Cross platform mobile app development methods for iPhone, Android, Blackberry and Java

Cross-platform support is one of the greatest challenges for developers, advertisers and service providers that offer their service through mobile apps. During the past 4 years we have evaluated the pros and cons of different cross platform development methods and tools and assorted the key advantages for a company from an app development point of view. The main methods and conclusions are presented below.

1. Web runtime apps (sometimes referred to as widgets)

For web runtime apps (sometimes referred to as widgets) the development is done in script languages typically using HTML, Javascript and CSS. Later they run within a shell on each platform (iOS, Android, Blackberry OS, etc). Solutions, such as Phonegap (adaptations thereof and other in-house tools) using this technology will only work for simple apps and are always behind in terms of supporting new OS versions and devices. The advantage is write once / run on many but the disadvantage is that most proprietary features won't work, e.g. pinching on maps on the iPhone. This is the lowest cost solution other than just developing a mobile website.

2. Source code porting

Source code porting means writing for one platform (e.g. Java) and then cross-compiling the code to other languages or OS. Innaworks is an example for a company using this technology. Although it looks very promising at first, we think that this is primarily a solution for game developers that don't require proprietary platform support. The advantage is that the developer can stick to one code base. The disadvantage is that the user experience usually becomes a compromise and that proprietary platform functionality is not supported.

3. Native development for each platform

Native development means to develop the app individually for each platform. Usually the iPhone, Android, Blackberry and Windows Mobile development will be done by different developers as each platform requires specialised skills and experience. The advantage however is that the apps are optimised for the best experience and
to exploit the functionality on each platform. The disadvantage is that, for each platform version of the app, the developer almost has to start over from scratch. This generally means very limited synergies in terms of cost and lead time. Even worse, the maintenance of the multiple platforms become more and more expensive over time.

4. Component based cross platform development

What developers want the most are tools and methods to make the native development and app maintenance more efficient. Golden Gekko has developed a proprietary platform called Mobile Media Engine (MME). It simplifies the development by though components for common challenges such as memory management on iOS, screen size scaling on Android, User Interface on Android (similar to iOS but with platform specific advantages), keyboard input on Blackberry, etc. The components allow the developer to use similar coding conventions across all platforms which vastly increases the quality and speed of development. At the same time it reduces the cost of maintenance when updating multiple platforms. Currently this platform is only used internally but in the future we plan to open this up for partners and third party developers.

For more information about the cost of mobile app development for each platform see our previous blog on the costs of developing a mobile app.

Recipe for success with mobile apps

I'd like to take the opportunity to update you on what's being going on in the mobile space during the past 6 months and some exciting opportunities going forward. I've tried to pick out the areas that are most relevant for you from our massive knowledge base consisting of over 150 mobile apps and services launched in the past 6 months.

1. Why brands should launch branded mobile apps?
2. In mobile fragmentation is forever. Deal with it.
3. How to choose mobile platforms (iPhone, Android, Blackberry, etc)?
4. Why is mobile web not enough?
5. Producing your own apps using Tino Mobile


1. Why brands should launch branded mobile apps?

Over the past year there has been an enormous hype around iPhone applications in particular and mobile applications in general. After 10 years of uphill struggle, mobile marketing and services finally took off with a big bang. Companies ranging from global consumer brands such as Coca Cola, Mercedes, Kraftfood, Nike, Zara and Pizza Hut to small and medium size organisations including festivals, restaurants, local touristboards, hotels and publishers.

An overwhelming 64.8% of marketers and publishers reported planning to invest in mobile apps this year, according to a December 2009 survey conducted by DM2PRO and Quattro Wireless. The most successful from a reach perspective are no doubt those with a great entertainment value (e.g. iPint, Barclays Waterslide, Volkswagen Polo Challenge and Malibu Bowling) while the ones delivering the greatest ROI are applications that provide a relevant
service to customers (North Face Snow Report, Yellow Pages, Timeout sponsored by Smirnoff, Absolut Vodka cocktail guide, etc). 25.000 people using a branded application every week potentially generates a lot more value than half a million people downloading a game which they play ones.

Rory Sutherland, Executive Creative Director at OgilvyOne probably describes it in the best way when talking about "mobile" advertising: "Never dismiss branded utility. It's a lot easier to be repeatedly useful than repeatedly funny."


