Mobile Phone Development

Custom App Stores

Simon Judge, Android Developer — Fri, 03 Feb 2012 12:27:53 +0000

As companies develop more and more apps for their employees, customers and even suppliers it soon becomes problematic to control who has access to apps, updating them and gaining reporting information. In such cases it becomes prudent to have a custom ‘enterprise’ app store that…

Restricts which users can download what apps
Ensures a device has the correct capability to run an app
Notifies users of updates
Provides reporting of who is using what and how often
Allows instant approval of apps

All this is possible on the dominant iOS and Android platforms. White label solutions such as Nukona have already started appearing. However, on iOS it’s only possible to distribute to employees and contractors rather than others such as customers, resellers or suppliers, due to onerous terms in the Apple enterprise developer license.

I see enterprise apps as an area of opportunity. Don’t just think enterprise but instead any organisation or group of people that want to control and limit access to apps. This might, for example, be large brands that create many apps, hospitals (mHealth), hobby organisations the publishing industry or anyone publishing lots of apps.

If you are interested in this topic, Motorola Motodev also has an interesting podcast on distributing apps to mobile workforce.

Amazon App Store More Discoverable than Android Market

Simon Judge, Android Developer — Thu, 02 Feb 2012 14:52:38 +0000

If you are developing an Android app, you might be wondering if it’s worth putting the app on app stores other than the Android Market. You might also question whether it’s worth developing an Amazon Fire specific app for the Amazon Market.

Ziggy’s Games has some answers to these questions based on downloads of their free Bus Jumper app…

It can be seen that the Amazon Market has now overtaken the Android Market and even SlideME has respectable numbers. Why is this? From empirical evidence from my own and my clients’ apps, I believe the Android Market has become so large now that it has become near impossible to accidentally discover an app. Conversely, the Amazon Market is smaller and together with the Amazon Fire’s popularity this means that apps are more likely to be discovered.

I also believe SlideME is gaining in usefulness due to the larger number of non Google Experience devices being shipped, mainly out of China, that don’t include the Android Market.

For developers, this means thinking more about distribution beyond the Android Market. This introduces new complications in areas such as device capability, app update, purchase authentication, reporting and in-app purchases.

For Google, this means it’s losing grip over developers in terms of lost commission and standardisation and control of the app ecosystem. Google can remedy this by improving app discovery and making it easier for smaller OEMs to produce Google Experience devices.

Related Articles:

Codename One

Simon Judge, Android Developer — Wed, 01 Feb 2012 13:07:18 +0000

An area that has shown promise but limited success is that of cross-platform app generators. The majority rely on HTML and Javascript and, as such, the look and feel and performance (particularly scrolling) is inferior. I previously posted some tips that identified areas to think about when assessing app generators.

My personal belief is that it’s necessary to move away from HTML and Javascript in order for these tools to be suitable for professional use. Mosync is one such tool but, looking at their latest advances, they seem to be moving towards rather than away from HTML/Javascript.

A very new tool/platform is Codename One which has origins in LWUIT used to provide a contemporary UI on standard Java ME feature phones. Some of the LWUIT team have adapted it to work cross-platform allowing code to be written once in Java and work on iPhone, Android, Blackberry, Windows Phone 7 as well as J2ME devices.

Some features (from their blog)…

Full access to the underlying OS including embedding native components in place.
A Swing like API geared for mobile devices with support for IO, GPS, Media/Camera etc.
Native Themes - Codename One ships with native device themes that are implicitly set to a project to ease the integration with the native OS.
NetBeans/Eclipse plugins.
Simulator environment - Codename One provides a fast, easy to use and flexible simulator environment.
Codename One Designer - cloud service to seamlessly build your applications for free. You can also compile locally and avoid the cloud service entirely but need a Mac for iPhone builds and Windows for RIM/Windows Phone 7 builds.
Cloud services - Codename One offers cloud services ranging from crash protection to on device debugging and access to web services, push notification and location.

While I haven’t used it yet because it’s still in beta, Codename One seems to have potential. I read in the FAQ that the Java API is CLDC 1.1 which is going to be limiting. I also wonder how the UI will adapt to say iOS vs Android idioms. For example, placement of tabs and rendering of lists which needs to be in native-like ways for an acceptable look and feel. Too many existing tools are encumbered by strange subscription licensing and I like Codename One’s free model. However, I question who will actually fund ongoing development as mobile platform APIs evolve and Codename One needs to keep up to date.

