Ballmer was not wrong about the iPhone. He said they would get 2-3% of the total phone market, not the smartphone market. He even makes that clear in the part that Gruber quoted: "... 1.3 billion phones ...". That's phones, not smartphones.
Gruber is right about Microsoft failing at phones and that there is not much they can do about it. However, I don't think Ballmer is in denial about that.
It was and is obvious that at some point all phones will be what we call smartphones now. I think Ballmer's point is that even when that happens, the iPhone will not be able to sustain its 16%. It will still have insignificant market share.
Android will not, though. It might get "60 percent or 70 percent or 80 percent", and that's the share that Ballmer would have preferred run Windows.
So yes, I think Ballmer asked the right question and got the right answer. He just doesn't have any way to respond to Android.
But when they are both free, they have to compete on quality. And on smartphones, windows loses it's main advantage -- a software library dating back to the dawn of the pc.
Microsoft cannot just take that market, but they can, and in time, probably will fight for it.
What will happen when smartphones become powerful enough to run windows? [Thinking long-term, tectonic shifts here]
There's constraints of display/keyboard size, and the form-factor encourgaes different usages (eg. less business software).
But I think people might start to dock their smartphones to large displays and keyboards at home (that's what I do with my netbook) - like a portable harddrive. Projectors might also become popular, giving a large portable display.
Or, future desktops might actually be smartphone components, wrapped in a desktop package (in the way microcomputers invaded the workstation, minicomputer and mainframe markets).
But yet he mentioned the number of total phones sold (1.3b) while talking market share so he didn't mean smartphones. Apple has a tiny sliver of the total phone market and that's what he meant. Microsoft has an even smaller sliver, but he said he wants/prefers a large chunk of that total market:
> But if you actually take a look at the 1.3 billion phones that get sold, I’d prefer to have our software in 60 percent or 70 percent or 80 percent of them, than I would to have 2 percent or 3 percent, which is what Apple might get.
He was exactly right about Apple's position but there's no obvious way Microsoft will be able gain mass adoption. Google will.
His points on Microsoft are well made. Given the strategic importance of this market it amazes me how badly they've done. Realize they beat Apple to market by around 4 years (depending on how you count the PocketPC/Phone hybrids) and Google by almost 5 .
Now they've tossed the whole thing out and started over with a product that won't be out for another 6 months and in terms of features seems to be equivalent to a first generation iPhone (which is relevant because I really don't think Microsoft is going to create a better user experience so they'd have to compete on features)
In the PC market, Apple competes on quality, Microsoft (with hardware manufacturers) competes on price. This is the world Ballmer imagined for the smartphone market, but when Google knocked the floor out by giving away the OS (or even paying the carrier a cut of the search revenue), there is no place left for Microsoft to position itself.
Unless Windows Mobile is a vastly superior product to Android (and how likely is that?), it'll be left out of the smartphone market unless they can give it away and try to make it up in Bing search revenue.
There is a place left for MS. Complete windows domain / exchange / current-telephony-OCS-or-whatever-it's-called-today integration. Automatic profiles, policies, security, calendar and contacts migration. There are loads of companies which would be happy with "locked to the domain" company smartphones, because they already have the backend ready (AD and Exchange).
Blackberry can provide a lot of those features, but not all.
If Microsoft goes after RIM and leverages their enterprise footprint they could quickly become viable. If they compete in the consumer market with Apple and Google they are toast.
I'm not sure that's true. The Google I/O conference made a pretty compelling case for Android's Enterprise chops and Microsoft has (ironically) backed itself into a corner by making Exchange and other servers open to non-Microsoft clients.
As someone who runs an enterprise's It dept. I'm not sure I can see Enterprise features that Microsoft could compete on (Security, Exchange Support and Remote Wipe are my real concerns and all seem to be covered by Froyo)
I think it is still too early to say that. Microsoft has yet to recoup the money they invested in the Xbox, and they probably won't anytime soon. And while they are still ahead of Sony in the console race, the PS3 has picked up quite a bit lately. Microsoft could very well finish the race dead last at the end of the current generation, with Sony at the top spot and Nintendo a close second.
The only problem is they have to actually do the catching up.
I'm not sure they can.
This market is much more complex, with a faster-moving platform, more hardware innovation, and competition (Apple, Nokia, RIM and Google) that can't be bullied or out-spent.
