Galaxy S III beats out iPhone 4S as top selling device in US | Hacker News Reader

Galaxy S III beats out iPhone 4S as top selling device in US

androidandme.com

158 pointslusob144 comments

"The overwhelming success of Apple’s iPhone 4S has finally hit a lull" - about 2 weeks before it's due to be replaced as the flagship iPhone. The iPhone purchasing cycle is very cyclical - huge sales after the new model is introduced, gradually declining up to the next model (which repeats the cycle). It would surprise me if the S3 hadn't taken the lead over the summer.

Yeah, the article is nonsense.

The linked chart is still intersting though. I'm surprised that the Galaxy S2 is still #3 on T-Mobile.

The chart is derived from nonsense too. The article is sourced from an "Analyst". Analyst is a fancy title for a pundit who is paid to have an opinion by some industry figure. In this case, most likely Samsung.

These analysts post "reports" that show the company is winning, and the company uses those reports to try and drum up sales.

Meanwhile the "analysts" send press releases to blogs who then report the results as if they were fact like this.

It is usually only years later (if ever) that the real results come out.

Only Apple reports its iPhone unit sales. Samsung, for instance, does not.

It wasn't until the lawsuit with Apple that they were forced to give up these numbers, and they were very anemic. (while at the time "analysts" were saying they were kicking butt.)

This is just hype being spread by a blog that is trying to make it look like android is "winning". None of the android makers report real sales numbers, including Amazon (so we don't know how many fires have been sold, whether it is a success or not.)

Keep telling yourself that. Yes, the industry plays number games, but when the multiple different indicators from many different analysts and direct numbers from retailers and actual usage numbers all point in the same direction, you can go ahead and call the trend.

The trend of many people with a deep vested interest is to lie?

It would explain why Jim Cramer is so consistently wrong.

I don't find the lede here unreasonable. However a couple of nitpicks:

The linked chart and article both imply these are only for sales at carrier stores. i.e. not including Apple Store sales which are significant and Samsung has no equivalent for.

Accurately weighting ranked charts with no numbers attached can be unreliable. Further the people on commission at carrier stores have an incentive in promoting devices that put more money in their pockets, making verbal sources like this even less reliable.

Has this ever happened in previous years (before new iPhone releases)?

Yes. Every year.

Edit: archgrove speaks of the pre-release lull, not the competitor's sales figures. By 'this' MatthewPhillips can only be referring to the lull. If MatthewPhillips had meant SIII sales numbers, he should have clarified the antecedent of his pronoun. Ergo, sp332 is still correct that this is the first time a competitor has outsold iPhone and I am still correct that this lull happens every year - these facts are not mutually exclusive.

Really?

Yes. Really.

Nope.

This person says otherwise: http://news.ycombinator.com/item?id=4479366

Not sure who to believe, anyone have some old links by chance?

The 'this' I was referring to was the subject of the article (which was iPhone being outsold by a single handset).

Being correct shouldn't always be the goal in discussion, by the way.

A good friend of mine once told me:

Don't just defend your right to be wrong, EXERCISE IT!!

Truly words to live by.

There is a lull before the announcement every year. That said, this is the first time a competitor has outsold them.

I'm not certain if a single handset has outsold the flagship iPhone previously (certainly Android as a whole has). However, I believe this is the first time a flagship quality handset from a competitor that is basically eating the Android market has come out in the few months preceding an iPhone.

This is true, mostly because only one company in this space lumps all their phones under one name even when they are on different carriers with different radios and firmwares... until the GSIII that is.

I could be wrong, but I remember reading that there is only 1 hardware model for the iPhone 4s. It supports both GSM and CDMA networks. The wikipedia article seems to support this claim: http://en.wikipedia.org/wiki/IPhone_4S#Hardware

That is correct - all 4s devices have both CDMA and GSM support; the difference between the AT&T and Verizon "versions" is just one of activation, basically.

> The iPhone purchasing cycle is very cyclical

I agree. It's also very critical. What we need is a competitor to time-strategize and release a competitive handset at the same time as the iPhone flagship so that consumers don't have a clear choice. We keep giving Apple the all-clear to continue with it's vicious cycle.

Nokia and Motorola are announcing phones today; it's not clear when they'll be released but at least the marketing will be out there.

As long as the new iPhone will have a pretty tiny screen as the leaks say, the large screen Android phones will keep outselling it even more.

Agreed. It's occurred to me that the only way that Apple will retain me as an iPhone user is if they copy Samsung this time around.

iOS offers a great UX, as far as the system software is concerned, but I honestly think the iPhone 4/4s has the worst physical ergonomics of any phone I've owned. Everything they could have done wrong, they did wrong... and I won't lie: the fact that nobody but me ever says anything like that makes me wonder what's wrong with my judgment.

