That was in may, now it's 160,000 daily
The desktop market was won by Microsoft in large part because of its appeal to the broadest segment of the developer population — cue the Steve Ballmer fight song, complete with sweat stains and manic enthusiasm. In mobile, Microsoft is AWOL because it has failed to attract developers in meaningful numbers. Instead, Apple and …
That was in may, now it's 160,000 daily
Spot on article, and it reminds me yet further of The Perplexing Case of Jobs.
Why, with all of computing history against him, does Steve Jobs persist in pushing a closed, restrictive platform? Sure, it has worked in the short (and maybe even medium) term but in the long run it's only ever going to be a losing strategy.
Maybe that's enough for him.
Let's hope that history isn't on Android's side - if it goes the way of Windows, it'll be nothing more than a copy of innovation and deliver nothing but pain, misery and at best mediocrity, and then it'll hit a peak and start a rapid downfall.
I don't agree with your dig at Apple, however, I think devices like the iPhone will flourish - the locked down environment isn't as locked down as developers would have you think. For the average consumer, ease of use is far and away better than an environment which allows the tinkering of the techy crowd.
... to stay closed and restrictive, grab the short- to maybe medium-term gains, then liberalise as necessary. Though he doesn't seem to be doing that. His plan actually seems to be, in the short to medium term, to leverage Apple's vertical integration, marketing and user interface nouse to jump into domination in nascent markets with the same software platform. How many Android tablets are there in the world? Probably a few thousand (since it's restricted to weird Chinese knock-offs for the time being; this will change). How many iOS tablets are there in the world? A week or so ago it was 3 million, with the rate increasing. I reckon he can pull off at least three more years of his current behaviour on that achievement alone.
Personally I hate the direction Apple, and now Google, have sent the mobile device market. Beautiful big screens - but dead-duck battery life ... touch screen keyboards - and massively reduced WPM typing ability. Hardly progress!
For both Apple and Google (indirectly through Symbian), everything has moved from productivity, to consumer orientation as a platform for those companies to generate continuous revenue streams. If you buy Apple, you are buying their media delivery platform - Google, their add delivery platform. Of course, this is getting fuzzy with both Google and Apple expanding into adds and media respectively.
What they have done to the mobile device industry is the equivalent to Sony selling televisions that only deliver Sony stable movies!
In this regard, Apple is worse insofar as they have, on numerous occasions, blatently exploited their control to stop other companies from monetizing the platform in areas they see potential for themselves to. Google so far has not, and it seems do not intend to. The closed approach of Apple WILL kill innovation long term as the prime aim is to grow and protect Apple's revenue streams which means doing everything Apple's way and for Apple's benefit regardless of whether it is in the consumers best interest or beneficial to them or not.
Give me (more) open any day!
Interesting comments. I was talking to some mobile industry analysts the other day and their view was that Apple lives by new product launches. (Makes sense given Apple device price-points giving them a limited market, and the high ratio of repeat sales to a loyal following?) They needed the iPad as a new device to keep growing. You are probably right that the iPhone is part of a strategy for another 2-3 years and then it will be of secondary interest and kept on life support.
I am a Linux person, but learned to edit video on Macs. I recently got an iPod Touch 3G 32 GB Newest (to listen to web radio via mobile 3G/WiFi.)
The way Apple treats its customers like idiots is truly eye-opening. While Microsoft has all the industrial style of East Berlin in the '70s, they are not nearly as creepy as the iCult.
I have a couple friends who, due to Leo LaPorte's scathing review of iP4, made the leap to xdroid phones and they have no regrets. A month ago they were loyal iFanbois.
Given that Steve (and Apple shareholders) are richer than most, Mac's are still around 25 years without being fully opened (but use open source where it helps), iPods, iPhones and iPads have all largely taken the mindshare (and profits) from their respective markets I would say closed restrictive platform seems to have worked pretty well.