Maybe I'm less anti-Microsoft than most (disclaimer: I worked there for a summer), but since WPS7 will be out before my current iPhone/AT&T contract ends, my plan is to switch to a Windows phone. I think the UI looks great (though practical usability is still in question), and I've always preferred Outlook to Gmail. My main gripe would be using an IE7-based browser, but as long as I can install Opera Mobile that shouldn't matter. I also wouldn't mind switching to the Zune platform - I've heard good things about the Zune desktop client, and iTunes has numerous problems. (I don't think Android offers any support for smart playlists, which would be a real pain for me.) I guess it might be obvious from the above that doing the core tasks well (email, browsing, music) is more important to me than how many googols of apps are available...plus, by using XNA as a dev platform, I suspect the learning curve for creating WPS7 apps will be lower than for iPhone/Android, because there are already tons of XNA devs.
Of course, this is all predicated on WPS7 meeting expectations, the pretty UI actually being usable, etc. Otherwise, Android. So I guess the point is that, at least in my case, Microsoft is still in the game, but they have only one chance to deliver.
Interesting you find the UI to be pretty. From the stuff I saw it looked very hacked together and disgumbled. I'm looking forward to seeing an actual phone though. I have no interest in switching (I just got an Android based phone and love it) but I'd like to see what the actual user experience turns out to be.
Aesthetics and UX are definitely linked; as you say, it's really hard to evaluate either until one can actually get a feel for the navigation, etc. But I think that as long as the UX is coherent, the visuals will be there.
The wm 7 ui is typography based with minimal visual flourishes and virtually no icons.
Its very different than any mobile ui out there. I really enjoy the experience on my zune hd and that's a preview of what wm 7 will be based around. From what I've experienced (people who have never seen this type of ui before), people seem to be intrigued and curious by it. If the mobile team at microsoft is paying attention to details in the way the zune team did leading up to the launch of the zune hd, then a lot of doubts people have about this platform will be gone.
Opera Mini is a great app for reading news and such, but as your main browser? It doesn't execute 99% of JavaScript. You can't load any smartphone targeted websites with your smartphone. Not very smart.
The first-gen iPhone had a really good web-browser. Even if Microsoft miraculously finishes IE9 and ports it to WinMo 7 in time for the launch, it's still going to be a substandard mobile browser, and thanks to the all-Silverlight development platform, there will be no other browsers available.
Although syncing with a computer may feel retrograde, I'm sure theres a lot of people, including myself, that would rather keep their data due to privacy issues instead of relying on Google or Apple to store it. However, I think most people would prefer cloud syncing since it is so easy for the user.
It'd better be an N Wifi if I'm going to sync 16GB of music to a new device, and even that's going to take quite a while. Of course, USB 2.0 isn't really up to the task either - but at least it's faster by a significant factor.
Downloading the initial sync over 3g isn't something I'd do if I could avoid it, but incremental syncs would be just fine. Even adding an album or two wouldn't take -that- long, and if the network can't handle a few downloads now and then it's just plain broken. Besides, everyone over here in Sweden is selling mobile broadband like there's no tomorrow, so their networks should be up to it.
Ah how I wish the US had Sweden's net/mobile infrastructure.
Carriers here like to complain that people are downloading too much rather than just upgrading the infrastructure to deal with it. I shudder to think what would happen if people started downloading over the air on a serious scale, it's already pretty bad with normal web surfing.
For example, when you buy an AppleTV, iTunes on the same network will detect that there is a new AppleTV that hasn't been synced. Then you synchronize a 5-digit code between them and off it goes syncing over the network.
I would love to see iPhones and iPads show up in the same way and not require a sync immediately. This would be right on track with the life-without-wires they were pushing with the MacBook Air.
I suspect it comes back to Job's strategy for the Mac back in 2001 (I think it was 2001, not sure). He talked about how computers had gone through several generations, and that we were moving into an era where the PC/Mac was the "digital hub" that kept all your devices in sync and playing nicely with each other. Google would probably argue that that's the responsibility of their servers.
With the iPhone and iPad doing as well as they are, I imagine conversations are being had at Apple about what the future of the Mac really is. Even if it has much of a future at all.
I guess it'll turn out to be iTunes.com, your Mac will stop being the hub, and it'll move to the cloud.