If the new iPhone actually turns out to be a taller (?!) version of the same package and sales don't drop off a cliff as a result, I'm in for some serious cognitive dissonance.

I'm having some serious cognitive dissonance over your comment. You're saying that if apple does not change the form factor of the best selling phone on the planet, sales will "drop off a cliff"? Solely on the basis of your personal screen size preference?

Yes. I am fairly sure that most people do not want a taller iPhone. They want a wider one, as evidenced by the success of the S3 and other larger smartphones... to say nothing of design principles based on aesthetic proportions that go all the way back to the ancient Greeks.

I have no idea what I would do with a taller iPhone 4S, other than complain about how I still have to spin it around to orient it properly 75% of the time when I pull it out of my pocket, and about how its display is still too narrow to be used safely for navigation and music playback in my car.

If I'm wrong, and there is a huge unmet market demand for an iPhone that is taller than a 4/4S but no wider, it will mean that I have no insight or understanding of the smartphone market as a whole, no product-design sensibility to speak of, and that I need to STFU and stop complaining. Nobody likes to be confronted with their own poor judgment, right?

I would like a taller iPhone because my only compliant about the size is how cramped the screen gets when I have the keyboard open or use apps that happen to have both a toolbar and a nav bar. Otherwise I think the physical size of the device is about perfect.

I disagree about the width/height. 640 on the iPhone vs. 720 on many newer Android phones. Not a huge difference there. Enough to notice certainly but it's the height that is far more dramatic. 960 on the iPhone vs. 1280 on many newer Android phones.

It's not the pixel resolution I'm complaining about, but the physical screen size. The Retina display has a much higher dot pitch than it really needs IMHO. Back off on the dot pitch, keep the resolution the same, and give me a bigger screen.

I'm not too excited over perspective of getting a murse to carry a phone. Taller iPhone still complies with a requirement for a phone to fit in the jeans pockets safely. And that seems to be a more thoughtful reasoning behind product design rather than "bigger = wider".

Funny enough it does seem as if Galaxy S3 is more popular with women here in Singapore. Might that be because they keep their gadgets in purses or I'm just wrong with my observation?

Interesting points. However I can see music playback work nicely: a music player in extra-wide landscape orientation would actually be quite similar to the backlit LCD screen of a classic car radio. You just need to way to firmly fix it somewhere in that orientation.

I agree that wouldn't really work for navigation, but for music it'd be kinda cool.

(note: I'm not actually thinking of buying an iPhone, going to get an Android, though probably not the latest model)

I feel the exact opposite. I had a Galaxy Nexus for a few months earlier this year. One of the reasons I switched back to my iPhone is that the screen is insanely too large for me.

The giant-sized phones have some appeal to some of the market, it's true, but for most buyers bigger is not better.

We're not talking about a wall-mounted display here. We're talking about something you hold in your hand, carry with you all the time, and hold to your face to talk. Bigger can be an issue for those one-hand-thumbing the device as well where they simply can't reach.

People are all pointing to the fact that the new iPhone is coming out, and for obvious reasons, the iPhone4s will not sell as well.

That said, people are forgetting that this is basically the FIRST TIME that any smartphone has sold more than the iPhone in the US. It's not as big news as it could be, but its definitely big news, and could be indicative of a trend in the US, namely, not just cheap/mid-range Androids are being sold, but a lot of high-end Androids, that cost as much as iPhones are being sold. This is a very interesting and important distinction.

It'll be interesting if the next iPhone isn't at least as popular as its predecessors. Until then, I'm chalking it up to pre-announcement lull in iPhone sales.

Anecdotally, the release of the Galaxy Note was the first time I witnessed average people anticipating and being excited by an Android phone. I witnessed the same to a much greater extent with the S3.

I do think it's indicative of a trend. I'm very interested in seeing what happens with the momentum I perceive once the next iPhone has been revealed.

Samsung kinda came out of nowhere with their original Galaxy S, and, along with other people, I absolutely love mine, which I've beaten the hell out of, left in the back of a taxi (this weekend - but got it back), and generally mistreated like a lover for 2.5 years, and it is still in perfect condition. There must be a gradual market effect from the fact that they've made some really good quality kit. I've seen so many iphones with cracked screens!

Hard to know what counts as a lot of high end phones. The four weeks around the sale of the new iPhone is like the music singles chart at Christmas, almost nothing happens before Christmas except people taking advantage of the lull, then the biggest sales of the year.

Or, it could be indicative of a trend away from the myraid of Android devices as Android customers gravitate towards a smaller number of models.

>That said, people are forgetting that this is basically the FIRST TIME that any smartphone has sold more than the iPhone in the US.

Well, didn't smartphones exist before the iPhone ? Also i don't think the iPhone would have become the "top selling smartphone" just out of the gate in 2007. It would have taken them quite a while to get to the top spot.

