 So the next speaker is Arvind Keshav Swami, he runs a software company in Bangalore called Deveta. They make mobile apps. One of the things they're working on is an app that allows you to practice current music through your phone with a girl who is real and working with you. But the entire interface has been created through a phone so that you don't have to be stupid or impressive. It's an interesting idea, I hope Arvind speaks for that when it comes to that. But Arvind's expertise is on both iOS and Android. Deveta brings mobile apps on both platforms. And in this session, he's going to begin his comparison and say, let's look at and write from the perspective of where it stands compared to other platforms. But on the other side, this is the other leading platform. Arvind is also the chair of the program committee for Android conference. Arvind is a series of conferences on Android around the world. It started off in Germany. They're now on the fourth year. Then the London edition is also on the fourth year. As of this year, this is an edition in Spain. There's one in Belarus in the eastern side of Europe. And the leading edition is the first outside of Europe. So maybe in the second edition this year, last year, we got about 500 people. And this year, we think in later, but later in the week overall, two days of sessions, three pallet drags, or anything in the damn bag. Arvind's team manage the schedule for the day. So it's their responsibility to understand what is the profile of the audience that's coming in. When you come in as someone who wants to attend a conference, you expect a fewer logical order. And deciding the logical order is what Arvind does. So that's his role in the events organizing. So today, he's going to be talking to you about Android in the context of the mobile ecosystem. Thanks, Kiran. Can I get you a name, please? Yeah, fine. So, well, I live in Bangalore now. I share a Chennai boy. I'm going to be in the Gopalapur, not Gopalapur. So I'm growing up in the 80s, early 90s here in Bangalore. I share a Chennai. Bangalore is the hottest place to be. It's the mall to come to, to hand out, to be seen and seen with kids from most of the colleges we have had. And it's amazing to me coming here and giving out some more after almost 15, 20 years. You see, this is a hotbed now for startups. So I think just the guys are running the startups. I think you're doing an amazing job. And also the guys here that has geek. See, I moved back to India about 10 years back from the States. There were a few startups there. I helped one of the startups there. I set up an India office here. And one of the things I have done 10 years back, right, was really hard in India. We have conversations with tech people and people in startups to discuss ideas, to bump things off and learn. So 10 years back, most young people, this was a world broad and they wanted to get on site and think there was a big space there. But the fact that you guys are here, I'm in Chennai on a rainy Saturday afternoon to come here and discuss jobs and enjoying the green. So first of all, thank you to all of you. And thank you to the tech guys. They're doing a good, credible job of just bringing together tech people. So in keeping with the startups, start tech conferences, I'm just going to be talking today about new mobile frontiers. And quickly, I want to ask people here. Just want to ask you a few things which is raise your hands. Tell me how many of you have filmed Android apps here? Excellent. Very good. And a few people who have iOS apps as well, or mobile apps as well. Excellent. Very good. So thank you guys. I'll sort of try to keep this a little interactive. I might sort of warn you guys we're going to both speak back as we sort of go on through this. Part of the objective of the session like this is so that you guys know what people are doing out here in the local ecosystem as well. And you know, this platform is like drawing con, which is happening in Nalanga. We're going to encourage people here in Chennai as well to come forward to come speak for things that you're doing to share our learnings. Since what point is this sort of community vibrant by learning from each other within the Indian context here? And to build our work together and to share it into those various companies that are either doing services or building our own products. And I think we have a lot to learn from each other as we do. So some of these things which many of you here are quite familiar with if you're closing into the mobile world but I'll cover them properly anyway. But first, just a quick question. Can anyone here guess of all the smart phones that sold worldwide in Q1 this year? What percentage of these phones were Android phones? Just a rough percentage. 60, 50, 60. 60? Close. I don't think so. Excellent shot. There you go. So let's get close to 55. And so what's remarkable about this, right, is you compare the numbers from 2011 up to 2012. The Android market share jumped from 36% up to 56%. So that's a pretty remarkable leap, right? And you can see enough of this on TV, you see ads for Samsung phones and others on TV, right? So the point I made here is you don't even need to be a mobile developer. You just can't be more of an Android anymore. It's just exploding. 