Android Fragmentation - Changing How the Numbers Add Up

Fragmentation has been a much discussed issue over the past few years. And it does not seem to be resolving itself in the most recent information which has been released. Gingerbread is still holding strong at about 39% and given that it was released at the end of 2010, it still has a huge market presence more than 2 years later. One would have thought that it would be tapering off by now, but that is not the case. Ice Cream Sandwich has finally grown close to the 30% mark for its share of the Android market. It was released in October 2011 and one would have thought that it would have grown quickly replacing Gingerbread.

All of this is from the latest report from the Android Developer site where they put out numbers monthly to show the status and distribution of the various versions of Android. As of this latest report, they are now showing Jelly Bean at 25% which is looking extremely good as compared to the slow growth of Ice Cream Sandwich. Jelly Bean was released as the latest Android version in June 2012, so things are positive in working towards getting the version distributed. But is that necessarily true? In the information from the site, there is a note regarding how they are counting things.

Note: Beginning in April, 2013, these charts are now built using data collected from each device when the user visits the Google Play Store. Previously, the data was collected when the device simply checked-in to Google servers. We believe the new data more accurately reflects those users who are most engaged in the Android and Google Play ecosystem.

It seems that the changes in the method of counting has put Jelly Bean in a very favorable light giving them a huge jump from the March Android report. In March, Jelly Bean was sitting at just over 16% and now with the April Android report, they are now at 25%. That is a huge one month increase and can partially be attributed to the change in the way that they are calculating the numbers.

On the surface, they are changing the method of measuring Android OS version to make things more accurate. But the end result is a huge jump in the numbers for Jelly Bean making it look to be highly successful in less than a year. In the end, this may not mean that much for Android Fragmentation. The next months reporting may be the one to compare month over month to see how that one looks as compared to the recent changes in how things are measured.

Android Fragmentation Will Never End

Gingerbread remains majority share holder for Android OS

Another report of the state of Android Fragmentation has come out and Ice Cream Sandwich is now at 23.7% and Jelly Bean is at 1.8%. The two combined have finally broken the 25% barrier and are going to continue to rise. But that rise is dependent on new devices being purchased to make that shift. The Galaxy S3 has done a lot to move the percentages higher with its great sales. But that is only one device among the large number of phones and manufacturers in the market. While we are now approaching the 1 year mark for Ice Cream Sandwich being available, being at only 23.7% is not speaking volumes for the version of Android. Even the recent Jelly Bean seems to be growing faster an Ice Cream Sandwich did.

Android Fragmentation has been a topic of conversation for the past few years. And it is not going to go away. It will continue to be discussed for many years, primarily because of the market place that it supplies the Android OS to. Unlike Apple, which controls both the hardware and the OS, Google must handle many different layers in order to provide their Ice Cream Sandwich and of late Jelly Bean to new phones. Apple is in control of the process and can chose to bypass wireless carriers in order to maintain a level of control not achieved by Google or even Microsoft with the Windows Phone OS. For Google, there are just plain too many layers they have to go through to get a phone updated with the latest OS.

One of the other big problems is that many of the phones which are able to support Gingerbread cannot support Ice Cream Sandwich, limiting the ability to reduce the fragmentation which is going on. And that is going to continue to be a problem until those phones limited to Gingerbread are retired. But that is not going to happen any time soon. Gingerbread OS phones are still being sold today and that does not look like it is going to be altered. Cheaper mobile devices will not support Ice Cream Sandwich and that is a fact.



Is there a solution for Google to try and resolve the adoption of Ice Cream Sandwich? There may be, but it will take far more effort that Google is probably willing to do. The current process works with Google releasing a new version, such as Jelly Bean. That version is then released to manufacturers who then test it on their mobile phones and add their pieces to it. Once that is done, the manufacturers turn it over to the wireless carriers who then add their pieces to the OS until it is finally ready to be released to owners of a particular phone. It is not in the best interests of wireless carriers to keep putting out updates. I have been with Verizon for a number of year and can expect to see only 1 significant update to the OS. For the Galaxy S3, that will be Jelly Bean. I would like to see what comes after Jelly Bean, but past history says that will probably not happen.

The solution for some of this may be to cut out the addition of all the pieces to the Android OS. The Nexus phone is pretty much a pure Android OS and Google could go down the path of trying to implement the idea of a pure OS being distributed to the owners of mobile phones. But doing that is going to take a lot of effort from Google to get those agreements from manufacturers and wireless carriers. That entails a lot of discussions and negotiations.