It can't possibly be the "FIRST TIME". Also since 2008, in the US, iPhones have cost $200 on contract which was par for the course for pre-iPhone smartphones (Blackberry etc.) including the higher end Android phones which came later. To say "as much as" is just as wrong as ""FIRST TIME".

Doesn't it blow your minds how many people are adopting a linux as a primary user OS? And how irrelevant that is, since they don't even know that's what they're running?

Also, the S III has a barometer. How cool is that?

Even if people don't know they are using linux, it's still important if it leads to more investment in the development of linux and related software/technology.

Can anyone shed some light on whether this is happening?

Well, now many people think it's immoral to tell vendors about local-root holes in the Linux kernel, because it could keep people from jailbreaking their Android phones. So there's definitely an effect on Linux.

You cannot jailbreak Android devices. Android devices usually come pre-jailbroken (able to run arbitrary software) out of the box.

What you mean is getting root access to the linux system that android is based on. Which is a cool hack, but not similar to a jail brake.

Since an iPhone jailbreak provides full access and is, in fact, usually used for tweaks and other modifications rather than custom apps, that is disingenuous.

An iPhone "jailbreak" both jailbreaks and roots the device. Android phones (except for very cheap ones from certain carriers I think?) only need to be rooted.

It's not disingenuous, just (perhaps overly) pedantic.

That is my understanding of it anyway.

"Jailbreak" is a made up term. As far as I know, it originally referred to accessing arbitrary files on the filesystem (getting out of a chroot jail), but it now refers to a process including gaining access to the bootloader and/or kernel. Either would be an acceptable use of "jailbreak" (though the former might be misleading), but the term has never meant "run unsigned apps". HTH :p

Hmm, my impression was that the process of "jailbreaking" an iPhone was somehow rather related to breaking out of a chroot jail. I don't know where I got that impression though.

Android uses a Java-based VM. iOS does not. Is that correct?

Can you boot your own OS (sans Java VM) on an Android device? How easy is that to do?

The Nexus devices are generally the most open of all Android devices. Consequently, there's a Firefox OS port, in alpha right now, for the Galaxy Nexus. http://forum.xda-developers.com/showthread.php?t=1805282

Many Android phones can also run various other distros using chroot.

Android users who don't have root are still in jail, because even if they can run arbitrary application software, they can't choose not to run arbitrary preinstalled software, so jailbreaking is still necessary.

Similar situation for *BSD on the iPhone/iPad/iPod touch, it is a golden age for Unix/Linux but, like you say, invisible to the end-user.

Do you know whether that has helped or hindered other BSD based OSes?

It has exposed a lot of people to BSD Unix, which I think helped some.

But outside of developers and hackers, most people have so little exposure to the Unix side of MacOS X (and practically none on iOS) that it wasn't as big an impact as some of us (unrealistically?) hoped.

Apple have sponsored some major useful things that the BSDs now use e.g. LLVM/Clang.

Apple's devices are based on Mach which has a different kernel than BSD. Though it does use some BSD code it's not accurate to call it *BSD.

The kernel is a drop-in replacement for a BSD kernel, so the OS is still BSD.

There's more to iOS/OSX Unix than the kernel.

The Darwin portion is for most intents and purposes based on BSD -- particularly FreeBSD-- running on top of XNU (a hybrid Mach 3.0 microkernel with big chunks of the monolithic BSD kernel embedded).

http://en.wikipedia.org/wiki/Darwin_(operating_system)#Kerne...

Concerning the BSD portions, see this from Apple's Kernel docs: http://goo.gl/1sp69

"Integrated with Darwin is a customized version of the Berkeley Software Distribution (BSD) operating system. Darwin’s implementation of BSD includes much of the POSIX API, which higher-level apps can also use to implement basic app features. BSD serves as the basis for the file systems and networking facilities of OS X. In addition, it provides several programming interfaces and services, including: The process model (process IDs, signals, and so on); Basic security policies such as file permissions and user and group IDs; Threading support (POSIX threads); Networking support (BSD sockets)"

That accounts for an immense amount of what a kernel normally does.

Besides, I have never met a NeXT or Apple employee that didn't consider Darwin to be BSD.

It's a radically modified BSD but I think it's still a descendant.

iOS ships the darwin libc. Does it use any other major components? I'm pretty sure I remember there being no unix userspace available on iOS. Note that the stuff you quote are kernel features. And again, it's really not a BSD kernel; it has BSD roots, but they're back in the 1980's. There's no relationship between the existing projects and the Free/OpenBSD open source work.

I'm pretty sure I remember there being no unix userspace available on iOS.

Not exposed to standard users, no.

There's no relationship between the existing projects and the Free/OpenBSD open source work.