Depending on what you mean by long run of course, but Mac's and then iPods have been around a while. You could argue Mac might have been more successful if they had not open; but other systems have come and gone and they are still around churning out a nice profit.
I suspect the market has room both for commodity open-source gear and some closed expensive designer kit like Apple stuff.
I don't develop for Apple or Android, but I have installed the Android dev system to see what it was like, and it didn't cost me anything to do that. That's one advantage over Apple I presume. As to the system itself, I have been quite impressed, it integrates well with Eclipse, and the simulator was a bit slow but functional. I was able to knock up a quick Android app with about 1/2hr of deciding to download the dev kit. This mirrors another reason why Android and Windows dev are similar - MS produced Visual Studio which was easy to use and enabled rapid (ish) development and debugging of apps, in a sensible IDE. Just like Google have done with Android.
As stated - I don;t have any experience of the Apple dev system, so would be interested to see what people think its like compared with the Android system.
"That's one advantage over Apple I presume."
The iOS SDK is free. There will be a charge if and when you build an app and want to submit it to the App Store, however.
The SDK may be free but you need a Mac to run it on.
"The iOS SDK is free. There will be a charge if and when you build an app and want to submit it to the App Store, however."
But the Apple Mac hardware to run the iOS SDK on is anything but free. For a PC user that wants to dabble with some casual mobile app development in their own time and on their own dime this is a bit of an obstacle. The Android SDK is equally at home on Windows, OS X and Linux.
But the mandatory Mac to run it on is not. If you're not already a Mac owner.
With comments like yours and others above, you will always have a comeback because you have already decided that Apple is bad and not Apple is good.
OMG, you have buy some equipment and you can't develop that next iPhone killer app on a netbook. FFS, find a better reason to stay away from Apple.
And are you *seriously* claiming Windows PCs are "free" now?
He's saying that in order to develop for Android you need a computer running Windows, Mac OSX or Linux.
In other words, if you own a computer capable of rendering a web page, you own a computer capable of running the Android SDK.
The same is not true for iPhone/iPod Touch or for Windows Mobile, both of which require a specific operating system.
You can get the SDK and the XCode IDE tools for free, although you must register, however if you want to get your app running on a physical device you must sign up to the paid developer program at $99 or some number of euros that I don't recall. It's not just access to the app store that you're buying.
Free devs can only get the current release versions of the tools, SDK and docs, paid ones get the betas well in advance of public release as well as access to the 'confidential' sections of the developer forums.
It is worth noting that in both cases you will be asked to indicate your acceptance of Apple's developer agreement, which as far as I recall is word for word the same for both free and paid devs. I cannot stress enough that you should take the time to read every word of it, and if you are uncomfortable with legalese show it to someone who isn't.
If you haven't seen many corporate NDAs before, then prepare to feel queasy, have a good long think about wheather you are comfortable working under what many people will regard as quite unreasonable terms but IME are boilerplate for corporate NDAs.
And yes, you'll have to buy a Mac which will set you back a few quid, but given the seriousness of the agreement you will be required to enter into, starting iPhone development is not something you should be doing on a whim. Despite much recent wibbling, it is unlikely that Apple's corporate legal ninjas will have put together an agreement that will fall down the first time someone throws UCTA at it.
Nobody said that; you just missed the point entirely so your choice of icon suits you very well indeed.
Obviously anyone interested in developing for a phone will own a computer, that's just common sense. Of course not all of those people own a Mac.
The point is to develop for Android, it doesn't matter what type of computer it is (Mac, Windows, Linux), the person is free to go for it without any additional financial commitment.
To develop for the iPhone, you have to own a Mac. If you are one of the many who doesn't, you must buy one first (or pirate OS X).
Something tells me those who complain don't already own a Mac. Do you get it now or was that not simple enough?
We don't want to see anyboby upsetting anybody. I always like STB's postings... makes me feel like a freaking genius.
"Perhaps, as Dave Rosenberg writes, Android needs a touch of the Jobsian strongman, someone to cut out the clutter of Android apps and enforce quality control."