Apple are building a new data centre which apparently will be built for the cloud, my guess it's going to be the backbone of iTunes.com and their other online services. http://www.wired.com/gadgetlab/2009/08/apple-cloud/
I installed it a couple of days ago and it works like a charm. I also setup 'Activator' so I can launch the wifisync app by pressing the lock button for about a second and it will automatically sync to my Mac.
His footnote is interesting: "Although there’s still no decent Android-based equivalent to the iPod Touch."
He's right. Based on price, there is no equivalent.
On features, the equivalent is a Google Nexus One without a SIM card.
Trouble is, that's about to not be available from Google directly. I'm wondering whether we'll be able to buy an N1 without a contract ever again.
Disclosure: I bought an N1 directly from Google and loved the experience of not having to talk to a carrier or any staff member at a generic retail outlet.
Although people who buy iTouches love all the apps and the ability to browse the web whenever they have WiFi, in the end, they're buying an mp3 player. They're buying it so that they can listen to music.
Not only is the N1 much more expensive, but it also has a much worse music-playing interface. Until someone fixes the Android music player or writes their own (luckily, an app can have all the privileges that the core apps have, so anyone could just write the Awesome Music Player app and put it in the market), the demand for a good Android iTouch competitor won't be there.
But I do think that getting an iTouch-like Android device out there is important. There are only so many people who are willing to pay $30/month for Internet on their phone, but pretty much everyone is willing to pay $200-$300 for a nice mp3 player with no recurring cost. And it's important to get as many Android devices out there as possible so that consumers believe that it's a good option (in many people's minds, popular == good) and app developers see more incentive to write good apps.
You wrote: Although people who buy iTouches love all the apps and the ability to browse the web whenever they have WiFi, in the end, they're buying an mp3 player. They're buying it so that they can listen to music.
I'm not sure you're correct. I have a simless old iPhone, equivalent to an iPod Touch, and it runs apps exclusively. It has barely been used for playing music.
In fact, the beauty of apps means the hardwares purpose is completely malleable. In my case, it is primarily a mini-Bloomberg reader and alarm clock.
You and others on HN are probably not the target for the iPod Touch. Just spend some time with my sister and her friends and you'll quickly see that a bunch of them have iPod Touches and most of them are just used for playing music (with a few games here and there)
My sister also has an iPhone, but she's not really tech savvy enough to download apps and actually use them. She's downloaded a few apps, but most of the time she just uses her iPhone as .... a phone.
Although people who buy iTouches love all the apps and the ability to browse the web whenever they have WiFi, in the end, they're buying an mp3 player. They're buying it so that they can listen to music.
I don't think you are right.
The touch music player interface is good, but not as good as my old 40GB iPod Video. Man, it has been almost two years now and I still miss the clickwheel.
I got my iPod touch because of the large screen, the wi-fi and the apps. I don't even hear as much music as I've used to because I really hate unblocking the screen to skip songs and change volume.
>> Although people who buy iTouches love all the apps and the ability to browse the web whenever they have WiFi, in the end, they're buying an mp3 player. They're buying it so that they can listen to music.
Our family has three iPod touches with no music on ANY of them. If you speak to families with young children, you'll find there are millions of iPod Touches out there being used as game/app/video platforms.
> Not only is the N1 much more expensive, but it also has a much worse music-playing interface. Until someone fixes the Android music player or writes their own (luckily, an app can have all the privileges that the core apps have, so anyone could just write the Awesome Music Player app and put it in the market)
I really want someone to fix this. Android needs a WinAmp type thing to play and catalogue any music.
The built-in player I've abandoned as it doesn't support FLAC.
I'm using andless ( http://code.google.com/p/andless/ ) which supports FLAC and is probably the cleanest interface. But it lacks a bit of polish (smooth transitions between screens, a tactile feel).
It also lacks album artwork read from meta-data and stuff like that. And it stutters occasionally on playback.
I really don't know about music file formats but just because I want a decent music player once I have some time I'm going to see if I can find any open source libraries elsewhere that I can plug in to see whether I can figure out how to fill these shortcomings.
Justin Frankel if you're reading this... we all know you could do this in a week or two and improve our lives in the same way WinAmp originally transformed the music experience on Windows.
[edit]OK, So I just realised I should do something about this, so I've just asked JF if he'd consider it: http://www.askjf.com/index.php?q=662s and I'm going to grab the Android SDK and work out just how inept I am at this. I guess the andless code is a good place to start.[/edit]
Might I ask why you think the Android Music app is bad? From what I can tell, it's nearly identical to the iPod app on iP* (it's more awkward getting back to your library from the Now Playing screen, but that's all).