That's simply not true. There's more to the BSDs than just the kernel – all of the userland stuff is incredibly important to Apple, especially as they try to further distance themselves from the GPL. There is a relatively strong relationship between the BSDs and OS X/iOS. It's true, the lower you get, the bigger the differences, but even xnu has some relatively current BSD sources. e.g. from OpenBSD:

http://opensource.apple.com/source/xnu/xnu-2050.7.9/bsd/net/...

The last time Darwin was synced with FreeBSD was FreeBSD 5, at which point the features available in FreeBSD 5 were also made available in Darwin, such as kqueue and various other API's.

I, and some other developers I know, have been hoping that Apple in the future will sync with FreeBSD again in the future, hopefully against FreeBSD 10 to get even more feature parity in kernel interfaces available.

There is a lot of a relationship between Darwin and the FreeBSD team. Apple's firewall import from OpenBSD (pf) for example was also used by the FreeBSD team to update their port.

Just because it is not visible on the surface doesn't mean the relationship doesn't exist ...

Was "darwin libc" copied from somewhere? Maybe a BSD project? Check your facts.

Does Apple contribute back to BSD on any meaningful level?

Define 'contribute back'. I tend to think of Darwin as being a fork of BSD, so in that sense Apple doesn't commit back into BSD source repositories. But Darwin is open source, so the BSD community is free to take any ideas that please them.

AFAIK Darwin is licenced under Apple's APSL which is copyleft-based so it's likely the BSD's can't lift code from Darwin.

There was OpenDarwin though, which the developers shut down due to lack of interest and other problems. Now there's PureDarwin but development seems very slow.

Not anywhere as much as they should. A few things I do know of:

Clang/LLVM (which has substantial support from Apple) will be the primary compiler for FreeBSD.

Apple's Grand Central Dispatch (libdispatch) has been ported to FreeBSD.

http://en.wikipedia.org/wiki/Grand_Central_Dispatch

The TrustedBSD Security Event Auditing and OpenBSM are based on the Mac OS X BSM audit implementation and donated by Apple.

As well, CUPS has been ported to FreeBSD.

>As well, CUPS has been ported to FreeBSD.

Well CUPS did not originate with Apple, they bought it when they needed a mature printing system. It existed for BSD prior to Apple's purchase.

I don't know if the BSD's have made any use of anything from Apple's Darwin release? Perhaps it's not possible due to the licencing (Apple Public Source Licence) which is copyleft'ish.

There is quite a bit of sharing between Apple and FreeBSD. For example libdispatch has been ported to FreeBSD. Also when Apple imported pf into Darwin; FreeBSD used that same port to update their port of pf from OpenBSD.

As others have mentioned, stuff like the TrustedBSD and Audit framework came from Darwin, and features such as Seatbelt and sandboxes are slowly being considered or improved upon by Capsicum for example.

BSD is a funny beast in that almost know one is required to contribute anything except a license attribution in their code. Apple doesn't contribute anything to BSD because you really can't, there is no one to answer to. All you can do is stamp the license on your code and other BSD forks may use it.

The Galaxy Nexus, Galaxy Note, and Xoom do too. It's used to speed up GPS fixing (knowing your altitude helps resolve your location more quickly). A Google employee said that he didn't believe the barometer in the Galaxy Nexus was weather grade. There is an app called pressureNet[1] that reports your barometric data and then builds a map of the results that anyone can see. Kind of interesting but I don't know if it's actually useful for anything yet (I welcome Google Weather though).

[1] https://play.google.com/store/apps/details?id=ca.cumulonimbu...

I haven't been surprised about consumer Linux from the time I got a Linux-powered TiVo back in 2003. Linux with a very nice UI = win for everyone (except, I suppose the FSF who argue that Google/TiVo's usage of the GPL v2 was a perversion of the intent).

What's more interesting is that it is (assuming you have a boot-unlocked device) running free-as-in-libre software.

"Also, the S III has a barometer. How cool is that?"

The number of sensors in these portable computers is astonishing. If you look at them as a tricorder-like sensor kit, there's very little else on the market that can match them.

The fact that I have something in my pocket right now that can almost detect individual photons, EM fields (within in a limited set of bands), barometric pressure, magnetic fields, orientation, intertial measurements, planetary magnetic bearing, gravitational fields, etc. and can provide global satellite positioning to within 1m, global communications network access on several wire-free networks via voice, video or data, and can translate my voice in near-real-time, scan barcodes,

and then

can play games, movies, music, record my voice, take photos, be a flashlight, compose music, let me paint a picture, take notes, access a global encyclopedia, the internet, be a web server, an ftp server, a wireless headset to my computer, act as a strobe light, act as a visual Morse-code broadcaster, rolodex, email system, several kinds of graphing calculators...and myriad other uses...

it's absurd. I'm living, literally, in a science fiction future.

While comments about iPhone 5 being due soon is probably correct, I suspect this is the first visible sign of a tend which began quite a while ago: iPhone has peaked (just like the iPod did) and right now Android us exploding everywhere.

