Fortune magazine says 2011 will be the year Android explodes and makes a number of predictions that track what I’ve been writing on Armed & Dangerous for the last year so closely that I have to wonder if their correspondents have been reading this blog.
Nice work, guys, even if you are a bit late to the scene. Me, I kinda thought 2010 was the year Android exploded, what with posting several quarters of market share growth in the 600% range and overtaking Apple in the U.S. userbase figures. But the article isn’t just old news presented as though it were breaking; there’s some fascinating info about a new chip from Broadcom that’s going to blow a Jupiter-sized hole in the price floor of the smartphone market.
Broadcom thinks its new BCM2157 smartphone-on-a-chip will push the retail price of a smartphone – not the handset bill of materials but the retail price – below $100, and might push it below $75. And they’re targeting it directly at Android OEMs. Phones based on the chip are expected to ship in volume before mid-year 2011, with the first possibly hitting the trade show circuit in a month.
The article doesn’t note that something like this was pretty much inevitable about now; there was just too damn much money waiting to be made from single-chip smartphones as the dumb-phone base turns over, especially in overseas markets at lower pricepoints. The only surprise, and it’s a minor one, is that Broadcom got there before Qualcomm. Expect Qualcomm to follow suit, and expect at least one more competitive phone-on-a-chip from one of the Pacific Rim manufacturers – my first bet would be on Samsung, but I wouldn’t be surprised if we get one from mainland China.
The prompt effect will be to open a lot of additional daylight between Apple and Android pricing. Of course Android phones have been less expensive all along, and that’s been fueling the explosive Android market share growth we’ve seen since late 2009, but the price gap is about to widen into a vast and yawning chasm. Like, with a width in three figures.
I think Apple’s hopes of retaining market share above 10% will disappear into that chasm, especially since app developers are quite capable of seeing where the price and volume winds are blowing. Given Apple’s history and Steve Job’s psychology, I do not expect Apple to try to compete seriously for volume in a market where midrange has fallen below $100; it is far, far more likely that the company will take a defiantly top-of-market position and hope they get enough lock-in from iTunes to defend that turf. Expect Apple’s brand positioning to shift further towards that of a luxury and positional good.
But that’s not going to be the biggest effect. Oh, no. The real impact of the BCM2157 and its kind will be on the cell carriers. I’ve been writing for two years that Google’s long-term Android plan is to break the carrier oligopoly over its knee, pry their customers loose from the contract system, and reduce the carriers to low-margin bit-haulers in cutthroat price competition with each other. And Jean-Louis Gassée has pointed out that around $89 retail is where shit gets real; at that point, the carriers stop being able to offer enough of a price break so customers will enter contract bondage to get it. This, in turn, is going to kill off carrier locking on phones.
That’s no longer a distant future prospect. It’s likely we’ll reach $89 before the end of 2011, and carrier locking probably won’t outlive that break point by more than a year. Fortune, to its credit, gets some of the implications (maybe this means the rest of the business press will be belatedly catching up with where I was in 2008 over the next quarter?). They note that increased customer mobility is going to exert downward pressure on data plan pricing. Yeah, will it ever.
Increased customer mobility between carriers is also going to pump up competition in the already frenetic handset market as customers get used to moving their phones and their Android customizations more or less frictionlessly between carriers. That means $75 won’t be where it ends; I expect a race to the bottom in the next three years, with cheap smartphones becoming nearly as disposable by 2014 as calculators are now. And the positive network externalities pulling people towards Android (most importantly, the size of the app ecosphere) are only going to increase as its market share does.
Big losers in this scenario start with Apple and the carrier oligopoly, for reasons I’ve already covered in detail. RIM? Stick a fork in them, they’re done – headed for a textbook disruptive collapse within that three-year timeframe. The minority Linux-based smartphone OSes (WebOS, MeeGo) are noble but doomed, no-hopers. Microsoft WP7 isn’t going to survive the smartphone price crash, either – their problem is that they need to make money from licensing fees, and that’s hard to do when your hardware platform is priced in the supermarket-giveaway range.
The largest remaining open question is whether Microsoft will drag Nokia down with it. The most optimistic scenario for Nokia is that they evade the lure and bail to Android; in that case they end up as yet another handset maker but at least they have a line of business playing the pump-out-handsets-cheap game they’re historically good at. The most pessimistic one is that Microsoft sells them on WP7 and they watch their market share drop like a rock as Android steamrollers right over it. It would be a billion-dollar fiasco, a classic don’t-do-that for the business-school casebooks.
If all this sounds crazy optimistic, don’t forget that people who should have known better said I was crazy optimistic two years ago when I analyzed Google’s grand strategy, and again almost a year ago when I predicted Android’s U.S. market share would break 50% right about when it happened. The trade press is catching up with where I was, but where I was ain’t where I is; I’m looking to the future, and I see that chipsets like the BCM2157 and their followons are going to be bigger game-changers than Fortune magazine yet understands.
Eric
You also predicted mobile phones would fleece mobile computing, taking over note/netbooks.
This looks the next step in this direction.
Why would the BCM2157 get the retail price below 100$? I had read that it costs less than 200$ to produce a modern smartphone:
http://public.tableausoftware.com/views/PhoneComps/WhatDoesitTaketoBuildaSmartphone
Even if you put all the discrete components on to one chip that normally cost you 100$, you wouldn’t sell it for 5$, you’d sell it for 50$.
So now you have a phone that costs 150$ to make instead of 200$. That is a big deal, but it won’t lead to some crazy market collapsing price war unless people can buy the phone for cheap from that manufacture directly. I’d love to see unlocked smartphones for 300$.
Subsidized, you can get almost any smartphone for a penny from Amazon.
I have to qualify my previous comment. I heard eric discuss the future of android phones on a pannel discussion in The Command Line podcast. I interpreted his futuristic ideas as an encroaching into the mobile computing sphere of smartphones.
I am not sure eric agrees with this extrapolation of his words.
Other interesting article:
The $85 Smartphone and the imminent extinction of non-smartphones
http://www.asymco.com/2010/12/27/the-85-smartphone/
I’m very curious to see if this pans out as you’ve predicted. I kind of hope that it does (a smartphone below 100$ means I might actually buy one), but I maintain a certain degree of skepticism. My skepticism is fueled not by a feeling that I know more about smartphones that you do (I don’t), but because when I read your entire archives a few months ago I noticed a some consistent biases in your predictions: (a) you usually predict things that you want to happen, and rarely predict things that you don’t want to happen, and (b) you exhibit a typical geek’s exuberance about technical advances that are always right around the corner. So when one or both of these biases seem to be in play, I adjust my credulity about what you predict appropriately.
OTOH, your record on smartphones in particular has been pretty good.
I agree wholeheartedly, but what I don’t see is Android app development really taking off. Apple has adopted an end-to-end strategy for iPhone developers that makes it easy to develop, deploy, and get paid for iPhone apps. They have really streamlined the process of developing, selling, and buying apps and that has made them, so far unshakably, the platform mobile developers target first, even as Android overtakes them in raw numbers. El cheapo smartphones hitting the market is likely to exacerbate the #1 complaint about Android from mobile devs — platform fragmentation.
Here’s the problem for Apple: As more people carry around smartphones and the majority of these become Android phones with music players preinstalled — and the use case for uploading music to these portable devices becomes “mount the smartphone and copy a directory full of MP3s to it” — whither the iPod?
@ErikZ:
Broadcom is the king of the el-cheapo system-on-a-chip. This is exactly what Broadcom did for WiFi networking equipment; virtually every wireless router you’re likely to buy is based on a Broadcom SoC; a few are based on the slightly more expensive Atheros chipset, But, as you say, low component prices alone won’t drive down the price the end-user pays; however there are two forces that will drive phones cheaper and cheaper:
One is that Android has pushed the software development cost near $0; the other is that intense competition among smartphone OEMs and wireless service providers will have a naturally tendency to drive prices lower. That’s what free markets do.