56% today. This number a year from now who knows what it could be. So as a platform, you can no longer even if you're building websites. Clearly enough companies are talking about how mobile at first needs to be the strategy. Even if you're building a website, you've got to think about Android at first going forward. And another factor here is just operating system versions, right? So for people who have been around sort of building old desktop apps in 15 years back, if you're familiar with all the complexities of dealing with different operating system versions and complexity that brings. And a lot of that's there in Android too. And if you look at these percentages, right? These are the... So we get a problem here in Android versions are sort of as particularly named. So we have Gingerbread, Honeycomb, Ice Cream, Jelly Bean. This is sort of the order in which Google names them. And the point that I'm trying to make here is Gingerbread, which is now a couple of years old, is still more than half of the Android operating system. There's a number of reasons in this case. Part of it is the OEMs when shipping out updates to the older phones. A bunch of other problems around that. But the complexity this brings in, right, is for an Android developer, you sort of need to think about it and say, you know what, should I be targeting the old phones out there, right? And to have things work in compatibility mode and either progressively degrade, you know, functionality that you make available based on device capabilities or should we print out some of the newer phones? Right? And Ice Cream Sandwich, which was released late last year, is still only one quarter of the total number of Android OS out there. So while this number is increasing, right, if you compare this with iOS, that's an example, most Android users upgrade to the latest OS within a month or two of it being available. So 80% of iOS users are on the latest version of iOS. Now Android developers, we just don't have the budget, right? We've still got half of the older versions. Here's another chart which sort of shows a trending line over the last several months. Since Ice Cream Sandwich started up here on a lot of newer phones and since Jeremy came out, then you can start to see more and more people are getting on to 4.0. But this is a huge chunk. That's not 2.0. Number of Cadillac is this slide that represents percentages. It's not absolute numbers, right? So in reality, I mean, these numbers probably haven't changed a lot. Because Android 2.0 is still on Android 2.0 and I haven't got a brain. It's just the total numbers of 4.0 to the 2.3 are probably in the heights. Now, similar to all the issues we face with different browser versions that we spoke about earlier today's meeting, a lot of this complex-based compute exists on Android 2.0. And just to, you know, these weren't enough. There's also complexities around size and density. And so the first one is respect to the phone sizes itself. It's physical size. And Google classifies these into small, normal, large, and extra large. So the small phones are with LG Optimus or 240 size phones. The normal size phones would be kind of closer to your 3.2 inch size phones. And then we have the larger size phones, which would be like Galaxy Note and others that you see, even holding it, but please, more like a larger brick, but, you know, it's not larger. And then you've got the extra large phones which are closer to the tablet class, if you will. And let me show you the size. You've got density to worry about, right? Since iOS Apple came out with their Latina phones, increasingly one and more than Android manufacturers are going into the XHDPI category, coming out with very high pixel density. So you could have a phone that's, you know, just 2 to 3 inches in size in the future, but its resolution could be 2,000 by 1,000 in the future. And you need to deal with it. You need to plan for all the pixels. Your artwork will need to be ready for that. So increasingly, right? These things are shifting. There's clearly a trend towards XHDPI. You start to see more and more of XHDPI in the future. Whether the start to see more of large versus small versus normal is still not totally clear. It's clear that I think in last year Android tablets didn't do that, right? Motorola Zoom came out and it was a disaster. The Galaxy Tab was okay, right? Until Apple started to sue them. And then just with the summer, since the Nexus 7 has come out and it's got a fantastic price point, it's pretty clear that the tablet's face is going to change quite a bit in Android. So I think we can start to expect a lot more growth with the extra-large side of this price. So one of the things I want to convey here is, you know, I hope that the Android developers talk about fragmentation and how it causes so much pain. The thing is, you know, we've got to stop the playing about it. Android is more fragmentation. And fragmentation is sort of the second side of the coin which is a beautiful diversity that Android offers. The fact that you have the power of choice to be able to choose from a whole wide variety of forms of different sizes and you can pick and choose what you're