It is doubtful that we are going to ever see the Android Fragmentation issues ever be resolved as some hardware is never going to support the current Android OS, what ever it is. The goal should be to see a majority of the Android devices being on the latest OS within 1 year of its release. For Ice Cream Sandwich, that is not going to happen. And for Jelly Bean, it does not look likely right now.

Until Google alters the current working model for distribution of updates, Android Fragmentation will continue.

Android 4.0 - Fragmentation Continues For ICS

Latest report show very small growth in ICS adoption at 7.1%

Google has posted the most recent numbers for Android on their Android Developer site and the numbers they are showing for Ice Cream Sandwich adoption leave something to be desired. As of June 1st, Android 4.0 has a share of 7.1% of all Android devices coming to the Google Play site which is a good indicator of usage. And it is those numbers which are showing very disappointing results for the expected rise in ICS adoption. It seems that Android Fragmentation is alive and well with no signs of going away anytime soon.

Over the past 3 months, we have been watching the numbers with an expectation that ICS would climb above the 10% mark with the June report. In the previous May report of 4.9% the growth back then seemed like things were beginning to make progress, growing from 2.9% to 4.9%. The growth from 4.9% in May to 7.1% in June may seem like it has seen some great increases in adoption. But if you realize that ICS was first presented with the Galaxy Nexus in December, things do not look so good.

Making things even worse is the fact that Android 2.3 has grown from 64.4% in May to 65% in June. If Android 4.0 adoptions where growing at a much faster rate, that percentage should have gone down, not up. With the slow growth for Android 4.0, it is starting to look like the most likely outcome is growth through attrition. As Android 2.3 phone owners replace those devices with Android 4.0, we will begin to see the numbers start to move up much faster. The problem with that possibility is that there are still large numbers of Android 2.3 devices on sale and that is going to continue for much of the year.

Given all that is being shown here, it is quite possible that finger pointing for the lack of adoption of ICS could pick up again. There is the path that ICS has to travel from Google, to manufacturers, to wireless carriers. Each stop takes time as they add their pieces to the puzzle in order to make it more of their own. Take the case of the Galaxy S2 and the ability to get Ice Cream Sandwich installed on that device. Here is the US, T-Mobile has announced it coming on June 11th and there is no word from AT&T or Sprint. It has rolled out in other countries, but not yet in the US. It seems that the roll out is moving at a snails pace and there are no signs of things changing very much in that direction.

Exactly when we can expect to see ICS take off and have at least a 25% market share is unknown. Whether attrition is the most likely method of growing ICS seems likely this year given what we are seeing. That is not the way this should be going because Jelly Bean in on the horizon and that will add to the fragmentation even further than it already is. Who is responsible for all of this is a moving target as Google says they have released it to manufacturers. Manufacturers have released it for various devices to wireless carriers. At some point, the length of time between Google releasing a version and the public getting it has to be radically shortened. Is that ever going to be possible?

Is This A Picture Of The Android Fragmentation Effect?

The discussions regarding Android Fragmentation have been moving upward this month, especially because of the most recent numbers coming from Google's own Android Developer site. The numbers for ICS adoption as of the beginning of the month show Android 4.0 with a 4.9% share of Android devices. That is less than 5% and it has been on the market for more than 5 months. That is very telling and we may not see it in double digit territory when they report at the beginning of June. That brings us to the topic about Android Fragmentation and just how bad it actually is. Open Signal Maps has created a chart which we are showing here. This chart represents the 3,997 unique devices they have found which have downloaded their OpenSignalMaps to a device.

If you look at this, it is overwhelming as to what we are seeing. The one green square represents the GT-i1900 which is Samsung's Galaxy S2 device which is very popular. Of the 3,997 devices, they have figured that 1,363 represent those Android devices which have a custom ROM build and installed before the download of their app. Even if you discount those custom ROM's, that still leaves 2,634 unique devices which they have to deal with in the app so that it will operate correctly. That is a staggering number of devices to have to deal with. They have to deal with 599 phone brands being sold with Samsung being the largest brand name with 40% of the market. Everything goes down from there.

So, you might ask what is the point of all of this. While the company could choose not to support some of the smaller brands, it would cause them to miss out on potential sales. Other smaller companies probably do just that which means that if you have a brand name phone which has a very small market share, you may not even have the opportunity to purchase an app from one of the app companies because they do not support your phone. And that is the point. With all these various phone models, it makes it more difficult for developers. You then have to factor in the Android Fragmentation effect with the potential for each phone model to have at least 2 Android OS versions which can operate on the phone which have been officially released by phone manufacturers. That starts multiplying things out and gives you an idea as to just how tough it is for app developers.