I don’t see that as a problem for the iPod, Jeff. That was the use case for MP3 players before the iPod, after all. (I’ve got a Creative Novad MuVo buried on my desk here; that’s was how you used it.) iTunes revolutionized portable device music management, and then revolutionized buying content for it. For Android to displace the iPod, someone will have to replace both iTunes and the iTunes Store, while making it possible for people to play the existing iTunes Store content they purchased – and make it work no matter what Android phone is being connected. This is a tall order, comparable in scope to standardizing the Android ecosystem’s equivalent of the iPod Dock connector.
I do agree that platform fragmentation will only get worse unless someone clamps down hard – and I’m not sure that’s even possible. Otherwise, consumers will be left with “this app works on phones X, Y, and Z, but only if your X is from this carrier and if your Y is not from this carrier and hasn’t been upgraded to the latest revision” and yada yada yada. DLL hell on Windows pales in comparison. Even CPAN won’t hold a candle to this.
I don’t think this will hold back Android adoption, but writing apps for Android in its current state would seem to be a losing proposition unless one is careful to stick to the lowest common denominator of Android capabilities. People who want an app-rich experience will find the iPhone much more congenial.
>Otherwise, consumers will be left with “this app works on phones X, Y, and Z, but only if your X is from this carrier and if your Y is not from this carrier and hasn’t been upgraded to the latest revision”
You know, people keep raising this as a theoretical possibility, but I have yet to hear about a case where it has actually happened. Can anyone supply one?
I tried several Google searches for variations on like Android app non-portable and Android fragmentation problem and 99% of what I turned up was the trade press going rooba-rooba-rooba without actually pointing at an instance.
In fact the only instance I could find was some thinly-sourced muttering about performance problems in a game called Angry Birds. Even in that case nobody was alleging that it wouldn’t run or even that it had platform-specific bugs.
If there were a real problem here, I’d think horror stories by the bushel would be far more easily found.
>Even if you put all the discrete components on to one chip that normally cost you 100$, you wouldn’t sell it for 5$, you’d sell it for 50$.
You’d like to, sure, but in a market with N other low-cost producers scrapping for share like frenzied wolverines you’re not going to be able to.
And that component price chart tells us that on a representative high-end Android phone (the Nexus One) the processor is just about 50% of the BOM. A good rule of thumb for consumer electronics is that your retail price floor is about 3 times the BOM; a little back-of-the-envelope figuring tells me that in order to pull the retail price down to $100 Broadcomm only has to get its volume price to $16 per unit, $12 if they’re shooting for $75 retail on the phones.
Are those prices achievable for a single chip produced at volume? Hell yeah. In fact once they’ve amortized the fab setup costs the price could easily drop by a factor of four from there. Broadcom’s prediction is conservative; they’re leaving themselves fighting room for the inevitable price wars when more all-in-one chips materialize.
So the next thing you’re going to ask is why the rest of the components should follow the processor down the price curve, and the answer is: because they can. There’s nothing on that parts list that doesn’t have a price curve asymptotic to zero as you push volume high enough.
@Jeff Read
Android “fragmentation” is a myth perpetrated by iPhone fanbois as they dig themselves deeper into denial over Android’s imminent market domination.
>Android “fragmentation” is a myth perpetrated by iPhone fanbois as they dig themselves deeper into denial over Android’s imminent market domination.
Incorrect, sir! Android’s market domination is not “imminent”. It’s already here.
Jay Maynard,
You’re right. End to end. I forgot. That’s exactly why people will continue to pay top dollar for the Apple stuff: the UX is absolutely seamless and users prefer that.
Ignatius T. Foobar,
Do you know from whom I hear the most talk about fragmentation?
People developing Android apps, who have to test them on every bleeding Android phone on the market.
Fragmentation is a real problem and it’s absolutely killing Android as a serious app platform.
Exactly. Just look at what MetroPCS is selling: the $99 Android 2.1-powered Huwai Ascend. 3.5″ screen, 3.2 MP camera. Not bad for the budget crowd, and you can bet when products based on these smartphone-on-a-chip boards start shipping, the big boys will have offerings at or below the $99 price point, whilst Huwai or other low-end Taiwanese producers will be selling Anrdoid phones for as low as $50.
Actually, that’s a good rule of thumb for many classes of consumer goods, not just consumer electronics, but it applies pretty uniformly across the consumer electronics spectrum with the exception of a few high-margin players such as Apple.
Eric,
I’m not following your reasoning behind this timetable for the collapse of RIM. Yes, I see this as marking the beginning-of-end for “feature phones” and dumb phones, but I’m not 100% following your logic with the collapse of RIM.
The Broadcom chip is hardware, and as “a rising tide lifts all boats”, reduced hardware costs affect all smartphones, not a specific operating environment like Android. While Android’s $0 software cost is a factor (lower price floor), isn’t that already reflected in the market with the exploding activation rate? Admittedly, all of my computing experience is after the minicomputer/home PC revolution; if there is some historical paradigm-shifting comparison you are drawing on for support, I am not seeing it. [Porting is the only thing that came to mind, but cannot be an issue in this case; this chip is based on ARM11, as are recent RIM releases.]
I am fairly sure (call it 55% in support) that these shifts along Moore’s power curve sound the death knell for RIM …eventually… but I am curious why you chose this timetable. From what I read in this article, we can foresee only the “feature phone” market collapse, not a consolidation in the smartphone arena.
>I’m not following your reasoning behind this timetable for the collapse of RIM.
RIM has three bullets flying at it.
One: Android phones will severely undercut RIM on price. There are several different reasons RIM can’t compete on cost; one that will stand for all is that it incurs a lot of NRE maintaining its own operating system – and those costs will be going up as they throw out their old codebase and transition to QNX.
Two: RIM’s defenses are fragile. What’s holding their customers is Microsoft Outlook integration and a reputation for being manageable by corporate IT departments. But they’ve been coasting on that too long; other than the admittedly nice physical keyboards their UI sucks really badly by 2010 standards. (I know this because I’ve watched my wife use a Blackberry her law firm gave her.) Decent Outlook integration on an Android phone? There’ll be an app for that; the opportunity is too big to ignore.
Three: RIM doesn’t have a credible strategic story for investors. Their market share has been eroding too fast for them to claim they’re successfully forted up in a base of corporate customers. Even if they could make that claim, it would be a classic retreat upwards, and nobody in the tech indutry really buys that anymore (look at what happened to Sun when they tried it). The question they’re going to be asked is “Where do you think your share gains are going to come from?”, and they just don’t have any good answer.
The acute danger here is that corporate customers will decide they don’t need RIM’s gold-plating, a classic disruptive collapse a la Christensen; if that happens, RIM’s market share could crash to statistical noise level within a quarter. I think that’s a strong possibility to happen within the next 18 months, and a near certainty within three years. But even if it doesn’t, where’s their growth going to come from?
If by ‘users’ you mean the trendy metrosexuals who hang out at Starbucks all day pounding on their MacBook Pros and iPhone 4s, with their brightly-colored Mazada 6s out in the parking the lot with Apple stickers placed neatly on the back, then, yeah.
The T-Mobile Comet (Huawei U8150 IDEOS) is an example of what an $100 Android smartphone might look like in 6 months. It is $175 from T-Mobile and $146 from Amazon, “prepaid” but not unlocked, and free on contract. T-Mobile is ahead of the curve in the US, and if I could get voice coverage from them where I live this phone would be “good enough”. It gets 4+ stars from 10 reviewers on Amazon, which indicates that buyers have realistic expectations that are largely met by this phone.
Eric will confirm that I’m not an Apple “fanboi”. They do screw up, and I call them on it. OTOH, their stuff, whatever else you can say about it, just works.
The posting above about the $99 Android phone is an example of what I’m talking about. It’s Android 2.1. What would it take to upgrade to the current, or even one-back, Android version? What apps will and won’t run on it? (Certainly those that depend on Android 2.2 won’t.) What other assumptions are there in common Android code that will break when run on that box?
And, critically, how much will people spend on apps that won’t run on it before they give up on buying apps?
I’ve got no doubt that Android will take over the smartphone marketplace, for all of the reasons Eric cites. I’m just not entirely sure that an Android smartphone monoculture will be any better for consumers than the Wintel monoculture is on the desktop. I’m certainly not going to toss my iPhone any time soon.
“Broadcom is the king of the el-cheapo system-on-a-chip. ”
Which is also not a good thing. Their cheaper product offerings are not known for their performance or quality.