To follow your Microsoft analogy, this issue has never mattered to Windows, or, maybe even to a greater extent, to Linux. There's good software for both, and there's an awful lot of rubbish too. But, you pays your money and takes your choice. Very few people complain that there's too much choice (with the possible exception of video and audio standards, which is a mess).
Also, Microsoft, for all their ills, never tried to police the market at all. It has pretty-much always (except for maybe deliberately trying to block rival versions of DOS in the old days) left the software market to itself, and this has, quite obviously, paid off for it. Apple on the other hand seem to be coming up with more and more restrictions, almost on a weekly basis. This, coupled with the seeming shifting sands of the iPhone API, (another thing that MS have been pretty good at maintaining; Linux less-so - which audio standard are we using this week?) will inevitably put-off developers. Why put huge effort in to writing something when you don't even know if it will work (or be allowed to work!) a year from now?
This is a comment I received from a customer who bought one of my apps. He found a bug which I verified and fixed. I emailed him to say that the new version was available in the Marketplace with the bug fix. It's looking like I may have to start writing iPhone apps, and I'm dreading that day.
"Incidentally, I must say I'm impressed with the way Google allows developers like yourself to download apps without the limitations Apple adopts for their iPhones."
* I say more open, as I'm fully aware that apps can be removed from the store and killed remotely. You're still are allowed to install an app from outside the store without voiding your warranty though.
but Apple and it developers will take most of the money.
For the very same reason the author pointed out. Android will win over in emerging economies.
Places where software piracy is endemic, where people have far less money to spend on apps and where ad revenues will be low compared to the developed world.
The only party to make any real money from Android will be Google's ad business. For the rest I fear it will turn into another race to the bottom.
I'm not sure how you can reach that conclusion.
The very markets you finger as high piracy are those same markets where Apple products are stratospherically overpriced and therefore have a tiny market share. I'm going to ignore them completely for the purposes of this reply, assuming revenue ~£0.00 on both sides.
The good/popular applications on the iPhone are swiftly starting to appear on the Android store as well, at similar prices. Angry Birds (coming soon) and F1 Timing (last week) are just two that spring to mind.
Nobody seems to dispute that Android smartphone sales will outstrip iPhone sales soon, and end up dominating this space. On face value then, more app sales too, and more profits to developers. The available market for Android applications will be huge, so sales revenues will end up being bigger on Android than they are on iPhone.
Apple will earn more per phone than the Android handsets on sale. But the volumes shipped could easily compensate for the lower profit per handset. You also have to remember that it is cheaper for the networks to put an Android smartphone in the hands of a customer than an iPhone. Therefore it will pentrate further down the social scale than the iPhone in every market. And those people will still buy apps.
As a former developer I can say that what I wanted to happen in the OS adoption patterns of the public very rarely happened in practice.
The problem with the mobile landscape is that something which catches the public's imagination can change the whole playing field in less than 12 months. Look what happened with the RAZR and compare that with the relative flop that was the RAZR-V2 and ROKR.
It's entirely possible that Nokia manage to get their act together and make an extremely desirable phone that works properly so that Simbian becomes the next big thing (again) or that Sony make a WinMo7 phone so desirable to the public as to make WM7 huge.
The only thing you can say about the future is that it's rarely what's predicted.
Now, if you'll excuse me I need to don my little silver cape so I can go on my day trip to the moon. Oh, wait, that prediction was wrong too.
Developer fees for Android: £17 once off payment.
Apple yearly recurring fees of about £75 a year.
Android: Free and pretty good (Eclipse plugin works well as does the Android device emulator)
Apple - pretty rubbish and mac-based.
Android: none, you post your app, it's there right away.
Iphone, 4-6 weeks if you are lucky enough to get it approved.
Android: Lots, you can do pretty much anything you want.
Apple: Not very much. It Jobs doesn't like it, or it may hurt their income streams, then it's banned.