The question is just when we officially start treating iOS as the minority OS, the one we develop for last.

To those thinking I disregard Windows 8 tablets, I don't. I know they're out really soon now, but it's just too early to predict its impact on the market. My guess is that it will eat into both Android and iOS's market-share for tablets, but how much is impossible to say.

> The question is just when we officially start treating iOS as the minority OS, the one we develop for last.

At least for me(as an indie dev), probably never. Developing for the Android is not enjoyable at all.

I'm curious about what makes android development not enjoyable. Is it the fragmentation or development tools or low economic incentive ?

Yes.

Not trying to troll here, but can someone explain this "fragmentation" issue please? There's a single API for Android; there are also different versions of the API, but there's a same issue in iOS. Is the issue the different screen resolutions? Does all the Android developement require low-level access to hardware?

It refers to different Android versions on phones which results in developers being forced to test their applications against a lot of versions. Usually like in the case of iOS this would not have been a problem but Android updates are notoriously slow due to various reasons which means that most of the Android users are still stuck on gingerbread while jelly bean is already out. This means developers can't reliably take advantage of features added in newer versions of OS. Of course all this even before we consider the screen sizes.

Isn't this a classic issue in software developement, where you have to require or assume some minimum OS version? Remember all those labels: "Windows 95 or higher required."

And especially with web developement, where you'd have to test your site on a various number of combinations of OS, browser, and resolution.

The way Android layouts work and the backporting Google has done relieves a lot of this pain and you can now support only 2.3 and up and still get the majority of users.

If you want to get the kind of pixel-perfect layouts people have come to expect on the iPhone then yes that's a hassle. But I suspect the trend is away from that kind of design towards more flexible, information-centric designs just as it's been on the web.

20% of Android devices in the field are 2.2 or earlier. Its hard to tell a company that they should cut out 20% of their Android users in order to take advantage of the latest APIs.

Is it? My clients have all been fine with it. Going back to 2.2 isn't that hard but the people on those devices aren't likely to be very valuable customers.

Isn't it similar for pre-iphone 3 users, who also frequently get left out in the cold?

> There's a single API for Android; there are also different versions of the API, but there's a same issue in iOS.

Not really, no. http://thenextweb.com/apple/2012/03/06/why-do-developers-pre...

The biggest issue with fragmentation is the extra hardware and the screen size.

The company I work for does Android development for the gov't, and we have had a lot of issues with difference in camera support for example. Yes, there is one API, but for each camera in a phone it behaves slightly differently.

The worst still is the different screen sizes. It will look and work perfectly on one or two devices, and then you try it on a third and all of the elements are laid out differently or are rendered wrong and we have to add a work-around for that specific device.

The other issue we have found is the different versions of the libraries that various different versions of Android ship with. For example BouncyCastle is extremely outdated, or sqlite versions differ enough that it works on one device but doesn't work on the other.

We've also had issues with the phone switching between the cellular network and then wifi whereby sometimes our app will get sent up to 50 notifications in rapid fire saying it is switching between the two, which causes our app to tear down/restart the connection process. We ended up putting a timer in that waits a couple seconds and uses the last notification that came in ...

That's not to say that iOS development also doesn't have its issues. I had one app that if you used the switch on the iPad for screen orientation lock that it wouldn't output sound due to the switch not being recognised correctly. A bug that still exists in the latest version of iOS, and still hasn't been fixed (here is hoping for iOS 6 (which I have not yet installed on this device), my radar got closed as duplicate). But iOS has less of an issue with fragmentation due to the fact that most of the hardware is the same, and people with iOS devices tend to upgrade en masse to the latest version when it is released so it becomes less of an issue to have an app require a new version of iOS and no longer function on older versions of iOS (although, I am still targeting iOS 4 for now when doing consulting/client work).

Thanks for your comments. I was hoping you would elaborate on what makes the screen sizes so problematic? Web developers have had to deal with different screen sizes and different resolutions for a long time. Luckily we have CSS which makes it much easier (and especially now with media queries). Is it that Android layout doesn't give you enough control?

The issue isn't layout control, although that could definitely be better, the issue is that even with all their fancy auto-resizing everything sometimes depending on the device/OS version you HAVE to hard code a pixel perfect layout or it just won't render correctly.

This means that even if you use Android's layout tools and create automatically resizing and fitting layouts it may not work correctly, and render incorrectly for example when you move from landscape to portrait, or move from one screen to the next. Highly annoying.

We've also had bugs that we could not reproduce on any of our test devices (all 27 of them that we have now ...) but that a customer had out in the field. We then attempt to locate the same device with the same OS version to test on.

Sometimes it has to do with themes that a manufacturer has installed not working correctly... this gets annoying fast when you've got the same phone on multiple carriers and they each customise it their own way.