Additionally, a HSDPA-only chipset (that the BCM2157 is) is not going to be the endgame, at least in the US. Qualcomm really does have get their Snapdragon chipset pricing down, or you won’t see cheap smartphones on CDMA networks, which is a critical piece of the game, IMHO.
Nothing predicts success like calling your competitors’ users “trendy metrosexuals”! Thanks to attitudes like this, my Ubuntu netbook has all of the UI polish of Windows XP, and I can’t convince my wife to give up her Macbook for anything.
There’s an Android boxing game in the style of Punch-Out!! called Super KO Boxing. I saw a friend playing it on his Droid and it looked pretty fun.
I found out, only after I had installed the app and only after it had downloaded its 50-MiB content payload to the SD card, that my poor Hero’s graphics chip is too weak to handle whatever OpenGL ES stuff they’re doing in that game. So the game, technically speaking, works — but all the in-game graphics become black rectangles. Meaning I’m staring at an empty screen.
This, simply speaking, Does Not Happen on iPhones.
As for Angry Birds — it works on my Hero, but it has severe input lag and framerate jitter even when I give it a nice beefy Nook Color to run on. Again, this is probably not an issue at all on later-model iPhones and the iPad (Angry Birds is a ported iOS app).
Those are people who buy Apple stuff as a status symbol. There are perhaps a greater number of users who buy Apple stuff because It Just Works and it enables them to do something they couldn’t do before.
Those millions of iPod-buyers and iTunes-users I and Jay were talking about are certainly not all “trendy metrosexuals”. The iPod (and iTunes Music Store) has crowded out most of the other players in the digital music space on the strength of ease of use alone.
What we’re going to see from this is pretty clearly the end of the non-Smartphone. It’ll probably also kill off the cheap Symbian phones that have been Nokia’s market saviour and puts RIM in a bad place due to lead times on their reboot of the Blackberry OS. WebOS is pretty much dead in the phone market already.
As to Apple, I don’t think it will have a significant effect on them except to possibly make it more clear that they own the Media side of the market since these new phones will almost assuredly be bad at media playback (especially video). It will change the numbers, but I doubt it will significantly affect their share of the overall cell market (what it will do is change the smartphone market to BE the overall cell market and make it clear that Android is the dominant player in the overall market). SoC’s are a great cost-cutter but don’t expect great performance from one, especially a Broadcom one. These phones will be the cheap and mostly crappy phones that fall in below HTC’s low-end stuff so don’t expect great things from them, expect crappy ubiquitous things. We’re talking the smartphone equivalent of those old eMachines Win98 boxes that sold so well back in the late 90’s/early 2000’s.
>As to Apple, I don’t think it will have a significant effect on them
When you say things like that, you seem to be ignoring the fact that Apple’s share is already falling, and has been steadily dropping ever since Google launched phase two
of lumpy gravyin 2009. It’s not good enough for Apple to tread water; they have to have a strategic answer that gains back share, or at least firewalls them against losses past the point where the app store and iTunes lose their attractiveness to suppliers.I’ve already said that I think media delivery will do that for Apple if anything can. The problem is that this depends on Google and the Android army not attacking that problem successfully. Not a safe bet; Google’s got an awful lot of smarts and financial mass to throw at the problem. If Google decided to set up an iTunes competitor, that’d be a credible threat from day one.
@esr: Yes, RIM is vulnerable on Exchange(not Outlook) integration, but the real enemy there is Dell with Windows Phone 7 , not Android. While there will no doubt be a viable Android solution for Exchange integration, Dell’s ability to package a solution and provide it directly to the IT markets as a value-add to their current Server business is the real threat. Dell can now be a one stop shop for a full solution rather than IT shops having one supplier for their backend and then RIM for the handsets and integration. Android can’t offer that until there’s a single supplier offering all of the bits (or at least both the handsets and the integration solution a la RIM). Don’t discount Corporate IT’s semi-ridiculous level of love for integrated single-source solutions, it’s what made Exchange/Outlook so popular in the first place.
Adam, if anything can save Windows Phone 7, it’s a scenario such as you describe…because nobody else wants it. And the only thing preventing Dell from doing it on Android rather than WP7 is that Dell felches Microsoft even more than Compaq used to.
@Jeff Read
“Device X does not have enough horsepower to run Application Y” is *not* the same thing as “the platform is fragmented” … and you know it.
When you can demonstrate an application designed for one device that will not run on an equally powerful device running the same version of Android, but from a different vendor … *that* is fragmentation.
I’ve never seen that happen. Android so far has been quite the “write once, run anywhere” platform. The inevitable fact that some folks are still using older hardware does not negate that. I don’t expect Angry Birds to run on my G1 any more than I’d expect Photoshop to run on a 386.
>I’ve never seen that happen. Android so far has been quite the “write once, run anywhere” platform.
I’ve never seen it happen either, and my Web searches are stubbornly failing to turn up examples. I won’t go so far as to say deliberate FUD has been spread about this, but I think it is one of those cases where the trade press has ginned up a controversy because conflict stories are juicy and drive page views. The more I look into this the more doubtful it seems there’s any actual there there.
@esr:
Thank you – that’s a lot clearer! I’m still not sure I agree with you entirely on aspect #1 being a future threat – I still expect the activation numbers to already reflect price discrimination. Likewise, #2 isn’t changed much by this announcement; that is a software issue, and thus we should wait for each Android revision to make new predictions on this element. (Of course, you didn’t post about the recent 2.3 release separately, so I’m assuming you rolled that into this new prediction.)
It’s #3 that I think you’ve really hit the money on. Many of your detractors don’t seem to fully understand the power of a narrative, and it will take one narrative-killing question (like yours) on a quarterly-earnings call for the investors to flee RIM “like rats from a sinking ship”. I also agree with your implicit closing statement-as-question that RIM doesn’t have a new growth field available: new “business professionals” (the CxOs, doctor, and lawyer crowd) don’t have a compelling case for choosing RIM devices.
new “business professionals” (the CxOs, doctor, and lawyer crowd) don’t have a compelling case for choosing RIM devices.
Sure they do. RIM devices can be remotely destroyed and all information on them erased beyond recovery if the device is lost or stolen, or an employee is terminated. That’s the killer app in Exchange Integration space. Until you can do that with another device, then RIM has the corporate market sewn up. And their hold will only get tighter as more and more data security regulations impose upon more and more sectors.
In other words, you won’t be carrying your e-mail with you on your Android phone in any company that has to deal with HIPAA or SOX if they have the appropriate level of paranoia in IT.
JCB Says:
> … or you won’t see cheap smartphones on CDMA networks, which is a critical piece of the game, IMHO.
I’m glad someone mentioned this. Until these phones appear on Verizon, there are huge swaths of the US that won’t be participating. Maybe the story will still play out as indicated, but there is a big market out here where the fast-as-a-charging-glacier AT&T network is useless and the big-city-only carriers like T-Mobile can’t be bothered to soil their wingtips. Can I get any sympathy for those of us in flyover country?
@Jeff Read:
I’ll give you three reasons while millions of iPod/iTunes users won’t be replacing their portable media device with a smartphone from any: 1) battery life, 2) battery life, and 3) battery life. Sure, you’ll get a few light users to switch, but the people that listen to music all day? No way. It’s too much hassle to constantly have to charge your phone.
>1) battery life, 2) battery life, and 3) battery life.
System-on-a-chip smartphones may help here. One of their selling points is reduced power consumption.
@brian:
Correction: That’s a killer feature for the IT department, but it’s a platform killer from the end user perspective. As a private individual I’m perfectly content with a password on my phone and being able to turn off service to the device with the carrier; in fact, with the exception of the two fields you mention (medicine with HIPAA and accounting with SOX) there is no need — aside from rampant paranoia — to implement a wipe & destroy policy for all devices connected to their network, otherwise all workstations would do that once disconnected from the correct Windows domain.
>That’s a killer feature for the IT department, but it’s a platform killer from the end user perspective.
You know, I hadn’t got this far in my thinking before, but you’ve just pointed out a fourth reason that RIM is in strategic deep shit. It’s that the balance of power between mobile end users and their IT guys has shifted in the same way we saw with the advent of the PC in the 1980s. The shift is not to RIM”s advantage.