buying is a beauty, a diversity, and a freedom that will confuse you. And fragmentation isn't going away. It's something we should plan for, figure out what the best practice is all to deal with it and how the approach it makes from targeting making certain phone sizes to certain phone resolutions and then taking a certain approach in the right way. Now, here's just another quick data point, right, which is just data from the OpenSignalMatch app. So, what OpenSignalMatch, that is, they have data from all their users and put it together under infographic which shows you what percentage of the users use devices from each of the hardware vendors. And as you can see, Samsung's a clear winner, right? I mean, there's more than half the forms of their name in the Android world which are all Samsung. Much further behind, come in HTC, SEMC, here at Sony AXL, and then there's Motorola. Now, there's a whole bunch of other Android manufacturers out there from Motorola, from China, from India, and so on, but they have a much smaller percentage. The clear leader with respect to profits is Samsung. In the other hand, they're much, much further behind. Now, how does this happen to you as an app developer? I think when you plan to make those choices with respect to which devices you want to test out your app on, you can easily pick Samsung, because both of the users have it. Now, where does it matter if I can use the app? Let's put stuff within our rounds. Now, obviously, each OEM round on top of it adds little nuances, the keyboard behavior is pretty different, little things that you need to work with, but you need to play with it. But having a sense of what this distribution is, lets you plan and apply your app for those fragmentation problems. So, you have an idea of what your target user is, like which devices they're using, and you can make some choices. So, pick a couple of Samsung devices and then sort of choose the main HTC so that it's more roll-out of what you're going to do. And once you've got an 80% case covered, you can kind of go from there once you get to some critical mass with respect to user base. So, let's see what we're talking about in monetization. And I sort of want to make this a little interactive. I'd love to hear you guys sort of talk about your apps, what your approach to monetization is, what your approach to monetization has been, and any challenges you've had around the way. I'm just going to quickly share just general insights, things that I've heard, my experience, things that I've heard from people, replicating your thoughts to on it. Of course, iOS, right? Two thirds of Android users don't actually pay for apps. There's been a lot of documentation around there, people that remember the surveys. In general, most of the Android users prefer free apps, but in most cultures, prefer free apps. That's not a joke, that's not a lie about it. But there's, I think, there's a bit more of a tendency in Android where to sort of go with a free app versus thinking about going with a paid app. So, the challenges poses for developers are how do you monetize this, right? And this ultimately is a small set of monetization models that can go with a mobile platform, unless you've got a service that you're running which expands out to the web and other platforms that you can monetize as well. So, paid apps, they appear to be a good monetization model for you, right? There's only so far you can go with them. You can look at advertising-based models, right? But from every one app spoken to, in general, CPM is the highest for the iPad. They're sort of a lot lower for the iPhone, and then Android is a little further behind it. CPM is a little higher if you put an app that has a media, right? It's not less satisfied the apps and things like that. So, while it's a model, especially with Android, here you've got much more scale that you can get out of it. Just honestly, the number of Android users is high, right? And if you're building an app for India, right, much more Android users in India than Apple users. That number is just really growing given that Apple's pricing media's are taking this way, right? So, that number that we're going to go up and I think just in terms of the sheer number, India and China have a lot of opportunity there, right? Just a number of users to get these models that you can look at. So, it's not really possible in India if you're here, but not Android apps, if you have any stories you want to share about monetization models, if you might do it now, that's fine. I'd love to hear from you guys if you have any feedback, things that you guys are doing. I think the font here is a little small, so I just want you to take this piece out. This is about in-app purchase. So, if you're familiar with in-app purchase, right? In-app, for the certain things you can buy within the app. So, people can try the app and then certain modules, like for instance, if it's a game, you can unlock a certain level of pain more. Or if you're buying an app that has music, the app itself could be free with some tracks free, and then you could buy additional music tracks. So, in-app purchase, right? Clearly over the last couple of years, it's become a more popular model through which