These numbers didn't make sense considering more than half the population can't afford to eat let alone pay for a smartphone. And then factor in the number of children, these figures would mean that pretty much everybody has a smartphone. Then I realized that the typical android user gets a new phone every year, so the number of users is a far different number than activations - about 75% less.
There are an estimated 7.1 billion people on the planet so your assumption that 1 billion is more than half the population would not be correct. I have no idea how often a typical Android user gets a new phone, but for American the average is most likely every two years as many go with two year subsidized contract thru a phone carrier.
With that out of the way, I completely agree that the number of Android users is a much different number than activations and my post agreed with that (I'm not sure if you missed that aspect). The post I was replying to suggested that the number was not accurate as well, but the reasoning he cited was factually inaccurate. It would be more accurate to factor in repeat customers as a reason instead of the incorrect assumptions that were being made. As mentioned, I account for three of those 1 billion activations and I'm only one person.
I'm not sure where you're getting the 75% figure (just made it up?). That number seems on the high side to me, but not to an unreasonable degree. It certainly doesn't seem out of the realm of possibility.
In the end I think we both agree that activations does not equal number of users.
They only talk about activations. Well, how many are currently ACTIVE? Meaning, how many deactivations or how many are just going to someone that is upgrading to a new phone so they can get a later OS because they won't upgrade an existing model?
And how many revs is it going to take until it becomes, well, stable?
How many years is it going to take until they have most of their active users actually USING it?
OT, but is Huddler having a little meltdown right now, or is it just me? I am using Safari 6, Mountain Lion. I see:
Notice: Constant CONFIG_ENV_DEFAULT already defined in/var/www/sites/huddler/releases/397dd1de9069631534e9b4ba3bdf1a8bf801db30/common/defines.phpon line 9
Notice: Constant CONFIG_ENV_MOBILE already defined in/var/www/sites/huddler/releases/397dd1de9069631534e9b4ba3bdf1a8bf801db30/common/defines.phpon line 10
Notice: Constant CLASS_NAME_PREFIX already defined in/var/www/sites/huddler/releases/397dd1de9069631534e9b4ba3bdf1a8bf801db30/common/defines.phpon line 13
etc. There seems to be like a hundred of these spilling out into my browser page on this very thread...
EDIT: This only seems to appear in this particular thread. Other forum threads appear OK.
EDIT2: It appears fixed now. The editor has changed to Basic Format mode, and I'm staring at embedded HTML tags. Ugh. No.
how true, the fanboys will be all jump up and down about the new version even though it might not run on their phone, but they will be all giggly about it.
And the .05% of the android population that doesn't have a newer Nexus product will get the update through XDA after they root their phone and side load the OS.
Well, okay...and security feature in the base OS level since I tried to search on this topic (which android version to target) for the last month and all I could find was the statistic that 79% of mobile malware targets Android.
Hence the out of date 2012 metric since I didn't find any late 2013 ARPU numbers. However, if the ARPU numbers still hold then Gingerbread (API 10) is the sweet spot with a 1.20 ARPU and 33% market share according to the dashboard. If I go to API 15 (ICS) then I lose a third of my potential market. If I go to API 17 I lose 93.5% of my potential market.
Google Play Services does not address API or OS level security flaws. Just the allows updates to the core apps that support Googles revenue stream (YouTube, Gmail, etc) and the Google APIs that support those apps. So vulnerabilities in those apps get patched but not the API/OS vulnerabilities that malware apps will use on 2.x devices.
And lord help you if Google Play Services itself ever gets hacked. It has essentially any privileges it deems necessary to do whatever it wants to your phone without asking. Also lord help you if it ends up being buggy because it always runs in the background. Watch that battery life crater if it doesn't want to sleep correctly on your phone.
It does help me as a dev if the new API I want to use is part of GPS but not for the code application framework APIs, hardware APIs or security APIs. But an API 10 device + GPS is still mostly a API 10 device if I'm not using google services (maps, account auth, google+ stuff, etc) heavily in my app. Button Bar, notification panel, lock screen, etc is all still API 10.
That's not to say this isn't a huge improvement...account authorization and in-app billing is very important as are the location and games APIs...but it doesn't really change the fragmentation story until all the 2.x devices are history. I still wont target above API 10 till that happens.
With respect to the iOS fragmentation, it matters more to users than developers whether they get Siri or not. I still get to target the latest API version in iOS and cover the majority of the iPhone market. The remaining few percentage points probably have lousy ARPU numbers anyway.
For folks keeping track, GPS is closed source and essentially Google just forked Android into a closed source architecture. No more free ride for Amazon. Or Samsung for that matter if they ever play hardball with Google. Sorry, you're not a Google branded/approved device. GPS wont run on your android and without it, half the APIs are gone...