We've also run into some interesting documentation fragmentation points. Let's say you make an app for relatively technically unsophisticated users. You need them to drag some files to a folder on their SD card (the files are so large that you don't want to try and move them over a network). To do that, they need to mount their SD card on their computer's file system. How do you write the documentation that explains that?

Some mount to the file system as USB mass storage. Some devices call it "Disk Drive" mode. Some have a button called "Mount USB". Sometimes the button's in the notification bar. Sometimes it's in the settings. Some Android devices use MTP and the user doesn't have to do anything (but the names of the folders are different!).

You either get to write documentation like a choose-your-own-adventure novel or deal with confused users.

Take a look at http://opensignalmaps.com/reports/fragmentation.php

Android manufacturers & carriers have little incentive to get you to upgrade [Apple sidesteps them and delivers updates directly]. Do you want to target a several-years-old platform to guarantee the most access?

Screen sizes are also a challenge. iOS (iPhone + iPad) has 4 different ones to choose from. Android has... a lot more than that.

Fluid layouts usually don't work that well -- awkward spacing here, awkward wrapping there. Notice how many web frameworks use fixed-width columns (like Twitter Bootstrap).

So far I'm finding it remarkably pleasant. Some things are more difficult than iOS but some things are vastly easier. Maybe my opinion will change over time but I don't see either platform as a clear winner for the developer.

If anything I think there's an opportunity for somebody to make a Rails of mobile dev. It's a lot more work on either side than it really has to be most of the time.

This is my experience as well, it is certainly a give and take.

Java isn't my favorite language to use and I really prefer Objective-C, but Xcode is a huge annoyance as well. I would say I prefer Android development as it is my favored OS to use but neither blow the other away as far as development fun.

Your judgement seems a bit hasty. Comparing sales of a flagship phone on launch week to sales of a one year old phone (that is a spruced up 2.5 year old phone) one week before a replacement is due seems a bit... Ridiculous?

Why dont we compare all of Android sales for the last half of September and see how they look against the new iPhone.

> Why dont we compare all of Android sales for the last half of September and see how they look against the new iPhone.

Doing so wouldn't tell us anything about trends.

When would? iPhone sales are very predictable, strong now and coming up to Christmas (plus any other major contract ending time that I'm not aware of). If you're aiming to be flattering to one side it's too easy.

The technorati are still mostly running iOS and as long as the opinion makers favor it the people that make apps to impress that crowd will make iOS a priority. iOS also makes it easier to make the kind of overly fussy photoshopped UIs that people with too much time on their hands fetishize.

But Android has already gone from optional extra to mandatory second platform for any serious consumer-facing app and the sheer number of users is going to make it increasingly important for those for whom an app is just a part of a larger business, i.e. the people that will be supplying the bulk of the capital in the mobile app business.

The key question for 3rd party developers is which platform will make them more money. My understanding is that iOS is still far and away the winner by that metric, even with the smaller installed base than Android.

Until the iOS market for apps becomes small enough that developing for it is no longer guaranteed returns I will continue to develop for iOS.

The biggest issue is that on the Android Market Place (Play store or whatever they are calling it these days) I still can't make as much money as on the Apple App Store for iOS.

Releasing a free app with ads just doesn't work for me. So in the long run, until it becomes no longer cost effective it is going to be a long time before I switch to treating iOS as my secondary platform and Android as my primary.

iPhone has peaked (just like the iPod did) and right now Android us exploding everywhere.

I've been thinking the same thing, although it's hard to prove either way. I have owned the iPhone since the original version and recently got the N7 tablet and switched to the HTC 1x phone. Android hardware and software has come a long way since I tried it before buying the iPhone4 (my last phone). Before ICS, the Android HW/SW simply wasn't very good. It wasn't as responsive as the iPhone and certainly didn't feel as nice.

The question is just when we officially start treating iOS as the minority OS, the one we develop for last.

About 55% of iOS devices are iPhones. The other 45% are iPads and iPod Touches. On Android I don't think anyone knows the exact number but it's believed the vast majority of Android devices are SmartPhones. So in the end it evens about pretty close.

That plus the percentage of iOS users actually buying apps is much higher than on Android. Apple's Appstore, as far as I understand, has still the biggest market volume for developers by a good margin.

I recently bought an S III and gave it back.

At least on my device, the back cover felt cheap and flimsy.

But more importantly, that PenTile screen just looked terrible to me. I constantly noticed a faint checkerboard pattern in bright surfaces. None of my friends see it, but it drove me insane. Am I the only one to notice that?

Apart from that, not a bad device, though.

PenTile is a complete joke, Boggles the mind how anyone can ship that on a premium device.

To be fair, you have to be < 2 centimeters from the device to see the pentile on the SGS3 due to the resolution (and at 2 cm your eyes will hurt. Mine do.)