The interface just isn't as nice overall. It gets the job done, but it's not as intuitive or pretty as the iPhone's player.
Also, the vast majority of people have their music in iTunes. The fact that they can't just plug in the device and tell iTunes to sync might be a turnoff
anecdote: friends and family have spent at least 20 thousand dollars on music in itunes. The people who even know alternatives exist are only tech people and they don't spend much money on music.
My iPod touch is most often used as a remote control for media player classic on my PC. I still use my nano for music though. Click wheel is much more usable and device is much smaller and lighter.
The current iPod touch is on the same level as the 3GS, maybe even a bit faster. It has the same amount of RAM and the CPU might be a bit better.
iPods are released yearly so there is no reason why they shouldn’t be just as fast and have just as much RAM as iPhones. It’s right, though, that the oldest iPod touch (from 2007) won’t get the update. 2nd and 3rd generation iPod touches will but 2nd generation iPod touches won’t get multitasking (just like their brother, the 2nd generation iPhone aka iPhone 3G).
AFAIK it's a combination of the hardware not being good enough, and Apple not really trying to optimize for those devices (they'd be happy if you bought a new one). The iPod Touch went through several iterations, just like the iPhone; the latest iPod Touch is actually a bit higher specced than the latest iPhone.
Note that this is why standalone MP3 players have been unable to garner marketshare in Japan: everyone's cell phones were already their mp3 player (and they bought the music OTA via their phone, to boot). As is, the days of standalone music players are likely numbered.
What about the ipod classic? For the moment this gives a lot bigger capacity than you can get with a phone and for a lot cheaper price. Plus it's very straight forward to play music.
Although if you were to only carry around one device it would be a phone, just personally everywhere I go that I would usually want a mp3 player I carry a bag making carrying two devices a non issue.
It's not just Microsoft who seem to be out of the game so to speak. What's going on with Nokia? It's taking them forever to come up with a device and/or platform that seems like a credible competitor to Android and iPhone OS.
Nokia switched tracks, abandoning Symbian for the smartphones and putting their chips on Meego embedded Linux and Qt for the UI.
There's a huge effort to make Qt into a higher level framework that can be used like Cocoa Touch or the Android UI elements. The Qt Declarative/Quick stuff is being advanced as quickly as they can work, but after working with the latest stuff they have a ways to go.
I have a hunch Nokia thinks they'll be able to quickly catch up.
And that makes me somewhat sad. I've been a user of Nokia for 10 years and I allways loved those devices. When I purchased my last phone, 1.5 years ago, I gave them a vote of confidence and got an E71 instead of an iPhone.
It's funny because John starts with "Google is doing this fine" and "Apple is doing that fine", and "we are all gonna get better smartphones with this kind of rilvalry" and then when you think he is goint to add Microsoft into the rivalry: "Oh, yeah, remember Microsoft? it seems like they don't remember the smartphone market"
Google is upping the ante on the iPhone here, though, by adding cloud-based data backup for Android applications
How many apps will actually support this feature? It doesn't sound like it's an OS level feature transparent to third party apps. Am I going to have to keep track of which apps support it and which don't?
> One area where the iPhone has been far ahead of Android is in terms of backing up and restoring data. Buy a new iPhone, or install a major OS update, and when you re-sync with iTunes on your desktop, all your apps and data are re-installed.
Gruber doesn't seem to get that Android is cloud based: buy a new Android phone, sign in, and your contacts, apps and bookmarks are available immediately, no finding a PC to tether to.
Music and (for some odd reason) text messages are currently excluded though.
I think he does, but he acknowledges that data isn't included in the previous cloud based offering.
I was blown away when I first got an N1 (my first android) and signed in on startup and immediately had my contacts, email, calendar, etc wired up.
No computer required. No special software required. It just works.
What Google are doing now is taking that further and making that include data like music, pictures, TXT messages, etc. So now when you replace your android that stuff automatically comes across.
What would be lovely is that instead of just sync, whether we could manage the older devices. i.e: Get a new mobile, sync it, declare the old mobile to be wiped and have the old mobile hear that and purge itself.