Absolutely Spot on. This solves fragmentation for Google. But it's a work around and a bad work around at that - heading in the wrong direction. What Google needs to do is force manufacturers to release the radios and drivers for the new release, create a hardware abstraction layer and a base OS layer for which that they can directly update for security patches and encourage manufacturers to release their skins and apps in the Play Store just like they are doing with their own apps. Some of it's tricky and may not be possible since at least some portion of the software is not just skins, but changes made in the OS. But Google needs to figure out a way to allow the manufacturers to do this. IBM did this 2 decades ago with the brilliant Work Place Shell in OS/2. 3rd party ISVs took the WPS to a completely different level, well beyond IBM's developers expectation, by subclassing the WPS classes.
so a company that makes profit from selling privacy makes a pact with a company that wants to privatize water supplies - amazing move, now I will look very closely not to buy anything from Nestle again
<p><span style="color:#454545">etc. There seems to be like a hundred of these spilling out into my browser page on this very thread...</span></p>
<p> </p>
<p><span style="color:#454545">EDIT: This only seems to appear in <strong>this</strong> particular thread. Other forum threads appear OK.</span></p>
<p><span style="color:#454545">EDIT2: It appears fixed now. The editor has changed to Basic Format mode, and I'm staring at embedded HTML tags. Ugh. No.</span></p>
I'm curious if 9to5Mac and AI are under the same ownership now. 9to5 is having some of their own formatting issues today and there's been a few articles here at AI in the past few days that seemed to be almost word for word duplicates of 9to5 articles.
That's one of the best things about the German language.
Unless you're designing Responsive websites these days: menu's, asides, whatever refuse to fit!!!! Grrr!
Actually, wrapping my head around "good typography" in Germany has been the hardest thing I've done for the last 20+ years here. How ironic considering wood and metal type, along with many a typeface was originally created and designed here; but to read a magazine or newspaper here is sometimes painful when ever single sentance has a hyphenated word!
Just a respite from hating on Google... carry on... :smokey:
"Activations" is a BS, made up stat. Putting that in the heaedline tells me the writer is part of the mindless "android is winning" apple bashing faction.
WTF is up with the editorial voice of this website?
Comments
These numbers didn't make sense considering more than half the population can't afford to eat let alone pay for a smartphone. And then factor in the number of children, these figures would mean that pretty much everybody has a smartphone. Then I realized that the typical android user gets a new phone every year, so the number of users is a far different number than activations - about 75% less.
There are an estimated 7.1 billion people on the planet so your assumption that 1 billion is more than half the population would not be correct. I have no idea how often a typical Android user gets a new phone, but for American the average is most likely every two years as many go with two year subsidized contract thru a phone carrier.
With that out of the way, I completely agree that the number of Android users is a much different number than activations and my post agreed with that (I'm not sure if you missed that aspect). The post I was replying to suggested that the number was not accurate as well, but the reasoning he cited was factually inaccurate. It would be more accurate to factor in repeat customers as a reason instead of the incorrect assumptions that were being made. As mentioned, I account for three of those 1 billion activations and I'm only one person.
I'm not sure where you're getting the 75% figure (just made it up?). That number seems on the high side to me, but not to an unreasonable degree. It certainly doesn't seem out of the realm of possibility.
In the end I think we both agree that activations does not equal number of users.
THey have to do something to attract the little children.
I like KitKat bars, but I ain't switching to Android.
Yeah, and 90+ % of the current Android owners won't get it the day it gets released.
What IDIOTS.
And how many revs is it going to take until it becomes, well, stable?
How many years is it going to take until they have most of their active users actually USING it?
OT, but is Huddler having a little meltdown right now, or is it just me? I am using Safari 6, Mountain Lion. I see:
Notice: Constant CONFIG_ENV_DEFAULT already defined in/var/www/sites/huddler/releases/397dd1de9069631534e9b4ba3bdf1a8bf801db30/common/defines.phpon line 9
Notice: Constant CONFIG_ENV_MOBILE already defined in/var/www/sites/huddler/releases/397dd1de9069631534e9b4ba3bdf1a8bf801db30/common/defines.phpon line 10
Notice: Constant CLASS_NAME_PREFIX already defined in/var/www/sites/huddler/releases/397dd1de9069631534e9b4ba3bdf1a8bf801db30/common/defines.phpon line 13
etc. There seems to be like a hundred of these spilling out into my browser page on this very thread...
EDIT: This only seems to appear in this particular thread. Other forum threads appear OK.
EDIT2: It appears fixed now. The editor has changed to Basic Format mode, and I'm staring at embedded HTML tags. Ugh. No.
how true, the fanboys will be all jump up and down about the new version even though it might not run on their phone, but they will be all giggly about it.
And the .05% of the android population that doesn't have a newer Nexus product will get the update through XDA after they root their phone and side load the OS.
