About this talk
When Amazon released AWS 11 years ago, it dramatically lowered the barrier to entry for software development, beginning a cloud computing revolution that lead to today's ecosystem of hosted developer tools. Pusher's hosted APIs make it easy for any developer in the world to build features like chat, news feeds, and notifications into their apps. This talk will focus on the evolution of software development and how developer tools are empowering everyone in the world to build apps without worrying about infrastructure.
- First off, I just wanted to thank the event organisers for putting this on, particularly Gen Ashley and everyone that's worked with us. Just an amazing event, we're really happy to be here and be involved with it. Can everyone hear me all right in the back end? Perfect, okay great. Okay so today, Amina and I are going to talk about why we think developer tools are key to empowering everyone to be able to access technology. So when you hear the term developer tools, it's kind of a dry subject I guess. But they've actually been really revolutionary in lowering various entry for developers around the world. So first I'm gonna go through some trends that were seeing and then we'll look back on the different eras of software development. And then I'll be talking about how developer tools make it easier than ever to build really cool stuff quickly, that it doesn't take much to get off the ground. And then after that, I'll hand it over to Amina who's our head of marketing. She's gonna talk about through those different eras coming up through the trenches and then now in our position meeting, push our efforts to take it even more developers empower around the world using developer tools. So living in London and working technology, I think sometimes I take for granted just how quickly these things are changing and definitely the privileges that I experience because of it. We buy our meals with apps, we get to go to these meet ups with technologists like yourselves. We get to experience connectivity all the time and I've had MacBooks and iPhones since I was an adult. But globally, were at this sort of infection point where a lot of these trends that are making software development itself much more accessible are happening at a dramatic scale. So in 2016, there were over 2.5 billion smart phones around the world in circulation, and that pasts the 1.5 billion PCs that were in circulation and it's just continuing to rise. So Ericsson predicts by 2020 that there will be 6.1 billion smart phones in circulation. Which basically means that almost a majority of the world's population has connectivity and is connected to the Internet via mobile phone. Connectivity itself is expanding really rapidly. So just this year, in India alone, 415 million people are going to get access to the internet. Which of course is larger than the whole population of the United States and you know, you think about that that the change that presents itself, when suddenly there's this networked population that have their own local problems but they're specifically now online and all the developers who live there in those places understand those specific problems that those people experience. Or suddenly they have this networked population that they can build digital services for, and new applications to scale solutions to those problems. So it's really exciting and I think fundamental change in the software market as a whole. So that stack overflow is a place where you can get answers to all of your questions about development and they do an annual survey. It's really interesting to look at the different demographics of software developers and the questions that they're searching for. So last year, they reported there were 16 million professional software developers. But what's really exciting is that there are over 46 million people who are software developers who are learning, not necessarily as their job, but also transitioning to that job. And so they're coming from all of these places and so kind of the question is and how are they entering this and where are they coming from? So people are running their code from amazing communities like this one where they're able to learn from each other in person and discuss things at meet ups around big cities like London, but they are also learning from places like coding academies, tech accelerators, online courses that are free and accessible to everyone and really just limitless Google searches about any possible technical topic you could think of that you can daisychain between learning one piece to the next. So for instance, when Y Combinator released a course online last week, Y Combinator's this great accelerator and Silicon Valley, but obviously because it's geographically located in Silicon Valley, it shuts a lot of people out that aren't able to immigrate to the United States, aren't able to have access to these resources. So they just added this course online that anyone in the world could take, and within one week, they had 10,000 sign-ups of people that want to create software businesses, specifically. So that's really amazing thing that's just happened last week. So in total the trends that you see are that there are more people than ever to build software for and there are more people than ever who are building the software. But the thing that's really exciting is that as these two trends converge, we're seeing that it's becoming easier than ever to build software that you can build a business around. So anyone can do this. So I'm gonna talk a little bit about how we got here and that it definitely wasn't always this way. But it's just this hugely exciting new moment. So back in like the dark ages of the '90s and the .com era - When I started my first job. - There's some crazy things that you had to do to be able to get a software business running. You had to have on-prem database and compute hardware. You had to actually have a brick-and-mortar office park building to