There’s lots more information on the Codename One blog.

Related Articles:

Mobile Internet and Smartphone Adoption

Simon Judge, Android Developer — Mon, 30 Jan 2012 19:58:18 +0000

Google has some new research performed by Ipsos on Mobile Internet & Smartphone Adoption (pdf). The research covers the United States, the United Kingdom, Germany, France and Japan.

Some interesting conclusions…

More consumers now use a mobile phone (feature phone or smartphone) than a computer (desktop or laptop) across all five countries.

The demographics of smartphone users are also shifting, with an increasing percentage of women and those aged 45+ adopting smartphones.

For developers, this means that it’s making increasingly more sense to develop for mobile first. The changing demographics also means that different types of app now become more viable.

Global Tablet Shipments

Simon Judge, Android Developer — Fri, 27 Jan 2012 13:13:01 +0000

Strategy Analytics has some latest statistics on global tablet shipments for Q4 2011. Global tablet shipments reached 26.8 million units for that quarter which is an increase of 150% year on year.

The iPad saw a contraction of market share that was taken up by Android tablets. While Apple iPads are still more popular, I believe it’s only a matter of time before Android shipments exceed those of the iPad. However, I also believe forthcoming Windows 8 tablets might damage both Apple’s and Google’s market shares.

What does this mean for mobile developers? As has happened with smartphones, I believe we will see many iOS app titles being ported to Android. We will also see side-by-side development of Android and iOS rather than just iOS. Some people will go for an Android first strategy as that market is less saturated with existing apps.

However, having switched over myself, as an end user, from an iPad to an Android tablet, I can see there are problems ahead for developers. End users have no way of knowing, in the Android Market, what apps have been optimised for tablets and which are legacy apps that run, but don’t look that good, on a tablet. This means that apps developed for tablets get lost in the hundreds of thousands of compatible but less desirable apps. For developers, it’s difficult to get seen even though there aren’t many tablet specific Android apps. The Android Market has a section for tablet apps but this includes both legacy apps and those written specifically for the tablet.

UPDATE 29 February: I now have a simple site at http://www.appsforandroidtablet.com/ listing tablet specific Android apps I have found.

The Long Tail of Android Devices

Simon Judge, Android Developer — Thu, 26 Jan 2012 10:55:10 +0000

There’s a growing number of devices that aren’t officially sanctioned by Google. What are the implications for developers such as myself?

First of all, sanctioned devices, called ‘Google Experience’ devices, have passed Google’s compatibility test suite. This allows them to run Google’s apps such as Gmail, Maps, and Market or as Google call them ‘Google Experience Phones’.

Non Google experience devices include…

Smartphones and tablets created by smaller OEMs who don’t want the extra implied costs
Tablets created by corporations, such as Amazon with it’s Fire, who want to try to control content on the device.
Specialised devices such as Casio’s VX-100 Android EPoS System and Parrot’s Asteroid

This is just the start. There are also set top boxes, vehicle entertainment systems and many more traditionally embedded systems that are moving over to Android. I am also seeing a greater number of enquiries for Android captive (kiosk) solutions. There’s a long tail of devices with new opportunities.

At first this looks good for the Android developer. However, with this new opportunity comes more pain. I have previously written about the limitations of developing for Kindle Fire and developing on non-Google experience devices.

A recent complication is the introduction of some very compelling tablets from China. For example, you can already get a $100 7" capacitive-screen ICS tablet. The very latest ones are using a MIPS Ingenic processor which means that consumer apps that use native code (the NDK) won’t work. However, for developers such as myself thinking about using such tablets for custom client solutions, there’s an add-on (or rather superset) of the NDK provided by MIPS.

As an aside, some non-trivial apps end up using the NDK either to allow porting of existing code or to bypass Android’s Java low heap size. This means, for example, it’s not possible to load full size images or large memory objects in Java. NDK code does not have this restriction.