Android is just about beating iOS in sales, and it's only going to get worse for Apple, by the end of the year it's easy to see a scenario where Android is crucifying it.
Crucifying Apple. Hmmm. Not bloody likely. Look at the early success of the iPhone 4. Now I'm not saying that Android phones won't sell more, but what you imply is that Android will not only sell more, but at the same time the iPhone will stop selling. That will not happen. The iPhone will always sell lots and lots, and there will be a huge market for apps - that will not change.
This competition thing is so stupid. Nothing is better merely for sellng more. And it's even sillier when you compare software to hardware.
The one thing that Android needs to manage is this fragmentation issue that is becoming more and more of a big issue.
Do I actually believe that? No. Am I offering it as an unqualified subjective statement in order to make a point about exactly how persuasive that sort of argument is? Yes.
I chose to judge the accuracy of your post entirely on the grounds of "Iphone, 4-6 weeks if you are lucky enough to get it approved" — you seem not to know what the device is called and seem to have missed the memo that 95% of apps submitted are approved within seven days. It looks entirely like you're asserting things you wish were true just to kick up an argument.
Competition isn't stupid. It drives change and improvements.
Whatever anybody may feel about Apple, iPhones and the nutters who fanatically support them (compared to normal users) is that they have substantially changed the mobile market for the better.
1) What are the sales like per handset, of apps though? Not seen any figures yet?
2) the only trouble with open source OS, etc is that there will always be someone who is willing to spend their time for free to produce an app, especially if yours is any good.
3) People can copy your app to all their friends
4) Google will remove non-malicious, no affecting apps with their kill
trouble is the only other market in town has Jobs in charge... :(
>> Android: Free and pretty good (Eclipse plugin works well as does the Android device emulator)
>> Apple - pretty rubbish and mac-based.
Last time I looked at the Android dev stuff, it was utter cack. Eclipse is vile, but that's personal opinion. Even when I was developing Java I paid for an IDE rather than suffer Eclipse.
Android documentation is either outdated, partial, splattered across the web on a dozen contradictory blog posts, or simply non-existent.
Apple's devtools are slick and well integrated, and *everything* is documented. You might not much like XCode, but that's, again, personal choice.
I've got a game on the app store. It took me less than 2 days to port it, and it runs well. I'm currently struggling with getting half the framerate on Android.
Java is a shite language for developing apps that have to perform on a platform with limited resources. Yeah, Froyo delivers a new Dalvik with an up to 450% increase in performance, but it's taken 2 years to get there, and it's only out on a couple of devices. And even then, it doesn't stop the "hic" of garbage collection.
For what it's worth, I don't see either Android or iOS as being a sustainable long-term platform. iOS is too limited (in general terms, not in terms of "what you can get approved"), and Android doesn't seem to be interested in being anything other than a "me too" iOS clone. That comes from seeing real innovation being passed by over and over again.
As for android taking over in countries where expense is everything? Dunno, Linux was supposed to do that to the developing world's desktops, but everyone seems to be happy enough running Windows. Go figure.
Nice, ya capitalist shill.
Yeah, I could never understand the use of interpreted code. It's like running all your applications in a CPU emulator.
What do people have against native code? C and C++ produce fast code usually. The browser you are using now is likely to be written in C++. You wouldn't run a Java browser, it would be slower and more resource hungry.
So why on earth have a JVM on a mobile phone where you have a fraction of the memory and CPU power, plus a battery!
I guess one reason is skills, there are plenty of people around with Java skills. So as usual it's not about providing top quality software to the end user, it's about attracting developers.
XCode is rubbish and Eclipse is brilliant?
I use Eclipse every day, it gets slower and slower with each release. Helios is awful, it takes about 3-5 times as long as the last release to deploy JSPs to my Tomcat.
If you didn't get much freedom on iPhone OS then there would be no apps.