This helps on the security side, as you could remotely wipe a stolen device. But it also helps with re-selling used phones safe in the knowledge that your data is intact and you're not leaving any on the device.
I think it's the next obvious gap filler in the cloud-owning-the-data space. We can move and sync, now clean up behind us.
Well, that and the fact that if data is persisted across all devices that you eventually will exceed the storage capacity of the device itself. But I'm guessing that network speed will keep pace with this and by the time we're at that crunch point Google will be able to serve that stuff on-demand and just use the phone as a local cache rather than as the persistent storage itself.
Gruber's thoughts can be summed up as: Google vs Apple, bring it on, but where's Microsoft? While he isn't wrong, it's not exactly much more than some common sense.
I also think he _seriously_ underestimates Microsoft.
It is certainly possible that Microsoft could turn around, but I doubt it: they have a huge cash reserve, yes but that is not going to make them relevant.
What would be interesting would be if they got an entire new board + CEO, Steve Ballmer isn't ready to lead and never will be.
Microsoft will be 3.5 years late to the smartphone party when windows phone 7 series will eventually be released. People would have already made their choice between iPhone and Android.
Apple will be 5 years late to the _smartphone_ party when the iPhone will eventually be released. People would have already made their choice between Blackberry and Nokia.
It's shocking how irrelevant they are in most of the things that are getting a lot of press these days. It's hard for me to imagine the scenario where Windows ever becomes a big player on mobile.
I wonder where they'd be if they'd have been broken up by the US government a decade ago.
I remember the early 2000s. I devoutly hoped that Linux would get widely used in some way, even if it never really took off on the desktop. I wanted open source to be fairly common and well-accepted, and not just something that weird people wrote and talked about on Slashdot. And hacker-friendly open-source Linux-based smartphones would be great, but that was just too implausibly awesome to even hope for. I wanted pleasant languages like Python to get serious commercial use.
Now, all those things have happened or are happening. Want to get on board with cloud hosting in any way? The standard options involve Linux VM images. Languages like Ruby and Python are mainstream and well-supported. Android is disrupting the smartphone market, and has the kind of widespread commercial backing I never would have thought possible for an open, Linux-based thing. Microsoft is no longer an unstoppable leviathan; they've stopped being scary.
Occasionally I'll reflect on all this and chortle happily. I wasn't optimistic enough back in the early 2000s, and that's surprising.
What is really funny is that we never really hit the year of Linux on the desktop. I mean, this has been speculation for years... Will this be the year that Linux really starts to make in roads in the Desktop market? For the most part, the answer has always been no. Redhat and Novell/Suse have all but left the desktop (commercially, not counting Fedora). Ubuntu has all but taken over the desktop Linux market. Overall, Linux Desktop experience has increased substantially since 2000, but it has never really caught up with Windows (now 7) and the Mac. (I actually think that the Mac has hurt Linux adoption on the Desktop more than Windows.
So, I too find it quite humorous that Linux is poised to be a, if not the major player in the next generation of computing - the portable/dedicated devices. But what I find more amusing is that it hasn't been the success of Linux on the desktop that has made this possible, it has been the decline in the importance of the desktop.
So, 2010 may finally be the year of Linux, but it still won't be the year of Linux on the desktop. And that's not necessarily a bad thing.
It's amusing, but not a very surprising outlook for the tech industry; platform shifts are opportunities for competitors to break in.
Linux came a little bit too late to make inroads on the desktop vs. the incumbent DOS/Windows platform of the early 90's - if it, and the GNU tools, had reached the 1991 level of development in the mid-80s when the PC clones first took hold, that story might have been a different one - but in the real world, it still devastated the competition on servers when the Web started its massive post-1994 growth.
Now we're hitting the juncture where, again, a new platform category is poised for growth; recall how netbooks started as Linux boxes, and only ended up running Windows after a combination of effort(Win7), strongarming(subsidies and licensing agreements), and initial advantages(the desktop ecosystem, which is largely netbook-compatible) from Microsoft. This time around, those tactics can't work.
It might actually work out that the desktop still shifts towards Linux in the end; the "bottom-up" nature of progress in tech means that people are going to want a desktop that does everything their mobile environment does (plus desktop things). Hence, desktops will start running Android or a compatible variant...which would naturally work out to mean you'd be running Linux, even if it isn't today's Linux environments.