host all this stuff. And so that was really costly. You had to hire people to manage that. So right upfront, to start a software business, you had to think about is our IT secure? Is it reliable? Is it scalable? And so if even one core machine in your stack broke, you would have service downtime until you fix that one machine and so obviously this is really a big cost upfront for anyone to be able to enter building a software business. And so, and then even way back when, before AOL and the pets.com era of Internet businesses, you had to actually release software on a CD-ROM in many annual releases. And I mean this is crazy because we all talk now about agile and how you want to get iteration from all the user feedback but you literally can only release once a year. So you had to do that in a way that would take a much longer time to get to market. And then finally, the educational barriers, a really big one to talk about, that in this sort of climate, we have this much higher educational barrier that was requiring a deep understanding of computer science and a lot less educational resources available. You know, big textbooks about C plus plus and really requiring IT management of hardware as the first barrier to entry. So in 2006 specifically, but also just generally in this time period, this really amazing thing happened where Amazon released AWS, Amazon Web Services, and all these other companies are also releasing cloud computing offerings basically. So what was incredible about this is that combined with all of the advances in connectivity in the developed countries and people getting all these new mobile devices, and suddenly developers no longer required on-prem database and compute management. So all of the sudden, you just had to worry about getting your service up on the cloud. So with all these people who used to manage hardware suddenly being able to do dev ops and basically spend their time thinking about how to deploy code faster, they could do things like getting closer to continuous deployment and basically to take iterations and immediately ship software to all these people that were using their product. Well what's really crazy to me is just that, this was only in 2008, was GitHub even launched. So the whole open-source community that makes so much of what we talked about in software development possible now isn't even 10 years old. So around this time, we had all these new languages, Ruby on rails, in 2005, new JS in 2009. These are things that are kind of taken for granted that they're really fundamental languages that people are learning now, but they haven't even existed for very long. And right around this time is when we see the emergence of social media, the iPhone is released in 2007. This is incredible opportunity for building apps on mobile devices. And then the whole Android ecosystem that came after that and how that's just scaled to just the millions or the billions of devices that we were talking about going online by 2020. So people suddenly had connectivity, could access all these new digital services and more developers could build businesses for them. So today, you still see the cloud advancing. In fact, there's increased statistic that's only six percent of enterprise companies are on the cloud even now. So the really big banks on these kind of companies. But the main cloud players like AWS and Google Cloud are still innovating at a really rapid pace. So you see company, and the unicorn companies, like Airbnb, running entirely on the cloud and never having to worry about opening a data centre to get up and running. But the current paradigm shift that we're really excited about, and we wanna talk about today is that hosted the developer tools. So even above the stack of the cloud where everything that we build a Pusher is hosted on AWS and you never have to worry about it. That makes it so that you don't even have to deal with cloud infrastructure to be able to get started with building an app. And for instance, if you wanna add payments to your application, you would simply send an API request to Stripe and they handle the hard part of processing that payment. So the barrier for entry has become so low that you don't even have to worry about the infrastructure, you just worry about the business logic of your specific application. So are you gonna build this kind of app or this kind of app? So as an example of that, let's say you we're building a food delivery app. It's really difficult to go business-to-business and get all these brick-and-mortar shops, these restaurants, to be partners in your new application and to add them to the inventory of food that you would offer in this app. So you could spend a lot of time doing that and still have to worry about this cloud infrastructure or you could spend the time and the hard part of building this business and then a lot of the development time could be cut just by simply integrating with these hosted services. So as an example, you could have your whole website hosted on Heroku and you could use a database from mongoDB and never have to host and manage all those things yourself. You could send an API request to Stripe to process payment. So as the order comes in, everyone is paid, the business is paid, the customer's able to do the transaction. You could send an order confirmation about what food they've ordered and they could see that all in an email through SendGrid, which is an API for sending transactional email. And finally, I mean the really exciting one for us, with pusher, you could build an entire experience that's real-time, meaning, as someone has the app open on the phone, they could see the driver coming to their house. They could get notifications as the order progresses and they could do a chat with the driver, if maybe they're just down the road. I've had this with delivery, where they're like, "Oh, I'm over here, this is my address!" and you can chat and contact them directly. So, the exciting thing about this is that