an app isn't purely an ad-driven. You can monetize it either by tech people or, you know, you can turn on ads if you pay so much or unlock or subscribe traditional things by paying so much. Now, with this, right? Some stats have shown that 22% of the people if they like an app right in the first session, they go ahead and make an in-app purchase. So, that's about 22% of people. But, 33% of them go through between two to nine sessions with the app before they actually are going to purchase it. And 44% of people use the app at least 10 times before they're willing to pay to buy anything. So, the thing is, user engagement is important, right? Two, three years back in the mobile space, you could build if you could be a part-app if it's a little funny and you'd get away with it because it's better to buck out to a traditional entertainment fact. Those are the robots, right? So, now we're in 2012, and users are looking for something that is quite a sustained value in some way. People are a lot more concerned not more than apps that are popular of their apps free. So, if you're coming out with an app that's new, that's different, it still needs to resonate with the user. So, driving this, you know, average number of daily users, monthly users is a big deal. The same sort of tracking you would do with websites starts to become just as relevant. And apps no longer are standalone. The main role of app developers a couple of years back is just build on an app put it out there and try to monetize it. And that equals to a strange approach. Now, with respect to monetization, let's give you a second view of how this is connected. Well, this whole native version of this came out by debate. There's a lot that's talked about it if you guys are probably familiar with it. I just wanted to share some of my experiences and thoughts that I've had. I wanted to give you thoughts too. I'd like to share some of the things that you guys have done. So, clearly, right? I mean, mobile apps are a bit popular. People are growing in energy. So, in the days with desktop apps back in early 90s, you know what, there's just a flag, people will build these apps for a year or two and then it's done and move back to the web. And let's just step back to see how this whole mobile app thing happened. So back in 2007, right, when Apple launched the iPhone, initially they did not offer SDK for native apps. Actually, all they offered was the web SDK. So people could build web-based apps and they really were pushing extremely hard to build a key part of building a web kit and everything else. So, but there's a huge approach in developers in that time who said, you know what, no, it's not going to happen on the web. It's difficult and web connections are slow. We want a native SDK. And then six months after the iPhone was launched, Apple was informed that they're providing a native SDK. And the unusual thing that happened is two things happened. One, people just took off more than anyone expected to. And people started building apps in the sediment race, right? So the whole native ecosystem just moved. Now, since then, one, the networks have gotten a whole lot better, with things like an E, 4G, all over the place. Two, HTML5 and CSS3. The types of things people are doing today are just some of the folks to demonstrate even before this session are increasingly completely changing the need to build native apps. So, where earlier people looked at things like Flash, to build their apps would be responsive and work in a consistent way across different browsers and different operating systems, more or less now have big issues with them. Now, however, you know, this is sort of one of those questions people often ask me. It doesn't mean that a couple of years down the road, we'll never need native apps. And it's always hard to say, right? It's like, all those questions of when the TV came out, would people stop visiting the radio? Right? We still visit the radio and still watch TV. We've got a whole bunch of different news sources. So, some of these things might exist and they may be relevant in certain scenarios. And I just wanted to talk through some of those since I think the area where we use one versus the other as things to happen and the talk process behind it is quite significant. So, in my experience my experience it makes sense in choosing a native approach. And just to clarify, when I say native it doesn't make exclusive environments without them. You could be using both in conjunction, using a web page, right? You might choose to build out a native app and container share with some things that are being built with your native modules and some things that are using a web view and web pages and either local web pages or a remote web page using HTML, time management, fire, or that sort of approach. Now, do you mean that you're looking to tap into a lot of inherent device capabilities? Do you think that if you were an app that's an augmented reality app obviously it makes the most sense to be completely native? You're trying to tap into device capabilities that will be a lot slower and in some cases it's just not feasible at all to do well across. Second, now the nature of mobile devices, right? There's been a lot of spend