The interesting things here are (1) that microsoft isn't even mentioned as one of the protagonists in the main battle (that's clearly apple vs google, (2) that google has done to microsoft what microsoft did to netscape, give away the product the other party hopes to sell.
It appears to me the biggest news from google I/O was the release of flash on android, just a couple weeks after Steve Jobs says it wasn't practical. Google just didn't demo it and give a release date several months away, but they delivered it (with reports of folks getting 2.2 pushed already). Seems like Apple needs to reconsider their position on Flash... already.
Flash could run at 100% native app speeds with no battery hit and Apple still wouldn't support it. Why? It's not part of their overall plan to give Adobe power over the platform.
So no, Apple doesn't need to (and won't) reconsider their position on Flash.
Then people would build Flash apps and just run it in browser and bypass App Store all together. If that happens, not only does Apple give the control to Adobe (and have to rely on Adobe to support iPhone Flash plugin, which is absolutely not what Apple wants), it also loses the benefits of charging 30% of the revenue if the apps would have been sold in App Store.
I know Apple doesn't want to support a flash plugin and was burned by Adobe in the past, but with Android supporting flash now Steve Jobs has less excuses to tell users why they can't view flash on websites. Most users won't understand why they're Android friends can access flash sites, but they can't. Apple could work with Adobe to get a flash plugin on the iPhone, they just don't want to.
Also, people can make flash apps in the browser and that's fine. It wouldn't hurt AppStore sales, cause native apps would be better.
Apple will create more excuses... I'm thinking security issues will be the next focus.
Flash could (and probably would) hurt AppStore sales. It probably wouldn't be significant but some developers would choose to use or keep using Flash rather than creating a native App.
I up-vote this, you are specifying what does it appears to you. I also agree with you, it is true that probably flash isn't the future from the web, but at least is part of the present.
Take the current URL from your PC web browser and push it to your device, over the air. If it’s a web page, it’ll open in the Android web browser; if it’s a Google Maps URL, it’ll open in the Android Maps app.
Meaning that now, not only can spyware pop up porn on my parents' home computer, it can also push porn to my kid sister's cell phone? Ditto for apps... I imagine spyware is a thing of the past for most Googlers, but it certainly isn't for the majority of users.
Isn't there going to be security with this feature? I don't think any app/website can send remote commands without having your credentials. If spyware has your kid sister's Google credentials, pop-ups are the last of her worries.
A program that keylogs your machine and grabs your passwords is a virus. You're right they are fairly common but once again a "popup" on your phone from a virus is the least of your worries.
I'm not sure Apple is ahead on backup & restore with the iPhone / iPad.
You shouldn't need a desktop (with an OS that runs iTunes) in the first place. I read that the first thing you need to do with iPad is to connect it to iTunes.
What Gruber was getting at is that Apple has a far better syncing strategy since Android wipes everything between updates and Apple neatly brings all your data with you.
That said, the cloud-based syncing that Android will eventually ship looks like the "right" solution. I doubt anyone at Apple considers USB syncing to be the end game, though; I'd be shocked if they didn't have some sort of over-the-air syncing in the pipeline. The question in my mind is will they fix this in 2010 or will they fix this a few years from now. People won't want to stay tethered that much longer.
As a sidenote- syncing over USB makes sense for the iPhone, in some respects, but the iPad lends itself to be a standalone computer. Tying that to another centralized hub is a hassle, and that's where an Android-esque cloud sync would be great.
Ah; I stand corrected... I misread what Gruber wrote. Looks like Android handles updates fine, but there's no way to cross-load data between devices. So buying the latest and greatest hardware leaves you in the cold, at least until Froyo.
Android as a platform doesn't require data wipes to update. A few handset manufacturers have chosen to go this route for certain devices but other devices such as the Nexus One preserve all user data during an update. The cloud-based syncing will certainly improve things for users migrating from one device to another.
Personally, I feel more comfortable having total control over my data first, and then choosing what I want to share with "the cloud". The direct personal control over syncing to a device I also control is the more critical factor over not requiring a separate computer for me.
According to http://www.zdnet.com/blog/btl/apple-iphone-smartphone-market... in the first quarter of 2010 Apple had 16.1% of smartphones and smartphones were 18.8% of mobile phones. That comes out to almost exactly 3%.
Ballmer also wasn't wrong about Windows Mobile. The "60 percent or 70 percent or 80 percent" was not a prediction, it was just what he would prefer.