Java ME is Not Surging Ahead of Android, and Other Fun With Numbers

Java ME is Not Surging Ahead of Android, and Other Fun With Numbers

Jan 3, 2012

While everyone talks about the domination of iOS and Android in the smartphone market, and the battle for supremacy among the two platforms. However, there appears to be a third competitor worth mentioning: Java ME. Yes, the same Java ME that powered many handsets long before Android came around, and before the iPhone was a twinkle in Steve Jobs’ eye. According to NetMarketshare by way of Fortune, Java ME took second place back from Android by the end of 2011 after Android surged ahead in October 2011.

The news seems shocking at first, but upon further examination, this does not point to the death of Android. First, iOS dropped a greater percentage of mobile browser usage than Android did, going from 61.50% in October to 52.10% in December – no one’s certainly willing to start making funeral arrangements for iOS quite yet, are they?

One explanation could be just different phone usage during the holiday season, as Electronista speculates. With many people on the go, especially people who might still be using ‘dumbphones’ powered by Java ME, they might be more inclined to surf the web on their devices while waiting in an airport terminal, or while their family discusses awkward topics in the living room.

As well, it appears that NetApplications is just measuring web browser traffic, and not all mobile usage entirely, based on their own methodology. iOS and Android users don’t always use their browsers to access internet content – apps for accessing web services are very widespread on the platforms. As such, Java ME may be more susceptible to fluctuations up or down in browser usage because the browser is the point of entry to internet access – not so for iOS and Android. However, such fluctuations did not occur in 2010 – this could point to an influx of new low-cost devices, perhaps in territories that are only now getting access to mobile data and 3G?

It could even just represent an error in data collection, as the statistics for all platforms appear to wildly skew from norms in October 2011, before eventually returning to more realistic numbers. Apple could also be to blame for the differences, as the October release of the iPhone 4S, which was available on 3 of the 4 major carriers in the US, caused a massive spike in iOS usage. Data on Net Marketshare goes back to January 2010, and for each month where Apple released a milestone iOS device (April 2010: iPad; June 2010: iPhone 4; October 2010: iPod touch 4; January 2011: Verizon iPhone 4; April 2011: iPad 2; October 2011: iPhone 4S), iOS usage spiked for a month or two before decreasing. This is likely a sign of iOS users using their devices more as they acquire them, before settling into normal, lesser, usage patterns.

Android was the only platform to not spike in October 2011, and this may be due to Samsung’s Galaxy S2, which was made available on Sprint in September, and on AT&T and T-Mobile in October. Buoyed by its branding, ad campaigns from both the carriers and Samsung, and the fact that it is the only Android phone to have such a coordinated multi-carrier release that could compare to the iPhone, this is certainly a rational explanation for why it enjoyed its spike like iOS did in October. It’s a natural usage cycle – people use their shiny new phones and tablets at higher rates initially, before their normal usage sets in.

Headlines decrying this as the fall of Android to Java ME are not looking at the big picture – Android is, year to year (December 2010-December 2011), up 4.77% in usage (of mobile browsers alone), and Java ME is down by a difference of 5.21%. iOS only grew by a difference of 2.91% in that same timespan. Mobile browser usage, including tablet devices, in that timespan increased from 3.44% to 7.67%.

The important to know about statistics is that they must be used and analyzed properly. Using small samples and selective usage of data can lead to flawed analysis. Looking at and analyzing the bigger picture, and trying to recognize potential errors is a huge part of data analysis. The analysis of that bigger picture does make it seem as if the story is not that Java ME is suddenly on the rise, that other circumstances either in usage or in data collection are in play here. Android is still growing, and growing quicker in a market that has itself doubled in market share in the past year. It definitely is being used less than iOS, but this does correlate well with the hypothesis that Android is cutting into the dumbphone market more so than they are into iOS, as more Android devices become available to low-end users, who have primarily been serviced by Java ME phones.

However, despite some faulty math and ham-handed analysis, it does not appear as if the once-ubiquitous mobile OS is making a sudden comeback, only that iOS and Android usage spiked in October, and the numbers started to level off and correct themselves in November and December. The numbers for January 2012, as well, will continue to tell the story of just where J2ME stands in a world where iOS and Android continue to grow.