When Blackberry launched, wireless digital was expensive and scarce; WiFi barely existed. There was a strong case for IT departments tightly managing mobile access, for the same reasons they used to centralize control of non-mobile computing more than they do. Expensive capital goods justify intensive control. RIM tied itself to that model, and BOFH-can-wipe-your-device is a sign of it.
That’s not the world we live in any more. Just as with PCs, the effect of plunging prices for network-edge equipment has been to make end user preferences more important and the end-user’s investment in his own data at the edge significant. BOFH-can-wipe-your-device has slid from a feature to a DO NOT WANT! And RIM is caught on the wrong side of that.
Agreed. But they are known for their ubiquity. Find me a cheap notebook with a gigabit NIC that isn’t made by Broadcom. How about a low-price point wireless broadband router? Even Apple uses Broadcom’s 802.11n chipset in their MacBooks.
Furthermore, I suspect a competing Qualcomm product is not far behind.
>Furthermore, I suspect a competing Qualcomm product is not far behind.
Bet money on that. Qualcomm had market leadership here; they cannot let Broadcom’s salvo go unanswered even if the answer means taking a margin hit.
@ esr
On media delivery and Android – do you think DVDJon’s Doubletwist [www.doubletwist.com] stands any chance in this space?
I used an early version to strip the iTunes drm crap from some oldTunes music purchases so they would play on other-than-Apple hardware. It worked simply and well.
I haven’t used the current Doubletwist implementation (I don’t have an Android phone at the moment) but it is supposed to both essentially replace an existing iTunes install and sync to Android devices pretty much transparently to the end-user.
>On media delivery and Android – do you think DVDJon’s Doubletwist [www.doubletwist.com] stands any chance in this space?
Dunno yet. Website looks like it’s got some money behind it, and underestimating DVDJon is for fools; on the other hand, I’m nowhere near his target market and don’t have a good feel for what’s going on there.
I will say this: if Doubletwist does what it claims and Apple can’t shut it down with lawfare, it will come near breaking Apple’s present business model. Apple needs the customer control it exerts via iTunes to defend the margins on its hardware.
Micheal,
Don’t worry, Sprint and Verison will also be getting inexpensive Android phones. If you read the linked article you will see it refers to a chipset from a different vendor which does the same thing for the Sprint and Verison networks.
Since I don’t mind a contract, I’m looking forward to ever more capable free upgrades to Android phones from my provider, Sprint, as well as free add-on Android phones for my children as they grow. And since we don’t use iPods, free Android phones will do just fine for our media desires. I know, we are stuck in an inferior user exerience, which I will happily accept is worth Jay Maynard’s non-metrosexual money for Jay. So far, however, iTunes and the iPod have not matched my family’ completely subjective price performance requirements. If Apple comes up with an iPod which includes insurance against theft and loss in the purchase price I’d be interested though. For some reason the most eager media device users in our family can’t seem to keep track of them, unlike our phones, which make noise when we dial them. :)
Yours,
Tom
The main example of Android fragmentation with which everyone is aware is Angry Birds. Read these two blog posts and skim the comments:
http://www.rovio.com/index.php?mact=Blogs,cntnt01,showentry,0&cntnt01entryid=47
http://www.rovio.com/index.php?mact=Blogs,cntnt01,showentry,0&cntnt01entryid=46
Quote:
“for the time being, the Android devices listed below are not officially supported by Angry Birds:
* Droid Eris
* HTC Dream
* HTC Hero
* HTC Magic/Sapphire/Mytouch 3G
* HTC Tattoo
* HTC Wildfire
* Huawei Ideos/U8150
* LG Ally/Aloha/VS740
* LG GW620/Eve
* Motorola Backflip/MB300
* Motorola Cliq/Dext
* Samsung Acclaim
* Samsung Moment/M900
* Samsung Spica/i5700
* Samsung Transform
* Sony Ericsson Xperia X10 mini
* T-Mobile G1
Furthermore, devices running Android platform version below 1.6 or custom ROMs are not officially supported.”
Users don’t care *why* Hot New Game doesn’t run on their phone, just that it doesn’t. Whether it’s CPU power, screen resolution, OS or what-have-you, this is a problem.
Don’t we have that now whenever the “equally powerful device” happens to be running an older OS?
…by which I mean, the existence of many versions of Android and many levels of processor capability simultaneously competing *is* what people mean by fragmentation. To insist on making everything equal first is to define away the core problem, which is that everything *isn’t* equal.
>>Otherwise, consumers will be left with “this app works on phones X, Y, and Z, but only if your X is from this carrier and if your Y is not from this carrier and hasn’t been upgraded to the latest revision”
>You know, people keep raising this as a theoretical possibility, but I have yet to hear about a case where it has actually happened. Can anyone supply one?
Netflix: http://blog.netflix.com/2010/11/netflix-on-android.html
>Netflix: http://blog.netflix.com/2010/11/netflix-on-android.html
This is not a fragmentation problem. Rather, it’s all about the DRM and the platform being so open that the DRM is critically exposed to attack. Netflix knows that the second it puts a client on an Android phone that can be rooted the reverse-engineering’s going to start. Hell, people will build custom ROMs just so they can watch the dataflow, and before you know it Netflix’s protocols and crypto keys would be busted. I don’t blame Netflix for not wanting to go there, but…what are they going to do when the market share of rootable phones gets too big to ignore?
esr,
Since Android very allows diverse hardware by design, won’t we necessarily have some quite noticeable software fragmentation, particularly with regard to games, some of which are almost always on the bleeding edge?
I have also experienced software fragmentation on my Samsung Epic. The first app I downloaded to use my LED camera flash as a flashlight did not work, perhaps due to hardware differences. This is not the first app I wanted to load where the comments or even the application provider claimed it did not work on some phones. Much as I like my Android, for me, software fragmentation is real. Maybe I just load more silly apps on my phone than you do. I am somewhat a sucker for sounds cool, is free combined with on my phone.
Yours,
Tom
>Since Android very allows diverse hardware by design, won’t we necessarily have some quite noticeable software fragmentation, particularly with regard to games, some of which are almost always on the bleeding edge?
You’d think so, but hard evidence of same seems remarkably difficult to find. Especially after you exclude Netflix dragging “fragmentation” out as a red herring to disguise their fear of having their DRM cracked, and the Angry Birds complaints that are pretty clearly down to underpowered hardware rather than API breakage.
Actually, your report of your flashlight app failing is the first real fragmentation problem I’ve been able to identify.
P.S. And I USE the flashlight app on my phone. My 1992 Honda Civic does not have a light in the trunk.
RE: Angry Birds and the Fragmentation problem, though their list has grown smaller, that list of unsupported phones was originally much bigger, and included newer phones like the G2. So it’s not imaginary, though certainly isn’t as big as it could be, and is not something Apple themselves can avoid for long.
@IGnatius:
>When you can demonstrate an application designed for one device that will not run on an equally powerful
>device running the same version of Android, but from a different vendor … *that* is fragmentation.
When you can get equally powerful devices, or even similarly powered devices, running different versions of Android, sometimes even from the same vendor, and all being sold as “new”, that is fragmentation. Apple has the benefit of controlling what device every vendor sells and what version of the OS goes with it and also every Apple customer has the ability to upgrade their own phone to the latest version of the OS (for that model), Google doesn’t have this control, nor do their users and this is a problem. Ultimately, I think that you’re going to see a form of software/minimum hardware tie to help keep this under control (i.e. an iOS 4 phone has at least these features, an Android 3 phone has at least these features, and you can’t sell your phone as an Android 3 compatible phone unless it does). Then when you go to either the iPhone or Android app store, you see a store with only the applications that will run on your device.
It’s either that, some form of developer responsibility for categorizing their apps by phones that they’ve tested or we’re back to the days of “This software requires an Android compatible phone with at least 60MB free storage, an 800 Mhz Processor, 256 MB of RAM and an NVIDIA XYZ graphics controller” being posted along with every software piece again. I think it would be ideal if we could get past that in the phone world. Ideally, mobile computing would be more akin to video game platforms, where games are “compatible” with a device, not with your software.