Don't worry, though ... they're *also* shunning the Android thingys as well. The only ones I've seen over here (Mexico) are the Motorola Dext and something called the Motoroi. And that last one is for the minority CDMA network... I'd like to see how it fares with the iPhone, which is an exclusive to Telcel, the carrier with 80% market over here.
I'd like to note that at my workplace everyone has smartphones. One guy has an iPhone. Unlike the Linux/Windows wars, the OS choice is linked to specific devices, which have a price tag. iPhone's price tag is somewhere near the "average income of a Mexican family"; the subsidized with 18-mo contracts are cheaper but few mobile users have contracts; most of them are on PAYG.
Blackberries are the ones that have been doing some inroads lately, as there are lots of them on sale on the $200 price tag, something that can be bought by middle-class people without wincing. Android handsets also seem to have a decent price tag, so those are selling even though there is close to none marketing for Android over here. Even though I've seen few Android handsets, there are more of those than iPhone handsets. And the few iPhones I've seen ... well... a couple of them are the hiPhone knockoff. I doubt Apple will ever have a raving success here in Mexico.
I am not sure what the pricing strategy is for the china and mobile markets for the android phones? This is not mentioned, presumably they are lower but no statement is made about actual costs of a desire in india? Of course in an anti apple rant I didn't expect to see actual comparisons of such things as facts, it doesn't suit the purpose.
Quoting Gartner for things such as ".. sees Android surpassing iOS by 2012 in the mobile operating system market. It's only a matter of time before it also displaces Symbian, the dominant-but-fading mobile OS leader." are completely pish, as good as Gartner may be at somethings, I almost always ignore most of these predictions. Not to say it won't happen, but using Gartner as evidence is a fail, in this case it is somewhat obvious.
Of course Android will outsell the iphone, its the only decent OS other manufacturers have, and only chance of putting out a decent smartphone. So no surprise it will outsell numbers wise. Question is does that make it a victory, and who for?
Google aren't giving it away without a reason (philanthropists, uh uh), your data, location, email, contacts, etc are going to be the price, as well adverts (Googles money maker). However they promise to do no evil.
The other tabular evidence is interesting too, we should be happy there are two important player s in the market, I am not so sure we should be happy if googles OS was the only choice (we have already seen how greedy handset makers are quite happy for older (>6m) phones to be left to die (please don't tell me users can flash their OS rom, etc)).
I doubt it. More liekly to be good advertising, which Apple has done successfully.
Joe Public really doesn't care if there's 800 billion apps for either platform, all they care about is being cool and trendy and blindly absorbing the adverts. Suckers.
Android competition is more diffuse. The coming Android army is likely to be populated with (i) those who made a specific choice about the handset; (ii) those who made a specific choice about tariffs; and (iii) those who don't care and took what the network offered. Even if Apple's hardware, software and marketing platforms were such that everybody in category (i) opted for an iPhone (please don't flame me here, I'm being hypothetical) then Android would still pick up a massive body of users from categories (ii) and (iii) because it has the momentum amongst the operating systems that compete in those categories.
Scenario 1) Sees ad on TV, likes gadgets, buys iProduct.
Requires someone into tech enough to buy pretty expensive toys.
Scenario 2) Coming up to a year on the contract. Phone company calls and asks if you would like a new handset. Android phone with trial data plan arrives in post.
Requires existing cell contract. Not hating it enough to throw it away.
Which one gets their gear into more pockets over a given time?
The only market which is significant in going forward for a vendor is a market where it can SELL ITS GOODS AT PROFIT. While it is possible to sell some hardware (taking the risk that it will be copied) in these markets, selling software for a profit in them is something which is nigh impossible even for someone like Microsoft who can bring in the USA diplomacy as "tactical support".
So as far as the success of phones as a software platform the amount which is sold in these countries matters very little and will continue to matter very little until it will become possible to sell non-local _SOFTWARE_ there at a profit.That is yet to appear on the radar.
