Act with Certainty

The ForeSee blog for CX professionals and the Voice of Customer community.

The Cross-Platform Mobile Problem

Alexei White is ForeSee’s Director of Emerging Technologies and author of the book JavaScript Programmer’s Reference.

As the smartphone market becomes increasingly fragmented, it’s important to decide if you want to choose a “write-once-deploy-everywhere” web-based mobile platform strategy, or one that provides better native usability and performance on each individual platform. There are advantages and disadvantages to each approach, but the tide seems to be turning toward web technologies.

With competition heating up in the smartphone world, the market is becoming increasingly fragmented. A recent Gartner study showed that worldwide third quarter 2011 smartphone market share was dominated by Android, with Symbian, iOS, and RIM left to fight over the scraps.

The ForeSee Blog Smartphone Market Share

This might be a bit misleading. When we think about the enterprise market, or even the urban, American consumer, we know intuitively that the story is quite different – Symbian is non-existent, and RIM still has a big presence. Indeed, if we look only at 2011, RIM has given up some ground to both Apple and Android, but this is hardly a sea-change. Microsoft is doing better with their Windows Phone 7, but is still being beaten out by even their own outdated platform, Windows Mobile.

There’s no clear winner here, nor should there be. The economics of smartphones don’t completely obey traditional network effects. Factors that drive consumers to one device or the other depend more on the intended use and the budget of the consumer than on how many total users that platform has. Gamers may prefer iPhones, while business users might be just as happy with an Android or Blackberry device. All three support Exchange and web browsing so there’s low switching cost unless they happen to be using RIM’s enterprise services.  In other words, if you’re waiting for a winner here, you might be waiting a long time.

How all of this plays out has big implications for companies who want to develop mobile apps. The development tools used to build apps for each platform are completely different. The cost of developing and maintaining your app is multiplied by the number of devices you want to support.

When I ask developers of iOS or Android apps whether they intend to support, for example, Blackberry, or Windows Phone 7, the response is generally that they are going to “wait and see”. The subtext is that they would love to increase their reach with a 2nd tier device but the economics are prohibitive. Instead, they’ll continue scanning the horizon for any perceptible shift in device preferences, and hope they’ll be the first among their competitors to discover the next niche opportunity.

This is actually a very widespread problem – and an obvious business opportunity.

First of all, one wonders how we got into this situation in the first place. There’s already a platform for applications that’s trusted and used by everybody, and it’s called the web. Given that the current generation of smartphones all have astonishingly modern web browsers built right in, it’s even more unfortunate that web technologies were not the prescribed way to build apps right from the beginning. One could point to the limited computing power of these devices as possible explanation (using a browser can be a less efficient way to execute code), but cynics could also claim that this was done deliberately to create incompatibility as a way to foster those all-important network effects.

There are now a plethora of options for pursuing a true write-once-deploy-everywhere strategy for developing mobile apps. Some of these are becoming very popular, even mainstream. The average user might not realize it, but companies like Netflix, Facebook, and Google have been using web technologies such as JavaScript, HTML5, and CSS3 in their mobile applications for a while now as a way to make supporting a wide range of platforms more manageable. Netflix has even gone to the effort of embedding versions of Webkit (Safari) on devices such as Playstation 3, Sony Internet TV, Nintendo Wii and the Logitech Revue to avoid having to maintain a separate version of their software on devices where they only expect a small number of their users to arrive. While the three companies I mentioned have the resources to solve this problem any way they like, there are plenty of ways for the rest of us to go cross-platform without breaking the bank. Here are some of the popular ones:

  • PhoneGap (Apache Callback) :PhoneGap is an open source and free framework developed by the Internet startup Nitobi, who was recently acquired by Adobe (full disclosure: I was a principal developer there for years). Apps using this framework are basically web pages housed in a native container, making them seem like native apps. It has recently been admitted into the Apache Software Foundation and is on its way to becoming a full Apache project.
  • Appcelerator Titanium: Titanium is an interesting marriage of web technologies and so-called ‘native’ development (meaning development using the widgets and user interfaces supported by the device out-of-the-box). Apps are written using HTML5, CSS3 and JavaScript, but output apps that have the look, feel, and performance of apps that were written specifically for each device. Like PhoneGap, Titanium has seen serious uptake at the enterprise level.
  • Rhomobile: Recently acquired by Motorola, Rhomobile takes a similar tack as Titanium in that is uses web technologies as the foundation to build apps that are in fact not web pages, but native apps using familiar widgets that adapt to each device platform they run on. Rhomobile also provides some back-end services to simplify the integration of enterprise data sources such as Dynamics, Oracle, SalesForce, and SAP.
  • Mono (MonoTouch / MonoDroid) by Xamarin: The recent entry of Windows Phone 7 has brought Microsoft development technologies to mobile. An open source project that has long had the support of Novell called Mono has been adapted to run on iOS, and Android devices. Now, the breadth and power of the Microsoft development stack can be brought to bear on virtually all major handheld devices except Blackberry and Symbian. The Mono strategy is yet another mixture of part native-development, part abstraction. The main difference here is that in the case of Mono, the needle is tilted a little more towards the native toolchain, making only parts of your application shareable between platforms. Anecdotally, about 65% of code is re-usable between all three platforms using Mono.

Choosing a Cross-Platform Strategy

Given all the advantages of using web technologies, why might someone choose not to go the Netflix or Facebook route and use, for example, a PhoneGap solution?

A big consideration is native look-and-feel. With a web page, you’re generally going to get exactly the same presentation on an Android device as an iOS device. Any attempt to mimic the device’s native look and feel generally falls a little short of being completely convincing.

Performance is another issue, making games not a good target for web apps. For business applications, this hit isn’t a big deal, but few would argue, for example, that Netflix’s iPad app even approaches the quality of a native app, and this can be attributed to the fact that it is really a webpage under the hood. However, with the performance of handheld devices improving in leaps and bounds, the quality gap between web and “native” is ever shrinking.

In the long term, the tide seems to be moving in the direction of web technologies. Here at ForeSee, we have so-far adopted a native strategy with our mobile apps partly because it’s important to us that we develop competencies in the tools used by our customers. Looking forward, it’s inevitable that we will be including more and more web technologies in our mobile roadmap.

We’d love to hear your feedback in the comments. For more on emerging technologies, subscribe to my posts or to The ForeSee Blog.


Categories: Mobile

About the Author

As ForeSee’s Director of Emerging Technologies, Alexei White leads the development of the company’s cxReplay, Mobile Web and Mobile App solutions. Alexei describes himself as a “software developer who is fascinated with everything web” and is the author of two books: JavaScript Programmer’s Reference and Enterprise Ajax.

Read more posts by Alexei White

Comments

  1. Eric Feinberg

    Fascinating. Thank you, Alexei. A clear elucidation of a question I often get when meeting with clients and prospective mobile clients – to app or not to app, that is the question. The fusion of web and native simply makes sense but I think the fusion takes us only 85% or 90% of the way there. To really make the user experience sing, I think companies will have to continue to invest in customizing the last 10% or 15% of the way there with lovely native techniques. Is that costly? Yes. Will it take unique skillsets? Yes. Will companies have to bring in outside help? Yes. But it will pay out bigtime in a satisfying UX. At least that’s my opinion now – things are changing so quickly and the need for agility in the mobile marketplace is increasing every day. Thanks for the great post.

    Reply
  2. Alexander Kolev

    Great article.

    I think that companies will continue to invest in native applications… Also I do not see any comments about local storage (SQL Compact / SQLite..) and working application without internet connectivity. How these two issues could be handled by the web app approach?

    Reply
    1. Alexei White

      Fortunately there’s a lot we can do with HTML5 API’s to persist large amounts of data between sessions, but its not as robust as a true file store. Losing your web documents stored with HTML5 is as easy as clearing the cache. I think when we write for web, the approach should be: work local, but sync periodically with the cloud. Stateless web API’s for syncing data can make this fairly straightforward.

      Reply
  3. Quin Lin

    Great article. Thank you, Alexei.
    I belive that the Web is definitely the right direction to go. In your opinion, among those open source mobile development platforms, which one you most likely pick to use if Foresee takes the Web direction? Thanks.

    Reply
    1. Alexei White

      Realistically, I see us taking a Phonegap/Cordova path mainly because we can solve the problem once for the web using a responsive UI approach, and then quickly adapt that for devices. As far as our off-the-shelf products like trigger and survey, we will continue to support whatever technologies our customers are using in house. This will necessarily mean maintaining multiple code bases for each product.

      Reply
Write a comment

Leave a Reply

Your email address will not be published. Required fields are marked *

*