each of these services has just a few lines of code of client and server API requests to be able to integrate into your app. But you have an and then business that's up and running as soon as you have food that's ready to serve. So you could get your business started a lot faster. So what pusher does is that we allow developers to build scalable real-time applications. So I'll talk about a bit about some examples of that in a second. But the really incredible thing is that we deliver billions of messages to millions of connected clients, phones with the app installed on them every single day. So this sort of scale is really hard to achieve if you're just a developer getting started. But with one API request, you suddenly have something that could scale to a million users overnight. So the really fun part about building developer tools is that we build something that solves the core business case for a lot of developers and then they use it in really interesting ways that we wouldn't even necessarily predict. So an example is that Lyft powers their entire rider experience with Pusher's API. So as you request a car, you see it moving on the map, it's coming towards, you get the notification that your driver has arrived outside. And again you could contact them with a message if they're not coming to the right corner or you need to cancel. The New York Times is an interesting example because they're actually powering live coverage of news events. So last fall, and unfortunately as an American, I have to recall this now but, we had debates, presidential debates in the US and the New York Times powered the whole experience of their coverage alongside the YouTube video of the live debate. So they built a Facebook lifestyle emoji reaction, which you can imagine had a lot of angry faces on it. But they were being powered with the 200,000 concurrent connexions on this page as the candidates went back and forth about the different issues that were discussed in those atrocious debates. And so, the fact that we could handle this can scale for them and they don't have to worry about it is really enticing for someone that they wanna just focus on political coverage so alongside, they have all the commentary from the top political reporters posting analysis and statistics in real-time, all powered on top of Pusher. It was pretty amazing, 'cause it was actually our the most concurrent connexions we've ever actually handled and it was fine. No one had to get up and like kick the server at night or anything like that. Then a third examples is that every kind of application you use in the workplace, there's collaboration and everything is moving towards a place where single page applications can handle this kind of collaboration in real time. So you don't have to reload the page to see that your email is sent on MailChimp or do you have to edit a draught in real-time kind of like Google Docs, and all these things are powered by the APIs that Pusher builds currently and then really, all these other cool real time use cases. But the thing that I really wanted, if you guys came away with anything from this, that why I love what we do and why I love my job is that developer tools have really lowered the barrier to entry for developers around the world. So what this means is that people come to us organically and they just start using our tool to solve problems that their app specifically addresses. So going back to all of these people who are becoming connected, 'cause they have access the internet for the first time and all of the developers who are learning to build these apps, we're seeing just in India alone that we have the next generation of ride-hailing apps, Amazon competitors, code learning platforms. So many other categories of technology that are being built but they're being built specifically for the people that live in those places, solving the problems that the developers that live there experienced. So being able to very quickly get to market something that goes something so vast at such scale, makes it so that everyone can just build and solve problems at a much faster rate. So today at Pusher, we are paving ahead on building sort of the next generation and incubating new ideas about how we can take a problem that a lot of developers have, build a tool that solves that and then ship it to market so that people can just make their request of the service with an API and never have to worry about hosting that infrastructure themselves. So with that, I want to turn over to Amina, who's our head of marketing. - So today, I think I spoke with quite a few of you at the Pusher booth, and there were few of us having reminiscences about the old days, I can see, you're the main frame developer, hello. Is it you, no? There was somebody who did stuff on mainframe. Oh, anyway. So I started my career in 1995. So in 1996, and the first years, I was doing a lot of product development, some IT projects. And it used to be just really hard and slow and boring. Because you were using waterfall, so expensive to do stuff. And most of the developers were even more men than now. And people typically had to study for years to get in. And it just wasn't that exciting. And then the world changed a whole lot. In the early 2000, people started to get smart phones, they had decent broadband at home. They were used to using computers. And there were a few changes, like the advent of open source and so on. And that made it really interesting, 'cause suddenly I could see a new generation of developers coming in. So mobile apps in particular became so cool. And you just felt like the world was changing a bit because suddenly technology was kind of interesting to getting into. They want customers that you could address, the tech was catching up. And it was just getting easier and cheaper to start building a business around technology. That was when I did my first startup because, I was playing around with some open source stuff with a developer friend of mine, and