You then add to all the various phone brands, various versions of the phones and then all the various versions of the Android OS and it presents challenges to app companies. On top of all that is now the various screen sizes which have to factor into the mix of things which they are trying to resolve. On top of that are the various screen resolutions as well and you start to understand the challenges of developing apps for Android phones.

While the physical devices stay static once a phone is released, the OS is changing which has to be support and more of a problem it becomes for a company developing an app. If they have been in the market for years, their app has already been developed for existing Android versions, so they would only have to address the latest version. But for a new company trying to develop an app for the Android OS family, this is a daunting task.

While all the news for developers is not bad, it is all these challenges which cause some to not develop for Android. We can hope that Google can change some of this and it looks like they may be moving in that direction. Rumors of Google selling Nexus phones with Jelly Bean may force this to change in the future.

The discussion on this will be continued by many.

Are Wireless Carriers Responsible For Fragmentation?

Carriers release few updates for phones as compared to the OS updates created.

 
And it is the lack of updates being released which should be raising some big concerns amongst all the manufacturers of phones. While there is no empirical proof of the number of updates which smart phone owners never seem to get, there are enough examples of this which point out problems. The upgrade process for the OS on a smart phone should be multiple times over the course of a year and yet it seems that there is often only a single update. That is unless you have an Apple iPhone where they force these on wireless carriers and prevent them from adding their items to the puzzle. For Android devices, this is turning into a big problem which is not going to go away until some things change in the process.

It seems that wireless carriers are at odds with the largest OS provider in Google's Android OS. Recently, there has been some finger pointing between AT&T and Google which in this case made AT&T look bad. This is all over the problems with the slow roll out of Ice Cream Sandwich (Android 4.0). The slow roll out is a big problem for Google right now as they only have 4.9% of Android devices on the latest OS for all active Android phones. Given that Android 4.0 has been out since November on the Galaxy Nexus, it makes one wonder exactly what is going on.

This seems to point squarely at wireless carriers for the delays, though manufacturers of smart phones are to blame as well. They all seem to want to add their pieces to the Android OS and that slows things down. There is the development time and then testing for each step along the way. If a problem is discovered in testing, it goes back to development for a fix and then a repeat of the process. If this process were to be changed so that the stock Android 4.0 was used and no other changes to the OS were done other than to support the individual hardware, that would significantly speed up the process. Or would it.

In a post from Gotta Be Mobile by Adam Mills, he is relating his experiences with his Galaxy Nexus which he purchased from Verizon 5 months ago. Since he purchased the phone, he has not seen an update from Verizon. Not one. He is currently running Android 4.0.2 which came with the phone when he purchased it. Since that time, Android 4.0.3 and Android 4.0.4 has been released. Android 4.0.4 started rolling out to Galaxy Nexus phones, but only for the GSM version. To date, there has been no release of the latest version of Ice Cream Sandwich at Verizon since the Galaxy Nexus was launched. This is a very bad sign.

I owned a Blackberry Storm for a time which I purchased from Verizon back in January of 2009. This was a few months after the 9530 model was released and it had a number of reported bugs associated with the phone which could be easily fixed by a software update. Ever 3 or 4 weeks, there was another updated OS from Blackberry to fix problems, but these were never released by Verizon. This seemed to continue for what seemed like an eternity. Finally, in October 2009 came the official word that Verizon was releasing the 5.0.0.328 OS update for the Blackberry Storm. This was great to download and fix problems, but that was the last update for the phone that I ever saw.

This is not to pick on Verizon as all the other wireless carriers in the US have probably done the same thing at one time or another. And it is probably more widespread than we would all like to believe. The pattern that I am seeing is that you get one significant software update and that is it. You are done until you purchase your next phone. I am sure I am blowing this out of proportion, but then again maybe I am not. In an article about Android Orphans, the number of devices which are one or more versions behind is not how it should be. I am sure that you can find other examples of this by doing some searches on the web.

The idea of only getting a single OS update on a phone is very discouraging and something must change. If not, people are going to start looking elsewhere for a phone where they are able to get multiple updates over the course of the year. And a phone where they get a major update close to an annual basis. At the moment, Apple is the only one that has been able to fulfill that and Google had better take notice. I am not an iPhone supporter and prefer the Android OS, but the idea of only getting a single update from my wireless carrier is discouraging.

Have you had a similar experience of only getting a single OS update to your phone?