like you're in the elevator you just quickly open a Facebook, check your status, you turn it off, you go back. So the attention span of those users is very small. We pretend to use these mobile apps for a minute or two in between and don't often use them on a sustained basis versus when you're back in a desktop or laptop we use Facebook for about 10, 15 hours maybe more to the view. But what I'm trying to say is the need for having a design is a response for the user to quickly put it pretty much where they left it when they stopped using the app earlier and building it in such a way it was interactions are quick. It's a big deal. So the challenge most mobile apps developers have had is building these apps with their views which you can find at times as responsive as this was in the app and using it to get it straight. Now while something is changing, right? I think it's still a good idea to think about areas within your app where the user might and see if you want to go native with those especially around navigation because especially on a navigation user's frustration can be if it doesn't feel like a UI's response. So overall navigation scheme it's with the key buttons within the app and things where the revision indicated often I feel that's useful to stick to keeping these pieces moving. The third piece is they don't automate people refer this whole native experience and there's a number of libraries out there. I think one of the earliest people referenced 4GaP and a titanium app separator there's Kirin there's a bunch of other libraries, platforms there's DK out there that try to offer your super bulletin says use these and you can have a common core base that focuses on different platforms that focus on just certain core areas of business that you need to build up and all of these have limits and especially having an enterprise app space has a lot of value since I think they fit the pattern of enterprise apps quite a bit where I think just the UI patterns are a little more simpler and I think it fits better but with some of the consumer apps when people look at an on-off controller or a slider they expect the same experience that they have from Google's apps or from Apple's apps and if you put the UI paradigm in your app in other words you aren't trying to build a web-based app which looks like a native app you're probably okay because user expectations aren't the same but if you try to mimic it and the user sees that you know what what you're building doesn't feel in data users tend to get a little frustrated so I think that's another decision point that I think is often relevant now where do you sort of need to estimate I think one is it's a speaker I think as one of these other speakers said you know a web-skills are commoditized while building good web apps it's still hard and requires people with people skills it is easier to get developers after-speed and building apps with HTML5 with JavaScript and less in CSS versus Objective-C and no matter what you say Objective-C other functional is probably more functional in nature than JavaScript without building apps but it still isn't okay in some of the motions that it uses and the medic says it uses and going out and building software with Objective-C and export this hard although Apple's done a good job of taking memory management out of the equation it's still this difficult but it's still Android right building lots of these apps with Java can feel a little you know you can get sort of a feel if you've got to be so much easier right and in some of these cases just the speed of being able to build this web with HTML5 is great right and you've got two choices with the web either you can have these HTML5 pages built and then put them in the app or you can have it delivered more you can bridge the pages you can catch some of the pages add a remote web page load local resources with a whole bunch of clips you can have and go to Apple and offer a link to that bridge through which you can share objects between Java and the Java side or between Objective-C and the Java side iOS as well and ultimately while both of that companies sort of need to work in an engineering cycle I've got a month long sprint cycle and I hope to build it up really most web companies use much shorter cycles right you probably do a one week cycle and might do it and it releases that sort of which is a couple of times and with HTML5 we have an opportunity to integrate a lot more so you can release early you can release often and go through the cycles a lot more and there's just several things that I think the web does that much better than native both in iOS and Android and especially when it's moving layouts more different ways text inhibitors it's just so much more work you do natively on either iOS or Android but these things are much more much better and those challenges of getting different different device sizes and their layouts are even a bit the fragments that Google provided since the last it's decade at least it still puts a lot of work on the development whereas doing these things with CSS like media queries simplifies various aspects in those so I think you know I think the bottom line maybe is it's not one size but it's all I don't think