@Jay: Dell only went for WP7 because it essentially cost them nothing and at the same time they got to standardize internally. There’s serious indications that the Dell WP7 phone was driven by internal Dell IT politics more than anything else (Dell doesn’t like having to go outside for their IT stuff) and selling it was simply a bonus. But yes, this is one of the only two strengths it has (the other being the only other real option to Apple for an integrated mediaphone and having used both, it’s only strength there is that Netflix and Zune Marketplace).
@esr: Is Apple’s share falling or is the smartphone market simply growing faster than Apple’s share of the overall cell market is growing? If the latter, Apple’s not in any trouble. I still think you’re mistaking the smartphone market eating the cellphone market for Android eating Apple’s share of the Smartphone market. Yes Android is growing extremely fast but I’ve seen little indication that Apple isn’t growing overall share in the Cell market even as they lose share in the Smartphone market.
>Is Apple’s share falling or is the smartphone market simply growing faster than Apple’s share of the overall cell market is growing?
Both, I think. I know for certain from my own experience that Apple has lost iPhone users to Android.
@ esr
> I will say this: if Doubletwist does what it claims and Apple can’t shut it down with lawfare, it will come near breaking Apple’s present business model. Apple needs the customer control it exerts via iTunes to defend the margins on its hardware.
I agree – and have been watching Doubletwist with some interest for that reason. It’s easy to imagine Cupertino readying for a fight, though …
@esr: Agreed on Doubletwist, if they pull it off all bets are off in the media marketplace. Both Apple and MS NEED the tie-in with their media stores to make their business models work. Apple’s actually in less trouble here as they retain a core market which won’t even look at Doubletwist but even then it would in fact break their backend relationships with the labels and studios who are bloody paranoiacs on the subject.
“If Apple comes up with an iPod which includes insurance against theft and loss in the purchase price I’d be interested though. ”
Are you also dismayed that the manufacturer of your automobile also does not offer insurance directly?
If I were an Apple shareholder, I’d seriously wonder what the hell they were smoking were they to offer such a thing. Especially since the things seem to be easily lost.
“Bet money on that. Qualcomm had market leadership here; they cannot let Broadcom’s salvo go unanswered even if the answer means taking a margin hit.”
Seeing as Qualcomm still has some patent lockup in the CDMA realm, they will probably maintain market leadership in that space. They’re probably going to lose in the HSDPA space without some major margin adjustments. Or possibly going for the “power user” boutique end of the Android market. I’m not sure, I haven’t done a lot of homework on this in a while. :)
Doubletwist rocks.
It was the last app I had to find for the Droid X that allowed me to leave my iPhone 3G happily in my past, along with the Palm Treo and the Blackberry Pearl.
Doubletwist looks and acts so much like iTunes that I’m shocked Apple hasn’t sued them yet. (Maybe they have.)
In light of this, you might want to tell me why Intel is building remote-kill and remote-Imperius-Curse into PC chipsets at the chipset and BIOS level — and PC manufacturers are on board. I suspect that with the ease with which a malicious employee can simply walk out with sensitive corporate secrets (or even classified information!) on, say, a smartphone or laptop and turn them over to Julian Assange on a perfectly innocuous-looking lunch break, corporate and government IT departments still consider remote-kill a critical feature if they want to mitigate the risk.
>In light of this, you might want to tell me why Intel is building remote-kill and remote-Imperius-Curse into PC chipsets at the chipset and BIOS level
You can bet this isn’t fitted to the white boxes you buy off the net…which is the relevant analogy for this discussion.
@esr: I’ve seen plenty of Android users go iPhone as well (and ditto webOS users). I’ve seen a few iPhone users go the opposite way but based on annecdotal observances what I’m seeing is new Smartphone users trying Android then moving up to iPhones.
Personally I picked iOS due to the content sources, I’m already an iTMS user and a Netflix user and my feature needs are pretty basic (email mostly) so iOS came out well ahead of Android for me personally. I was a webOS user, so I’m not a good general example though, I’d been using an old 60GB 1st gen Video iPod alongside my Pre and am replacing both with an older iPhone 3G.
@esr,
If breaking the netflix DRM would require rooting your android phone, how is the scenario any different on iOS?
>If breaking the netflix DRM would require rooting your android phone, how is the scenario any different on iOS?
Ultimately, it isn’t. They’re both Turing machines.
The practical difference is Steve Jobs’s reality distortion field. He has sold the studios on the belief that rooting of iPhones is a sufficiently isolated and unusual phenomenon that DRM cracks are only going to be a problem out past the net-present-value discounting horizon. The studios needed to believe there was some digital-delivery play that didn’t leave them screwed, blued, and tattooed, so they went for it.
Now go see Doubletwist…
Maybe Apple realized that all the world isn’t a PC.
Maybe they want to create an app ecosystem where users don’t have to contend with the case that an app downloads and installs but exhibits weird and jarring failure modes because some hardware capability it was expecting isn’t there.
Maybe they think end users would be better served if the App Store didn’t let them see anything that didn’t work with their device.
And maybe having a single product line helps them manage — and developers query — device capabilities in a way that Android, with its tensor of capability settings cannot.
Just some thoughts…
@Jeff K.:
This is an incredibly nearsighted method of security. First: it only addresses some remote media devices, not all. Private Bradley Manning claims to have carried away secrets on CD-R: let’s see you cast your “Avada Kedavra” on that [remote kill, for those not Harry Potter fans]. Second: you open yourself to attack from malicious elements within your network. How quickly could most business recover from a rogue administrator using this wipe-and-kill command to destroy their network? [Consider separately the cases of user-devices only, and the scorched earth scenario, where the remote kill command is enabled for servers and backup systems as well.] Finally: this policy can only covers the known, authorized devices interacting with the network. Do you really believe your security to be so complete that files cannot be transferred outside that device set by any means? [Email comes to mind first.]
In Russia, Open Source loves you.
http://www.techi.com/2010/12/putin-loves-linux-orders-government-transition-to-open-source/
The $99 Android-powered smartphone will have a disastrous effect on Android. I expect that Google and Microsoft will be battling aggressively to keep phone makers building devices with their respective software platforms and those manufacturers will first slash prices in an attempt to get market share, and then start paying carriers premiums to take their phones.
The result will be significantly decreased margins for everyone but Apple. This translates to a scenario similar to the computer market where commodity Windows computers hold the vast majority of the market, while Apple uses its highly polished products to stay far and away the most profitable computer maker.
Moreover, the ability to make money in the app market on Android will be increasingly difficult, other than through ad-supported apps.
In the end, Android phones will be a crapfest, with every app displaying increasingly intrusive ads.
>The $99 Android-powered smartphone will have a disastrous effect on Android.
Oh, yes, such a disaster it will be. Google, deprived of the critically important revenue stream from Android, will collapse. Handset manufacturers will pine for the days when they had to plow millions of dollars in NRE into supporting half a dozen different OSes because they so hate taking that money as profit instead. App writers will just loathe having potential customer bases numbered in billions rather than millions. Awful, just awful; I don’t know how we’ll stand it.
@Wet behind the ears:
You’ll have to pardon my ignorance, but to me this reads as two completely unsupported statements. First: Microsoft already appears to be “battling aggressively to [have] phone makers building devices [with Windows 7]”; at the moment, there are about a half-dozen models of Windows 7 phones, while several handset makers have already had that many different models of Android devices released. Secondly, you imply that makers will “start paying carriers premiums to take their phones”; however, the day the carriers decline unlocked phones from their networks, I expect the class action lawsuits to fly and the FCC to step in.
Please enlighten me on what source(s) you are drawing these statements from, as I see the market forces pushing in other directions.
Alex K.,
d00d i knowz wut l33t hax0rz can d0 2 ur computar!!!!1!!!!!
(Sorry; you called me Jeff K. I couldn’t resist.)
I know that implementing remote-kill on a device still leaves security holes you can drive a truck through. You’re still not thinking like a CTO of a Fortune 500 controlling untold millions of dollars in budget. To such a person, yes — having that feature means you still have security holes, and not having it means you have more security holes. Ergo, remote-kill and remote administrative takeover are going to be must-haves in your corporate laptop and smartphone procurement strategy.