Mobile Marketing Flaws

Simon Judge, Android Developer — Wed, 25 Jan 2012 14:32:05 +0000

There’s a thought provoking post at Mashable on "5 Flaws in Your Mobile Marketing Strategy". It talks of the following flaws in current apps…

De-Emphasizing the Unique Nature of the Medium
Treating Connected Consumers as Stationary Targets
Ignoring the Power of Contextual Relevance
Employing Old World, Unidirectional Strategies
Not Offering Value in Exchange for Value

These are the flaws from a marketing viewpoint but what about the technical side? How can brands create apps that are more than, what might as well be, on a web site?

One way of looking at this is to explore what’s available, technically, and see how you might weave it into your app to make it more compelling. Phone capabilities include communication (phone calls, text messages, email), contacts, Bluetooth, location, sound, motion, WiFi, camera and alarms. Think how you might use these in your app screens to either trigger things or to provide additional input to, or output from, your app. Triggers also need not necessarily be in your app screens but instead might cause notifications that cause the user to use your app more often.

The Implications of App Use on App Design

Simon Judge, Android Developer — Tue, 24 Jan 2012 14:01:23 +0000

I came across an interesting paper on Falling Asleep with Angry Birds, Facebook and Kindle – A Large Scale Study on Mobile Application Usage (pdf). It describes a study of 4,125 users, between August 16th 2010 and January 25th 2011. The main conclusion of the study is that mobile device users spend almost an hour a day using apps but spend less than 72 seconds with an app at a time. Here are some further insights…

Mobile phones are still used mostly for communication (text and voice)
Some apps have intense spikes in relative usage (e.g. music and social apps)
Other apps are more broadly employed throughout the day
When people actively use their devices they spend less time with each app
Short sessions with only one app are much more frequent than longer sessions with two or more apps
When people are traveling they are more likely to use multimedia apps and they are surprisingly less likely to use travel apps

This research tends to support the commonly held theory that people ’snack’ on apps. From a design perspective it follows that it’s best for apps to allow people to easily get to the main purpose of the app rather than have to go through intermediate (menu) screens. Also, just because people only use apps for a short time doesn’t mean the apps have to be idle. There are types of app that can work in the background and notify users when something happens such that the actual time spent in the app can be minimised.

Another thought I have is whether all this equally applies to tablet apps. Unlike smartphones, tablet apps are great for applications such as news reading where an app session can potentially be a long time. In general, it’s easier to read information on the tablet and there might be more inclination to linger and view more. If so, this might have design implications for smartphone as opposed to tablet apps.

UPDATE 26 January 2012: If the paper is too dry for you, there’s an infographic based on this data at appazaar.

Updated Mobile Market Research

Simon Judge, Android Developer — Mon, 23 Jan 2012 11:00:55 +0000

I have updated my listing of Mobile Market Research to include some research from the last few months. Some of the older links from as far back as 2004 no longer work. However, I have kept them because the titles hold some information.

It’s also often possible to use the WayBackMachine to find the original article. (Right click a link and do ‘Copy Shortcut’ in Internet Explorer, ‘Copy Link Location’ in Firefox, ‘Copy link address’ in Chrome and paste the address into the WayBackMachine.)

News Hound for Android Tablets

Simon Judge, Android Developer — Mon, 23 Jan 2012 08:56:28 +0000

One of the problems with having a blog is having enough to write about. When short for something to write about I tend to search for new recent news articles that either inspire me to write or can be used as a basis for something on which I can comment.

For years, I used the Delicious subscribe facility that gave me inspiration via the many uncovered articles on mobile. Delicious was great because it had the right type of more technical user while Digg, StumbleUpon and Reddit seemed to have less useful more consumer oriented topics. When Delicious was sold by Yahoo, the new owners removed the subscribe facility and I moved to mainly using Twitter. The problem with Twitter has been too much noise and duplication. This has led me to write my own article discovery app.

News Hound runs on Android 3.1 and later tablets. You enter the topics and it goes away and does lots of heuristic based stuff to get quality articles. While I initially used it for uncovering content for blog articles I now also find it useful to isolate the very latest news articles when something newsworthy happens.

Bloggers and people posting regularly on social medial sites should find the app useful. I suspect it could also be used by companies or organisations to discover where they have been very recently mentioned or by anyone to learn the very latest on whatever their passion might be.

Once you have discovered articles you can view within the app or, for more control, in the tablet web browser. You can also share article links to Twitter, Facebook, Evernote, Email etc. News Hound is free and available on the Android Market.