2. In mobile fragmentation is forever. Deal with it.

A couple of weeks ago TechCrunch had a great article about mobile fragmentation by Richard Wong at Accel Partners. The article (http://techcrunch.com/2010/03/04/mobile-fragmentation-forever/) highlights the fact that there is no long or short term solution to fragmentation and nor is there a simple answer to which platforms companies should invest in. As one of the pioneers in cross-platform development Golden Gekko has invested heavily in R&D and education of customers and partners. We have managed to make it easier and more cost efficient to develop and maintain cross-platform apps but there is still no simple solution. Multiplatform development platforms work for simpler apps but in reality there is no out-of-the-box solution if you want to deliver the best experience for each platform.
Our recommendation is

1. Plan for cross-platform support from day one unless you are absolutely certain that you only want to launch on one platform - fixing it later is very expensive
2. Ensure that your development partner has the experience to develop for multiple platforms and that if an porting tool or multi-platform tool is recommended that you understand the limitations agree on processes and scope for design, testing, acceptance, appstore uploads and other distribution early on for each platform as this will save you a lot of time and effort throughout the project

3. How to choose mobile platforms (iPhone, Android, Blackberry, Java, Widgets, etc)?

The number of mobile OS platforms and development SDKs keep on increasing with new platforms including Samsung Bada, MeeGo, 5+ different widget platforms, Palm Webapps, Windows Mobile 7, etc. Which platforms should I chose?
We think there are 4 key considerations:

1. What is your objective? PR or Mass market reach?
2. Who is your target audience? Markets? Niche segments?
3. Where will you get the most attention for the least money? Many operators and handset manufacturers offer very attractive distribution deals.
4. How much are you willing to spend?

Our platform recommendations vary from case to case but our general advice is:

• PR, communication and experience - iPhone apps
• Innovation & controversial - Android is the first choice as there are few restrictions in terms of APIs and approval
• Youth market - J2ME for non-connected apps
• Mass market reach - Launch on as many platforms as you can afford starting with iPhone and Android
• Business users - Blackberry and J2ME

Android or Java are our preferred base platforms for cross-platform development as there are clear development synergies between J2ME, Blackberry, Android and e.g. widgets whereas iPhone development has limited synergies with anything else. One of the main limitations with J2ME today is that most users won't connect to the network due to issues with settings, data costs, security prompts, etc but the number of downloads for a like for like iPhone and J2ME app is still 5x to the advantage of the J2ME app.

Finally the other key considerations is distribution and media spend. Many of the handset manufacturers and operators are desperate to get new innovative apps on their appstores and will therefore give you all kinds of free promotions and traffic. A free promotion by e.g. Nokia OVI could easily be worth 50-100k Euro in media budget and will give you a lot more reach and downloads than an iPhone app would without putting the same kind of media budget behind it. We are currently getting similar offers from Vodafone, O2, Samsung, RIM/Blackberry and various other partners so it's always worthwhile discussing these opportunities.


4. Why is mobile web not enough? Consumers are disappointed with mobile web experience.

A recent survey of 1001 mobile Web users found that users have high expectations for mobile Web performance. Ultimately, the majority stated they expect Web sites to load as quickly, almost as quickly, or faster on their mobile phone, compared to their home or work computer.

The survey "Why the Mobile Web is Disappointing End-Users" also revealed how unsatisfactory mobile Web experiences can negatively shape a consumer's opinion of an organization. In the survey:

• 60 % of respondents have encountered problems when accessing websites on their mobile phones in the last 12 months
• More than half are unlikely to return to a Web site that they had trouble accessing from their phone, and 40 percent said they'd likely visit a competitor's mobile Web site instead.

What do we think?

The mobile web is great for text based services like news, search, directory, etc but as soon as you want an interactive experience with maps, embedded video & sounds, location, animations or anything more advanced then apps are currently the best solution.

For more information about the survey see a summary here http://www.slideshare.net/Gomez_Inc/gomez-mobile-web-experience-survey-why-the-mobile-web-is-disappointing-end-users


5. Producing your own mobile apps using Tino

Finally we want to give you an update on our self-service tool for mobile apps Tino Mobile. Tino allows you to produce your own mobile apps with more standardized features. This does not replace bespoke development of innovative new apps but allows all kinds of great apps to be produced.

During the past 3 months we've continued to enhance the tool with

• New features including carousels, embedded video and audio, questionnaires, more flexibility for image/text formatting, SVG fonts, banner ads and lots more
• Major UI enhancements for larger screens
• Enhanced device support for Blackberry
• Making the online tool easier to use

During the coming month we will also introduce support for Android and iPhone as well as a myriad of other features.

If you want more information visit http://tinomobile.com or contact us for an online demo.


Magnus Jern, CEO