The most important aspect/definition of fragmentation to me (and I would guess many mobile devs) is API penetration rate and ARPU.
http://adtmag.com/articles/2012/11/02/android-version-developers.aspx
Well, okay...and security feature in the base OS level since I tried to search on this topic (which android version to target) for the last month and all I could find was the statistic that 79% of mobile malware targets Android.
Hence the out of date 2012 metric since I didn't find any late 2013 ARPU numbers. However, if the ARPU numbers still hold then Gingerbread (API 10) is the sweet spot with a 1.20 ARPU and 33% market share according to the dashboard. If I go to API 15 (ICS) then I lose a third of my potential market. If I go to API 17 I lose 93.5% of my potential market.
Google Play Services does not address API or OS level security flaws. Just the allows updates to the core apps that support Googles revenue stream (YouTube, Gmail, etc) and the Google APIs that support those apps. So vulnerabilities in those apps get patched but not the API/OS vulnerabilities that malware apps will use on 2.x devices.
And lord help you if Google Play Services itself ever gets hacked. It has essentially any privileges it deems necessary to do whatever it wants to your phone without asking. Also lord help you if it ends up being buggy because it always runs in the background. Watch that battery life crater if it doesn't want to sleep correctly on your phone.
It does help me as a dev if the new API I want to use is part of GPS but not for the code application framework APIs, hardware APIs or security APIs. But an API 10 device + GPS is still mostly a API 10 device if I'm not using google services (maps, account auth, google+ stuff, etc) heavily in my app. Button Bar, notification panel, lock screen, etc is all still API 10.
That's not to say this isn't a huge improvement...account authorization and in-app billing is very important as are the location and games APIs...but it doesn't really change the fragmentation story until all the 2.x devices are history. I still wont target above API 10 till that happens.
With respect to the iOS fragmentation, it matters more to users than developers whether they get Siri or not. I still get to target the latest API version in iOS and cover the majority of the iPhone market. The remaining few percentage points probably have lousy ARPU numbers anyway.
For folks keeping track, GPS is closed source and essentially Google just forked Android into a closed source architecture. No more free ride for Amazon. Or Samsung for that matter if they ever play hardball with Google. Sorry, you're not a Google branded/approved device. GPS wont run on your android and without it, half the APIs are gone...
Absolutely Spot on. This solves fragmentation for Google. But it's a work around and a bad work around at that - heading in the wrong direction. What Google needs to do is force manufacturers to release the radios and drivers for the new release, create a hardware abstraction layer and a base OS layer for which that they can directly update for security patches and encourage manufacturers to release their skins and apps in the Play Store just like they are doing with their own apps. Some of it's tricky and may not be possible since at least some portion of the software is not just skins, but changes made in the OS. But Google needs to figure out a way to allow the manufacturers to do this. IBM did this 2 decades ago with the brilliant Work Place Shell in OS/2. 3rd party ISVs took the WPS to a completely different level, well beyond IBM's developers expectation, by subclassing the WPS classes.
Google searched "confectionary starting with K".
KitKat.
A chocolate biscuit that you fragment to eat. It was an obvious choice.
That vid reminds me of Apple product videos. Copy on Google.
so a company that makes profit from selling privacy makes a pact with a company that wants to privatize water supplies - amazing move, now I will look very closely not to buy anything from Nestle again
Guys,
Let's try not to focus on the unimportant aspects of this story (activations) and focus on the important ones.
Like how Google named their next operating system 'Kitkat' after having pussy on their mind.
(http://appleinsider.com/articles/13/08/29/6-months-after-andy-rubins-exit-another-top-executive-leaves-googles-android-team)
Sorry, Sorry.
I didn't mean to mention the fact that Hugo Barra, Vice President Of Product Management, wasn't happy with getting his product managed.
Or that an Android employee forked off...to China.
Or that Sergey's wife is now threatening to fragment...his bank account.
God, I could go with the Android jokes all night.
Rather than discuss KitKat I'm going to stop typing now....and take a break.
Can somebody please get this thread back on track?
I'm curious if 9to5Mac and AI are under the same ownership now. 9to5 is having some of their own formatting issues today and there's been a few articles here at AI in the past few days that seemed to be almost word for word duplicates of 9to5 articles.
Actually, wrapping my head around "good typography" in Germany has been the hardest thing I've done for the last 20+ years here. How ironic considering wood and metal type, along with many a typeface was originally created and designed here; but to read a magazine or newspaper here is sometimes painful when ever single sentance has a hyphenated word!
Just a respite from hating on Google... carry on... :smokey:
WTF is up with the editorial voice of this website?
Send the Apple bashers packing.
And not the warm, eccentric, rich, and deep Wonka…
But rather the Michael Jackson Wonka.