you can say you know what if a shiva guy is best or maybe it's best we've got to look in use cases in the app you need to look at and say you know what there are some portions where all you're doing is you're rendering things to meet display there's no user attraction if there's no user attraction I think it's a great candidate for a shiva guy if there's areas of users interacting with it a lot I think that's a good case looking at it or is it something very important sort of longer cycle so I think it's not one size it's all I think it's a mix of both and I don't think maybe that's a very good idea soon but I think clear indication that the trend let's go with a shiva guy with all of these it makes more sense to invest your engineering resources there and trend when it makes more sense so user feels at home on the mobile platform that he's used to like if I'm a user I'd like to have to feel at home not like a webpage if I'm an iOS user I'd like to have like an iOS app and I'd like to feel at home when it costs some of the time so I can keep on with being on such a team going there no brawn is going before that's the first thing that I'm doing with sorry thank you sorry but keeping the team today Android is being used in places that we would never ever imagine people are using it in robots people are using it in televisions people are using it in fridges there's an app which Yahoo recently came out that I acquired by the company which built this where you can just set up your phone you can provide this app to the TV and turn on the TV and start to watch your favorite channel and the app and the figure out which TV channel you're watching which TV show you are watching and give you information about the TV show the characters so the types of things people are doing with Android today are just amazing it's being used in a number of different areas and I think the sort of opportunities we have in a way to the ecosystem it doesn't work one of my friends he's working on a startup in Bangalore where they can come up with tablets in the education space for teachers and children and they can source a low cost Android tablet from a Chinese manufacturer who is willing to give them a great tablet for roughly about 2,000 or 3,000 rupees and they don't get a cheap tablet I mean a good responsive tablet with a screen and everything else so the fact is the device the hardware piece of it is so commoditized you can get really cheap hardware and if you can build good software on top of Android the potential and the things you can do with this huge smart homes you can imagine Android running in your French in your poster in your electricity meter so that you know exactly how electricity used to be used what's in all of the neighborhoods in your apartment and everyone else over here they compare the potential is huge it's mind boggling and a lot of these demos have actually been coming together ad write-on so last year at write-on one of the engineers demoed a robot on stage nothing I do know so this year is likely I think with this one session proposal was discovered and has been funded for write-on but I was looking to show in it coming with an Android app there's a pretty interesting app demo that we have on write-on so that's my shameless product for write-on because you guys should look at the write-on site and see what's coming up I thought I would give you guys just some trivia something right here something right here yeah what a case right but this data from Google just makes some people accessing the Google Play Store that's very much I'm just going to record it for a time zone record for a time zone yes absolutely a lot of wizards just popular this place here see if you guys can identify as who this person is a little clue a little bit of a legend around him but he did not influence Apple actually I'm a wizard we want to have a second we should have others go on anybody my question is Alan Turing very good Alan Turing it's 100 years with Alan Turing a man who gives a Turing machine mathematical models that I think have since then it has come out with all these beautiful devices all over the place 100 years of Alan Turing and the little legend of Alan Turing is that he committed suicide Alan Turing was gay he committed suicide after a lot of persecution in Britain back then and he ate that nice apple and one bite of an apple there's a bit of urban legend about influencing the Apple logo with the Turing machine for the next session so and we have anything else I think the mobile space has just changed so much it's hard to predict where things will go I think it's been incredibly exciting I welcome all of you for going on and those of you who are a part of mobile ecosystem here in Commonwealth for our session I think it's a great opportunity to share what people are doing and develop with each other and you know I think in the next one year we'll see a lot of new things in the space with the Windows 4 coming out the new updates are coming out later this year should I think open it up potentially it's proved to be a challenge to Android as well it's something we need to wait and see but I think it's an exciting year so same receipts there's a great talk for ISM coming up right after this