@brian:
> In other words, you won’t be carrying your e-mail with you on your Android phone in any company that has to deal with HIPAA or SOX if they have the appropriate level of paranoia in IT.
Few organizations have more paranoiac IT folks than mine, but my Android is running Good Mobile Messaging (http://good.com/products/good-mobile-messaging.php) right now. The Exchange content (contacts, e-mail and calendaring) lives inside a walled garden with its own password/encryption protection. But that means I can run all standard Android apps without fighting with my IT folks for app permissions. My user experience is amazingly better with this Android solution than the Blackberry Enterprise Server I used for more than six years.
I don’t see why contracts would go away. Wouldn’t they just get cheaper? Instead of the 30-40-50+ people pay now it would drop to 15-20-25. Maybe even lower.
There are a lot of people who like package deals, my late mom did, for instance.
By making things cheap at entry, carriers could still make big money on overage charges off of that huge pool of folks who just do not pay attention to their contract or their usage thinking that it’s all cheap. These are the same folks who are not going to want to shift from carrier to carrier just to grab a few dollars off their monthly fee. Plus, give them “free” apps that eat bandwidth and you’re just printing money.
So I’m saying there is still a business model of taking advantage of the unsophisticated phone user.
Not that *I* would act that way…
>By making things cheap at entry, carriers could still make big money on overage charges
How do you do “make things cheap at entry” in a market where people own their own phones (because they’re dirt cheap, and you don’t want to lose your customizations just because you change services) and signing up with a carrier consists of popping in a SIM card?
In fact, we don’t have to speculate. Outside the U.S., a lot of cellphone markets look like this (Nokia made billions by pumping out cheap handsets to serve them), so by looking at them we can project what the post-price-crash market will look like in the U.S. The kind of chicanery you describe isn’t common, pay-as-you go monthly plans are much more so.
OT : http://newsfeed.time.com/2010/12/22/internet-legend-tron-guy-banned-from-seeing-tron-legacy-in-famous-suit/
You, sir, are truly awesome
Those millions of iPod-buyers and iTunes-users I and Jay were talking about are certainly not all “trendy metrosexuals”. The iPod (and iTunes Music Store) has crowded out most of the other players in the digital music space on the strength of ease of use alone.
True enough. But listen to what ESR started out by saying; Broadcomm’s new chip will result in cheap *smartphones*. The iPod is a different product, serving different needs, and has a different user base (though likely one that overlaps a lot with the smartphone market). Yes, you can make a smartphone that plays mp3s, but that’s where Apple’s iTunes store kicks in, to give the iPod protection from the mere cheapening of the hardware.
Though I suspect that once cheap smartphones playing music flood the market, it will take less time than Jay may think for the software problems (backward compatibility with iTunes) to be solved.
>Though I suspect that once cheap smartphones playing music flood the market, it will take less time than Jay may think for the software problems (backward compatibility with iTunes) to be solved.
In fact, as I learned today, that problem is already solved, by DVDJon’s Doubletwist code.
@ Adam Maas @esr: I’ve seen plenty of Android users go iPhone as well (and ditto webOS users). I’ve seen a few iPhone users go the opposite way but based on annecdotal observances what I’m seeing is new Smartphone users trying Android then moving up to iPhones.
No doubt, but as you admit that’s only anecdotal evidence; the market share figures are telling a different story.
> Are you also dismayed that the manufacturer of your automobile also does not offer insurance directly?
No. I’m not dismayed by the lack of loss / theft insurance, but I’d be interested if Apple provided it. I would be interested in such on other MP3 players, for my own selfish reasons. I do buy insurance from my carrier on my phone, since the contract lock in on my expensive phone makes this a good idea. Since Apple is more expensive, it’s loss / theft is more damaging, increasing my interest even more.
> If I were an Apple shareholder, I’d seriously wonder what the hell they were smoking were they to offer such a thing. Especially since the things seem to be easily lost.
And if they had done their due diligence, written their user contracts carefully, and obtained the needed underwriting I’d expect you to be satisfied.
> Actually, your report of your flashlight app failing is the first real fragmentation problem I’ve been able to identify.
I’m also seeing a lot of comments from both application developers and application purchasers via the App Market which lead me to believe that people believe this to be a real problem. Since it has changed which apps I download, I would expect that developers have incentive to fight this perception if it is merely FUD. Since I have a good reason (more diverse hardware) and user / developer reports I’m thinking this is a real problem, Eric. Why do you disagree? Is Google that good at subtracting away real hardware differences via software layers in an immature market?
Yours,
Tom
>Since I have a good reason (more diverse hardware) and user / developer reports I’m thinking this is a real problem, Eric. Why do you disagree?
Because I’m hearing rumor rumor rumor and, other than your report of your flashlight app failing, no actual evidence for the proposition. If it really were a common problem, I’d expect my Google searches to have turned up a lot of end-user horror tales. I’ve seen nary a one.
On the flashlight problem – the reason for fragmentation here is that there is no way through the Android API to interact with the camera flash. It’s not supported. Therefore, every app that does this must go native, and have special code for each and every phone. Obviously, this is impractical for a single app, so you see tons of flashlight apps that only support one or a few models.
I’ve been surprised on every new Android API release that this has not been included.
These look like Android app fragmentation issues to me (all from the last month):
http://forums.directv.com/pe/elementDisplayRedirect.jsp?elementID=10812216
http://www.appbrain.com/app/extended-controls/com.extendedcontrols
http://www.meizume.com/meizu-m9-apps/12521-android-apps-compatibility-issues.html
http://stackoverflow.com/questions/4497831/how-to-make-an-application-compatible-in-all-android-phones
http://osdir.com/ml/Android-Developers/2010-12/msg03454.html
>These look like Android app fragmentation issues to me (all from the last month):
OK, let’s look at them.
The first was fixed by an upgrade of the app, suggesting that the problem was a plain old bug. If it had been the kind of API divergence bug people mean when the “fragmentation” specter is raised, it probably wouldn’t have been possible to get the app to run portably. Half point, awarded only because it’s possible the fix was some kind of wacky JNI hack rather than going through the Android API.
The second one might be real; it’s difficult to tell from the listed complaints. I’d buy the fragmentation-did-it theory more easily were it not for comments like “But sadly the new version which came out a few weeks ago is full of bugs”. I’m sure shoddy and rushed coding can simulate any level of fragmentation issues you want to stipulate. You get a nod for turning one up with complaints from real users, though; I haven’t seen that before. You get a half point for this one on the possibility that it wasn’t the developers just fucking up, but I’m being generous.
The third one looks real. This is absolutely what people mean by API breakage due to hardware divergence. There’s a problem with gigging Android for this, though, in that it looks like the problem arises from trying to run the hardware at a resolution that Android isn’t designed to support. Oops. Of course you’re going to get “fragmentation” when you do that. Duh! No points.
The fourth and fifth ones are just lame. You damaged your own case by bringing them up. Nobody runs around croaking that the PC is “fragmented” because there are different screen sizes and resolutions in the world; it’s the developer’s responsibility to cope, and coping is not rocket science. As the reply to the fifth one says, “Or, design layouts that adapt well to different screen sizes, and supply appropriate drawable resources for screen densities.” You get minus points for being silly.
Your score: two weak maybes, one they-broke-it-themselves, and two that you should have been ashamed to bring up. Really, is this is the best the fragmentation hunters can do? Tom De Gisi’s broken flashlight app was a better indictment.
> First: Microsoft already appears to be “battling aggressively to [have] phone makers building devices [with Windows 7]“; at the moment, there are about a half-dozen models of Windows 7 phones, while several handset makers have already had that many different models of Android devices released.
So Microsoft starts paying carriers to take WM7 phones. Google responds in-kind. Microsoft ups the ante.
I’m sure you can take it from here.
> Secondly, you imply that makers will “start paying carriers premiums to take their phones”; however, the day the carriers decline unlocked phones from their networks, I expect the class action lawsuits to fly and the FCC to step in.
It doesn’t need to go that far. They only need to make it such that you can’t get the SIM card without a contract. The carriers will just make the contracts reflect the new reality, you’ll be locked-in for several years just to get inexpensive service.