Android is the Linux scam of the mobile world. Corporate managers are pushing a badly engineered architecture on the market like it was better than anything before it... the fact is that it is the worst possible option.
SymbianOS is the only real viable platform that could be better than iPhone if only Nokia pushed it properly.
But no.. there is the low quality Android from Google everywhere now... which just sucks.
Google couldn't even define a real and proper hardware specifications standard for manufacturers.. so application developers have to deal with a real nightmare trying to support an increasing number of incompatible devices.. it's the Linux anarchy.. a real mess with no real engineering plan.. that is the opensource scam for the masses...
i hope google sort out their marketplace, and clear out the smut and at least put some controls on it. i'm an android user and a little nervous (security-wise) about installing apps from there. i generally choose poplular ones, but don't get that nice feeling like when i'm downloading apps from ubuntu's software centre.
maybe google will sort it after there is a massive malware issue due to a rogue app.
had an android phone for only a few weeks and haven't paid for any apps yet. although not to say i won't ever. i'd probably pay for an app if it was cheap enough and useful enough (e.g. locale?). wave secure (remote wipe app) looked a bit too expensive for my liking.
The issue of wooing developers is not a new one.
Back in the 90s, OS/2 and Windows were battling for developer mindshare. IBM sold its SDK through its Developer Connection program when you could get it for free from Microsoft. IBM used to sell their command line CSet++ for mucho dinero when Microsoft were selling a visual equivalent for a fraction of the cost. Microsoft made development easy with wizards and visual editors, IBM belatedly threw in some C++ foundation classes but not enough. IBM documentation was dry and often unreadable "red books" while Microsoft implemented intellisense, context sensitive help and filled the bookshelves with accesible and cheap documentation. It's almost like IBM purposefully went out of their way to piss on interest in their platform.
What has this got to do with Apple & Google? Well the costs of developing for the iPhone are WAY more expensive than Google. Apple requires enrollment to the developer program costing $99 PER YEAR, vs nothing on Android (except a small 1 off fee if you want to use Marketplace). Apple requires devs use Macs for development. Google lets devs choose their platform and OS. Apple forces devs to use their app store, their ad service, imposes odious approval conditions and takes a cut of the profits. Google has no such restrictions.
Apple would do well to broaden the appeal to developers by cutting the iPhone development loose from the Mac and lowering the fees. There are money fronts to the war of course, but Apple is losing badly on this one. If they don't start making iPhone development more attractive they will lose out the way IBM did to Microsoft.
While not being able to dispute many of your points, I think you over egg your argument in the following areas:
• Google takes 30% of your profits if you sell through their app store, just like Apple do
• Apple allow you to use any ad platform you want (or indeed none at all)
• the iPhone fee is for marketplace listing and on-device testing only, not for access to tools or support
• your claim that Google "has no such restrictions" would seem to be contrary to the news that Google have recently used their kill switch to remove software on customer handsets
On Android you don't have to use the marketplace if you are able to find a different route to market. Conversely, you may publish to the Android marketplace only if you're from Austria, France, Germany, Italy, Japan, the Netherlands, Spain, the UK or the US.
You may agree with Google's use of the kill switch, but that's still an area where they have unilaterally decided what software is acceptable and they have unilaterally decided to kill an existing piece of software. The correct argument is therefore that Apple are significantly more restrictive, not that Google are unrestrictive.
1) You don't have to use Google Marketplace. Companies like Gameloft sell direct. Even if you DO use Marketplace you could use it to release a demo / trialware version for nothing on marketplace and then take users to a pay site to unlock or download the full version. That's just one way and it's not hard to envisage other models. In other words there are many ways to keep all 100% of the money.
2) Apple don't allow you to use any ad platform. Most advertising platforms rely heavily on analytics to deliver relevant adverts and Apple have banned 3rd party analytics. Without analytics the ads aren't as relevant and clickthroughs fall through the floor. They have deliberately banned / crippled the competion so people are forced to use Apple's instead. Their excuses for doing this "to protect privacy" are horseshit, otherwise they ban apps like Facebook, Twitter etc. who use / sell data in all sorts of ways.
