those out, and check out all of our tour dates for 2018, all the cities we’re going to on nofluffjuststuff.com. Thanks for your time, thanks everybody for being here, and thanks for listening.