I had a SGS (pentile), SGS2 (no pentile), SGS4 (pentile), various iphones (no pentile) and others.

I can only see the pentile on the SGS.

If that's not enough, I score 21 on EU visual tests (maximum is 20, above is/are bonus points for exceptional vision)

I believe the only jokes here are this post and the parent :P

I don't understand people who say this. My friends have pentile phones, and say they don't notice, but I looked at a GSIII, and it was immediately obvious to me. The screen looks sort of grainy, and the colors are all weird.

I prefer Android, so I've been holding out for a high-end one with a great camera and a non-pentile screen. If there isn't one this year, I might have to switch :(.

Are you sure that's not due to being an OLED display? They generally have a slightly grainy look to them whether or not they are pentile.

HTC One X has great screen and camera.

I've owned a HTC One X; the camera was far inferior to the SIII's, while the screen was way better. Not happy about the tradeoff.

The color issue doesn't have much to do with pentile, that's why. Yes colors are interleaved, so it sounds like it would be logical - but that's not the case

The SAMOLEDs from Samsung have weird colors because some colors get used faster and their intensity diminushes over time.

On the S3 and S2 (which has NO pentile, still has weird colors) you can mitigate it in the color profile settings

On the S1 there are tools to completely fix the colors, yet it's a low rez pentile display. While you can see the pentile matrix on this one, specially well if close, you can't see the color difference once calibrated. There's just no way.

There's no "grainy" effect that I can see tho. The resolution's just too high to see it at normal distance (on the S3 that is. Definitely something you can see on the S1.)

I definitely see the checkerboard at slightly-closer-than-normal distances. Like, slightly less than a 30 cm / 1 foot.

Maybe that is because I am slightly short sighted? I don't know.

Uh, sorry, but at "normal" viewing distance (a foot or more?) I can tell the difference immediately. In particular, the blue tinge of the pentile display is immediately obvious when viewing pictures of people on twitter / facebook.

The HTC One X, in this regard, is the superior phone. In all other regards, the SIII is (to me).

The color issues are not pentile related, they're on all SAMOLED samsung makes. See above post.

I think the challenge Apple has, even with the 5 around the corner is this. Not sure I'll explain it as well as I'd like but it goes something like this:

iPhone "1" is mind blowingly awesome. A fantastically complete device. Subsequent updates to the hardware and OS make meaningful but incremental improvements. So an original iPhone 1 is 80% of what a 5 will be.

Android early phones are fascinating but generally pretty average - especially compared to an iPhone. Yet/Therefore every new hw and OS update provides substantial improvements over what's come before.

Therefore two things happen. Android phones close the absolute gap to iPhone helping for competitive wins. Secondly, Android new/upgrade phones excite the heck out of existing Android owners because they (along with the OS) are a real improvement.

I see iPhone 5 sales driven by people who held off getting an iPhone for the last two months because of the news. I don't see that other afterburner of sales though - 4 and 4S users who see a real driver to upgrade.

I bought the second iPhone. 2 years ago I bought an HTC Evo 4g to test the Android waters. I still use it (upgrading soon) and never really loved it. But it "converted" me for a few reasons:

- The screen size. I write iOS apps at work and get annoying having to use that tiny screen. I had hopes the iPhone5 would address this. Making the screen a little taller is a weak answer.

- More customizability. The Verge had a thread recently where they asked you to show off your phone/tablet home screen. It was a stark contrast to see the way android users personalized their home screens vs. the same grid of icons the iOS users posted. The widgets I have on my home screen tell me all kinds of things without having to run any apps.

- More control. I can run whatever app I want on it, and I can sell my apps to other android users however I want with my own billing methods if need be. Selling Android apps is easier than dealing with the iOS App Store.

I (obviously anecdotally) think the scales have tipped. I'm running into more people loving their Android phones instead of tolerating them. The latest fad apps have launched with Android counterparts (Draw Something, SongPop). When we show our software to potential clients running on a SGSIII and an iPhone4s, they all want to touch the SGSIII phone and play with it. Side by side with the screens turned on, there's no comparison.

It's a fun "fight" to watch, and we're benefitting from it.

This is obviously just a subjective opinion, but I was using both these phones the other day (in the UK), and I was blow away to the extent that the S3 just demolishes the iPhone in form factor and just overall "sexiness."

After being on iPhones for years, I've just bought an S3 for my wife. She broke the iPhone's screen last week and needed a new phone fast and.. Apple's lawsuit both made me aware of Samsung and somehow put the thought in my head that the S3 must be a pretty good phone. I'd have never even thought of buying non-Apple before.

As it turns out, it's not a perfect phone but I've been pretty wowed so far while setting the phone up and might even skip the iPhone 5 and get an S3 for myself too. Oops, Apple..