3) You pay to enrol in the iPhone Developer Program which is $99 annually. Since you can't distribute apps in any way except the App store, it is effectively a mandatory charge unless you never want to release anything ever. So what about distributing in-house enterprise apps? Well that's only $299 a year.
4) A non sequitur and not a very good one. The marketplace has terms and conditions and people who upload apps must abide by. Compared to the app store these are very liberal. If you are extra paranoid about marketplace, don't use it in the first place. Distribute the apk from your own web site for example where you are not subject to the T&C of marketplace.
As for the kill switch in general I agree it has the potential for abuse and I would like Google to spell out and commit to the occasions it could be used for. I don't believe for a second that Apple has nothing equivalent and their record for declining to list apps and taking them down after the fact demonstrate they have no moral highground.
Actually, Microsoft not IBM was the original source for the OS/2 2.0 development kit. IBM was working on OS/2 1.3.x and LAN Server
Back in 1989, the prices were $2600.00 for the OS/2 SDK, the Windows 2/286 SDK was only $200 or so [this was just before MS's developer subscriptions]. Most developers preferred the Borland tools to Microsoft compilers anyway.
While the core of the API [API names beginning with 'DOS'] was available, the UI [presentation manager] was unfinished - MS OS/2 2.x used the same UI as their OS/2 1.2.1 and Windows 2.x
For that matter, the Microsoft OS/2 3.0 NT project [later renamed to Windows NT 3.0] still used the same File Manager / Program Manager UI.
You are missing the point.
The difference between Apple + Apple fees and their competition is a couple of dinners in a decent restaurant. The difference between IBM costs and Microsoft from 17+ years back was on the order of half an annual salary. If we inflation index IBM prices for compilers and kits you are looking at 20000K+ for some of them in todays money.
The cost of a couple of dinners or a weekend in a 1 star motel in a third rate resort is something that most people can swallow. In fact it is not that different. The average cost of a documentation book from Microsoft Press (and they were excellent by the way) around 1991 was 30-40$. In todays money this is not far off from what Apple is asking for.
Half an annual salary however is... how to put it... a slightly different proposition.
So the comparison between Apple/Android today and Microsoft/IBM from 20 years back is a bit bogus.
Apps are part of it, but it's also cost. You can get an HTC Legend for £20/month, which is the sort of money that iOS just doesn't have an offering to compete with.
And of course, that was always the thing with Windows vs Mac - you could buy a PC for a lot less than a Mac.
Unless Apple stop raping the consumer price-wise in certain territories, and wise up with regards to the App Store, and produce a non-Apple version of the dev tools, then in a few years the iPhone will be way behind.
.... are the main advantages of Andriod over iOS4.
Saturation is always a battle that once it's won usually helps to dominate the overall competition war. As mentioned by some comments, Andriod on the face of it could be an inferior platform but if it saturates the market with so many different devices and price points against the 1 model that Apple are offering at a fixed price globally than Andriod will surely come out top. The quality of the product vs the superioty of another won't help as the VHS vs Betamax battle proved.
Apples best move at the moment would be a recall, they can afford it and they could do it well enough that they would not only reinforce those customer loyalties that were waning but maybe even pull in some more customers out of respect for the move. I think the atenna problem is a severe chink the armour that is Apple's roadmap to success and market domination.
Personally I prefer Andriod, due to price points and choice.
There's very little to dissuade Joe public from jumping between iOS, Android, Blackberry, Palm or WinPo7, All their MP3's, MP4's, videos, email, contacts & calendars will just slide across, and at 99c a pop I doubt people are going to lose much sleep over sunk software costs.
The best situation for us (consumers), is for lots of competition, not a dominant player that sits on their arse and rots, been there, done that already.
I expected this would be fucking obvious to an IT readership.