The “class action” lawsuit would be on the basis of monopoly power, but the FCC (Congress) sold the spectrum for exclusive use in the first place.
Once you know you’ll be locked-in, the cost of the phone is a minor issue. People are more than used to paying $199 now, and getting locked in for 2 years. Eric’s magic $89 phone only discounts the NPV $110 over 24 months.
Bang! You could have had an iPhone 4 (or a high-end Android phone) for another $4.60/mo.
The $89 phone won’t have an OLED screen, or gobs of flash/ram. It will be stripped.
Moreover, that ‘dual core’ part … runs at 500MHz. Existing single-core Android-focused SoCs run at 1GHz or above. So ask yourself and your readers these questions:
How well does Android schedule Java threads onto multiple cores? We’re not talking about the linux kernel here (though an MP linux kernel is slower than one built for uP), we’re talking about Dalvik (which would need to be modified) and the userland pthread libraries.
And the biggie:
How many apps are written to take advantage of multiple cores?
Over on the other side of 101, Apple’s been pushing its developer base with OpenCL and GCD for over a year now. libdispatch has been released under an Apache license.
now…
ask yourself when you might see support for blocks in Android.
I sure hope you’re right. I have refused to have a cell phone for a long time now, because I’ve found all the available deals to be a monopolous rip off. The contracts have all been absolutely unacceptable to me, and without a better option, I’ve just decided to boycott them altogether.
I do have a cheap landline from my cable isp with good service. I feel fine about that since I’m not paying the phone companies I hate so much.
But boy would I like a nice smart-phone. Just the technology, seperate from the venders. Preferably with anonymous pre paid cards as an alternative to subscription service.
> Broadcomm’s new chip will result in cheap *smartphones*. The iPod is a different product, serving different needs, and has a different user base
An open question has been, “where are Android the clones of the iPod (Touch)?” Samsung has announced the Galaxy Player, but we won’t know pricing until CES. If it comes in significantly less expensive than an iPod Touch, then your argument is supported. If not…
As for Eric’s “I think Apple’s hopes of retaining market share above 10% will disappear into that chasm, especially since app developers are quite capable of seeing where the price and volume winds are blowing.”
Cricket already sells a Huawei Android phone for $99 (this is discounted from $149. You’ll need to pre-pay $55 for a month of service though.) This is $10 away from jlg’s magic $89 price point. The Broadcom chipset can’t change things much from where they are.
I still predict a race to the bottom for low-end Android handset makers.
And that Apple won’t be affected much.
If you think that price alone is going to dictate who wins in the cell phone market, you need to ask yourself a couple questions:
Why do people buy brand name cars at a premium (say Honda) .vs worthy transportation (Isuzu)?
Why do people spend more for branded (Nike) running shoes?
Why do people line up to buy Apple’s phone, and not for any Android phone?
@wet behind the ears:
You’ve made a few good points, so I think I’m following you a little better… but I’m still unclear on one thing:
I still haven’t seen support for this statement. Cellular carriers don’t care what device you use, as long as [like you said] the end user signs a contract. So, to me this handset maker payola story reads as castles-in-the-sky dreaming; it has no basis in the current market and there’s no visible market forces moving this way. Be glad we are not in wikipedia-land, where you would get the passive-agressive “[Citation needed]” stamp.
As to your new points:
This question is an unknown: as of today, we have not seen the public release of any multi-core mobile hardware, Android or iOS based alike. [Not that I expect either one to have problems, mind you: both are built on technology that performs admirably on the desktop, and there is no major changes from that to the mobile hardware.]
Again, as we have not seen any multi-core hardware released yet, this first question is hypothetical and academic; in fact, it begs what I see as the more important question: how many applications need multiple processing threads? [By which I mean to exclude the UI-handling thread, which every good application should use.] If your application continues background processing when the user focus moves, Android already forces a lot of thread separation into notification service classes and, as this processing would not change with user focus, for most applications I would expect this to be adequate.
This is not to say you don’t have a point about never seeing GCD-style block support in Android: however, this is because Android’s existing AsyncTask class seems to fill the same light-threading role as GCD, but abstracts away all thread handling, and is a reasonable learning basis for using classical style threads.
@wet behind the ears:
Please pardon the aggressive rephrasing, but I immediately read this as “Why do people line up to buy Apple’s computers, and not for any Windows PC?” We’re not saying Apple is going to get cut out of the market, but that when you compare a $200-with-contract phone to a $100 no-contract one, the majority of people are going to go with cheap, no matter the difference in quality. This happens with cars, shoes, computers, furniture [Ikea], food [McD’s]… if you see there being an exception to this market rule for phones, you’ll need to explain why.
@esr
>The first was fixed by an upgrade of the app, suggesting that the problem was a plain old bug.
Ultimately, aren’t most fragmentation issues going to be fixed with an app upgrade?
>Nobody runs around croaking that the PC is “fragmented” because there are different screen sizes and resolutions in the world;
OTOH, up until now, apps for your phone has been a lot more like buying for a game platform than buying for a computer.
>“Or, design layouts that adapt well to different screen sizes, and supply appropriate drawable resources for screen densities.”
It’s worth pointing out that a lot of developers skip the whole different screen size issue by a) running at less than full screen and b) when running at full screen, choosing a sufficiently low resolution that is likely to be on most screens. Neither of these is a particularly appealing option for a low resolution 4 inch screen.
>Ultimately, aren’t most fragmentation issues going to be fixed with an app upgrade?
Yes, the question is whether the fix involves a platform-specific JNI kluge or not. If it does, I’m willing to say we have fragmentation. If it doesn’t it’s a simple failure to use the Android API properly.
>OTOH, up until now, apps for your phone has been a lot more like buying for a game platform than buying for a computer.
True, but there are lots of reasons other than platform fragmentation that this was never going to last as smartphones became functionally more like computers.
>It’s worth pointing out that a lot of developers skip the whole different screen size issue
…by being lazy and short-changing their customers. This is a pretty weak basis on which to indict Android.
Really, I must say that Android is weathering this criticism better than I had expected. The fact that critics have to advance such lame and desperate examples of supposed “fragmentation” suggests to me that there aren’t enough real ones to matter.
@tmoney:
On the gripping hand, game platforms now have to support various “high definition” television resolutions, and you don’t tend to hear complaints about new XBox or PS3 games that fail to run when switched between standard definition, 720p and 1080i. [You can find stories from when high definition first appeared, but then again there were also problems with old iOS apps on the iPad when that first released.] Anyone trying to argue that resolution alone is a “fragmentation” problem of smartphone platforms will need to explain what is so unique about Android hardware that the same complaint cannot be leveled elsewhere — as Eric implied, crappy coping methods (such as “choosing a [universally] low resolution”) aren’t the fault of the platform.
Or c) making the application cope well with the various resolutions. If the developer can’t be bothered to do that, then his application is a pile of garbage. Even the GNOME and KDE developers, whom the Apple crowd loves to beat on senselessly for being clueless about GUI design, get this one right.
It’s boneheaded GUI design, not market fragmentation.
Eric, this is off-topic here, but I couldn’t find a recent post where it really fit:
>P.S. I have a long bet that SELinux is an NSA backdoor. Any takers?
From Why I Don’t Use Tor
I find this discussion on the imminent collapse of Android rather silly.
Whatever the relative strengths of iPhone, RIM, W7, Symbian, which is more likely to sell 2 Billion smartphones? Except for a $75 Android.
Because that is the scale of global market demand. And cheaper simply means more phones sold.
# esr Says:
> Really, I must say that Android is weathering this criticism better than I had expected.
I think there is another point to make here. There is a big pareto thing going on here. 80% of the users only use the 20% most popular apps. If these 20% apps work (which they do), then the impact of broken apps in the tail isn’t really all that impactful.
However, the argument can be made that there are key apps (“hot new game” or “watch my movies” come to mind) that don’t just work well. However, as Android becomes more dominant then the makers of these applications will move their resources to Android.
FWIW, I work in Windows software development. One of the lessons you learn in that field is that people will put up with all sorts of crap if the application does what they need done. (I am, of course, not referring to the apps I write. ;-)
The bottom line is you have a choice: tightly restrict the hardware configuration: pro — tightly control the platform for apps, con — decrease velocity of innovation; or loosen up the hardware configuration and reverse these. Life is full of choices — who knew?