we went, "Oh my God, this is amazing. "We can start building things that used to cost tens "or hundreds of thousands dollars, "but pretty much for free "because we can bolt some stuff together "and there's some great user communities "that we can start asking for help from." And that was really exciting. And then coming forth another 10 years, and it's even better, 'cause the thing that for me has really changed is actually around the education piece. So I look around, events like today and some of the inspiring groups that we at Pusher have been working with, like known girls, women who go, who are providing workshops and free technical training just to help the next generation of women into coding careers. And at the same time, there's more jobs available and employers are also getting a lot more open about, "Interview is great, but we're also very open to employees. "It's time to go look. "We want more people coming through nontraditional paths." Because actually, technology is starting to become so important for almost every business. I'm probably getting way too enthusiastic here but I feel so privileged to have been on a journey through the tech industry. To be able to start seeing these changes coming through. I'm really excited about what's gonna happen over the next few years as we have more and more people, not just in lucky countries like here, but in emerging markets where you will start to see a lot more great local solutions to local problems. Basically, I feel absolutely privileged to be able to be working in area like developer tools. I never thought I'd say a sentence that sounded like. Just be court, because you really do feel like you're starting to be part of helping people to be able to build a new world, and it's great. So that was kind of my personal bit. Now I'm gonna do a quick little thing to explain why Alex is over there. So one of the things that for Pusher is really important is that it's in our DNA that what we wanna do is we wanna help developers be better at what they do. That's why we exist. But we also know that there's a lot of great meetups going on with great talks, great content. But if you can't make it, then you miss out. And we want to make some that stuff much more available to more people. If you're a meetup organiser, come and talk to us because we're always looking for more meetups to come and join us on this. 'Cause we just think it's so important to get this content out there and to make it more available to people. One of the things that we know is that the way that people learn development is that they go and look for tutorials to try and find the answers. So we're also encouraging people to write really great tech tutorials. So we've got a guest writer programme. If you like learning about new tech and writing about it, find us in there, apply, and we'll pay for each one that gets published and promoted so that we can also help people develop better apps. And oh, finally my last one. Also, obviously we're hiring. We're hiring a lot of people. So if you go to pusher.com jobs, we're looking for obviously developers because half our team are developers, but also people across marketing, commercial roles. Basically anybody who's interested in the tech sector, we wanna hear from you. And that is all for me and happy women's day. - [Woman] Hi, I have two questions. So you mentioned in your presentation that you have a network of 150,00 developers. So my first question would be, as Pusher evolves and it improves it's code, how does communication happens with this big community of developers? Because you can't just simply go there and upgrade the API. So I'm curious to know how you approach change and the evolution of the platform. And my second question would be, you also mentioned that you were trying to tackle local problems that developers have. So how are you collecting that data? - Answering your first question, which is effectively a communications question. We would be treating that as basic as emailing our developers, getting in touch with them in different ways. We make sure that we're pretty active on social media. So we communicate out in as many ways as possible. We are planning some really interesting product releases soon so that we'll start to become something we need to get some pretty interesting solutions for, but some of the things might be doing notifications in our dashboard, and so on. But yeah, we're planning for it. - Yeah and then the second question, it's something that's really challenging to do. I basically would have a coffee with any developer who will give thoughts. That's easier in London. I've done some travel to go talk to people 'cause I think when you sit and talk about the problems that people are solving, it's much better than call or an email. But generally, as part of my job, I'm just trying always to always be talking to anyone who's reaching out just in saying, "I'm trying to solve blank problem." And we're trying to get an understanding of like, is it actually that problem you're solving or is there some underlying abstract service that you need that could do this that then could solve 10,000 people's problem. So a good example is just giving people, reiterating in every possible way, that we want feedback on quite through these since they happen and getting intercom, emails, getting Zen Desk emails and all the support questions that we have. The most challenging things that is just, like language barrier's obviously. We have a staff of 35 people, so there's just no way that we can cover every human language in the world. But Google translate is actually pretty effective, especially with developer questions where we have stock answers and the code is all the same. I don't know, support team really enjoys the challenge when it comes in Japanese or in Portuguese, actually. - Portuguese is fun. - We have a lot of people who are from Portugal. But yeah, so that's kinda answering. - [Woman] Thanks very much to Amina and Jordan, thank you.