Given that the iPhone 5 will be announced next week, a lull in 4S sales isn't so strange.

I'm sure those sales numbers will plummet once Apple pays off some other judge to block the S III from being sold in the US.

Samsung's new phone is crushing last year's iPhone!

Are you Macalope? :)

I'm pretty sure we all read this same story last year, at least once. Well, at least twice, if we count the "Verizon subscribers don't really want the iPhone" stories that came out in Q2, as iPhone sales on Verizon tapered off as people expected the new iPhone to launch around July.

Or more likely anybody who bought an AT+T iPhone 4 when it launched went out of contract in July and is waiting for the launch of the '5' before signing on again or switching to another carrier.

I think number of sales is not anywhere as important as the percentage of total profits, and Apple has an overwhelming dominance in that area. The reason I say profits is more important is because you end up with more money to allocate to R&D and innovate more. In addition, the bigger your war chest is, the better of a position you are to jump on opportunities that come up.

Perhaps this is why Apple managed to come out with two revolutionary products (iPhone and iPad) within the span of just one decade, whereas competitors are mostly stuck in reactive mode.

I'm certain Galaxy S3 has tremendous success.

But I also think iPhone is in a league of its own, yesterday my jaw dropped when I saw this 'news' on TMZ http://www.tmz.com/2012/09/04/iphone-5-release-date/ . I really don't know how to interpret it, but I think it is safe to assume it is not common for a smartphone to have an entry on this kind of site.

(don't ask me how I ended-up there, I don't remember. But I'm sure ashamed of myself.)

I don't see enough numbers in the link to do the analysis, can anyone tell whether this is more due to iPhone slowdown or due to Samsung consolidating premium Android device market share with the Galaxy S III? In other words, is the iPhone really stumbling badly or is the Android market coalescing around Samsung and we are seeing the result of a 2 horse race rather than one big horse (iPhone) vs many little horses (Android)?

It's also the second most popular Android device in use at the moment, according to AppBrain[0].

[0] http://www.appbrain.com/stats/top-android-phones

It's not directly comparable. Basically iOS users have the 'choice' of one model, while Android user have dozens of models to choose from.

Enjoy your 1-2 weeks at the top, I say... The 5 is around the corner

I like that

Isn't Apple looking to ban the S3 in the US? The new iPhone (based on rumour sites) is looking like just a taller version of the current.

Apple is looking to ban all Android devices in the US and abroad. From what the iPhone 5 looks like, they've run out of ideas and are now resorting to litigation to keep their profits up.

To be fair, we should probably wait until the product launch to comment on that. Apple are very good at keeping new features secret, so who knows.

I doubt they will change the width, there must be 1000's of apps that will break or look wrong if they change that (without updating).

That's not what I mean. They could release a new integrated service, do something with the sound, make the thing out of some new material.

They almost certainly have something their sleeve to hype that they can claim competitors don't have. If they just make the screen bigger and some minor changes to the case, they will just be seen as playing catch up.

They need to do something unique like a small always on screen on the back to show statuses.

> They could release a new integrated service, do something with the sound, make the thing out of some new material.

Those are all relatively absurd examples. They tried the integrated-service route with Ping, and it was an unmitigated failure. What could they possibly do with the sound or the material?

Something tells me I'm going to be severely underwhelmed in a week or so.

They weren't intended to necessarily be good examples. By an intergated service, I meant something like Siri.

What about squeezing the device as a user input method, squeezing the sides could unlock it.

They can change the physical width of the device while making the change transparent to apps. I wouldn't put much stock into this.

We know what's going to be in iOS 6 now and that doesn't leave a whole lot of room for unexpected hardware features.

That's a good point. They surely must have something hidden up their sleeves other than a bigger screen though.

To be fair, Samsung almost certainly had a hand in banning the original iPhone for two years in South Korea.

Well, if this is true then what are the great advances in Android device hardware that Apple is suing to stop (instead of advancing their own hardware)?

I don't like how the new iPhone looks. It's just slightly taller, which is not really "bigger" as in "I want a large screen iPhone like those Android phones", and it offers no advantage to those who were writing Ph.D thesis after Ph.D thesis on how 3.5" is the "perfect size" for a touchscreen phone, especially if you want to reach it with your thumb.

Of course that's complete nonsense, as even 4" is quite tiny for a touchscreen, but hey that's what they said! And that's what they thought.

> it offers no advantage to those who were writing Ph.D thesis after Ph.D thesis on how 3.5" is the "perfect size" for a touchscreen phone

Those people would've said 2" or 6" was the perfect size, if that was what Apple had decided on. I don't think their opinions are worth that much.

These are the same people that could go on for pages explaining why copy & paste was not needed on the iPhone and then spun on a dime when it finally showed up.

Apple does get a lot of things right but some people mistake that for perfection.

Reply on news.ycombinator.com