I should also say that I agree with Jeff Read in that Apple have made the end to end experience for developers great. In particular they have made it easy to get paid (though that whole capricious app approval process sucks.) From what I hear (and I’m not an Android developer) this is much more difficult in Android.
Now the real question I want to know the answer to:
Catherine Raymond Says:
> True enough. But listen to what ESR started out by saying;
Do you REALLY call your husband ESR? What’s wrong with “pookie” or “sweetums”?
>Do you REALLY call your husband ESR? What’s wrong with “pookie” or “sweetums”?
No, Cathy does not call me ESR in private. She’s making a concession to the milieu of this blog.
I am pleased to report that she does not call me “pookie” or “sweetums”, either. (Ugh!) She does have a funny pillow-name for me which she occasionally uses around our longer-term friends; Jay Maynard will know it, I think. But such intimacies are not for the rude gaze of the net.
# wet behind the ears Says:
> Why do people buy brand name cars at a premium (say Honda) .vs worthy transportation (Isuzu)?
Because they have never heard of them and/or they are worried about quality. Which is to say the customers are paying Honda (in the car price) to ensure the availability of quality information about their cars.
Now, you tell me, why are there more Honda’s on the road than BMW’s?
> Why do people line up to buy Apple’s phone, and not for any Android phone?
Because Apple has a throttled retail channel? Since when was “lining up” to buy something a good thing?
As a smartphone game developer, I can’t say that device fragmentation within the Android platform is of any major concern to us.
There are three significant factors that vary between Android devices: screen resolution, input options and device performance.
Additional input options (keypad, trackball or whatever) are only an issue insofar users expect you to support each of them. The baseline setup of accelerometer and touchscreen is always available.
Screen resolutions I think we’ve covered.
Device performance is, to me, slightly trickier. At this point, as we’ve touched upon, many users (primarily non-technical users) do not think of their smartphones as computers, and expect any software that can run on it to run well. This has been mitigated to some extent by Googles 24 hour grace period (if an app is sluggish, you had ample time to uninstall if for a full refund). It will be interesting to see if there is any noticeable difference when they lower it to 15 minutes. If we want to label varying performance ‘device fragmentation’ is a separate matter, of course.
>As a smartphone game developer, I can’t say that device fragmentation within the Android platform is of any major concern to us.
I think this is pretty telling. Games generally stress a platform worse than anything else, especially the sorts of graphics-intensive games that Tommy’s company (Polarbit) does. Their failure to encounter serious issues, together with my failure to turn up end-user horror stories by Googling, combined with the weak and silly examples of soi-disant “fragmentation” advanced by Android critics here, confirms pretty effectively that “fragmentation” is at best an extremely minor problem being way overhyped.
I should note that as a veteran of the Unix portability wars of the 1980s I’m a little surprised by this. That is, from first principles I would have expected the problem to be worse than it actually is, with awkward bits like Tom DeGisi’s unsupported camera flash sticking out all over and being at best thinly concealed by a layer of Google denial. Before this comment thread, I had difficulty reconciling that expectation with the conspicuous absence of Android fragmentation FAIL stories from the web.
Now it appears that Google has done an intelligent job of containing the problem and I had my expectations of them set too low. A bit funny, that, given that I credited them with tremendous strategic foresight (apparently correctly) and that I’m regularly accused of being an Android fanboy. Perhaps I’m not fanboy enough.
>Anyone trying to argue that resolution alone is a “fragmentation” problem of smartphone platforms will
>need to explain what is so unique about Android hardware that the same complaint cannot be leveled
>elsewhere — as Eric implied, crappy coping methods (such as “choosing a [universally] low resolution”) aren’t the
>fault of the platform.
I don’t think anyone is arguing that the resolution alone is the problem, more pointing out that it can contribute to fragmentation.
>It’s boneheaded GUI design, not market fragmentation.
Agreed. The problem is, however, that cell phones and their screens are much less forgiving of boneheaded developers than modern computers, and that ultimately the user doesn’t give a crap as to why the application doesn’t work, or even if the developer is an idiot, they only care that the android application doesn’t work for their android phone. Would that we could change this about customers, tech and tech support would be a much more relaxing world. Alas, we might as well wish for unicorns.
Also, we have to agree that at some point a developer must assume some sort of minimum simply to display the information needed. I just set my computer resolution to 320×240 and I can assure you that none of the applications on my linux box handled that well. But I wouldn’t call those “piles of garbage”
It’s like I said in some thread a while back in regards to android sales, a flip phone dumb phone running android would be an “android phone” but certainly not for any practical purpose. I wonder if we’ll begin to see phones categories (smart phone, feature phone, dumb phone etc) begin to be solidly classified based on specific hardware availability.
Incidentally, in my digging about this issue, apparently Google has gone with the “developers specify the minimums you need, and we won’t show it to people who can’t use it” method
http://developer.android.com/guide/appendix/market-filters.html
Which might explain why, if there is a fragmentation problem, you don’t see many complaints from users.
@ Jessica
>Because Apple has a throttled retail channel? Since when was “lining up” to buy something a good thing?
Depends on your perspective, but almost always. See concert tickets, theme parks, Tickle-Me-Elmo
Although, gas, water and food are usually not good things to be lining up for.
So I guess, lining up for any non-necessity is generally a good thing.
tmoney Says:
> So I guess, lining up for any non-necessity is generally a good thing.
Nonsense. Lining up for all these things is bad. Since when did people go to Disneyland because they liked hanging out for two hours in the line for the Haunted Mansion? Since when did Mom really delight in hanging in the rain to fight over the latest ridiculous trendy for kindergartners’ toy? They put up with the line to get what they want. How much better to get what you want without the line.
From the point of view of economics, it is the same thing. Constrained supply leading to lines and disappointed consumers. In some cases it is deliberate, because doing so might offer some benefit to the supplier. In no case is it beneficial to the consumer.
Remote wipe capability is an app download away on Android. Lookout Mobile has it, for one, and I am pretty sure the Exchange integration in 2.2 supports it as well; otherwise my corp IT dept would continue to force us to use an app they site licensed.
>Remote wipe capability is an app download away on Android.
Perhaps, but for user data other than the stuff the Outlook client owns? That I’d need to see hard evidence of before I’d believe.
>They put up with the line to get what they want.
Which is in part why lines to buy some new product are considered a good thing because it provides some measure as to how much people want the object at the end of the line. Obviously there are other factors going into this, but it’s not like new iPhone releases have 10 phones at each location (well, some of the AT&T stores were pretty poorly stocked). IOW, when they stop lining up around the store / block for new iPhones is when the iPhone is dead.
Wet Behind the Ears:
http://www.readwriteweb.com/mobile/2010/12/samsung-revealing-android-ipod-touch.php
Puggg: he mentioned that in the very next sentence.
Apple isn’t going anywhere, and all this talk about carrier lock-in and price competition misses the real point: what drives Apple sales.
And the answer is (1) a high class reputation with (2) a fantastic infrastructure for modern high-tech living.
The cost of Apple products is irrelevant to people buying IPhones. Apple’s products aren’t computer products–they are luxury goods. The price tag is part of the product. People don’t buy a BMW because they care about affordable quality. They buy a BMW because that makes them a BMW owner. Sure, BMWs are great, but that isn’t the point.
So what if Androids get cheaper? Hell. Give ’em away. We’re talking Hondas and Toyotas now.
Carrier lock-in is irrelevant for much the same reason. Most people don’t care about carriers or contracts. Most forget they even have a contract, and the Apple buyers never cared anyway.
Bottom line: what we have here is a market that people are buying into–the smartphone market. It’s growing. Some buyers rank price over reputation. Those people–and they will probably be the majority–will buy based on price and be attracted by the android. A microscopic minority will care about tech specs and buy android based on openness and interoperability with Linux.
Many millions of others will want their chance at the BMW experience, and they will buy IPhones. And they won’t regret it. Apple’s phones are and will continue to be great. They will integrate with a beautiful infrastructure that boasts the unique claim of “it just works,” which is huge. Apple stores will continue to offer wonderful customer service that makes the competition look like dog shit.
Price. Carriers. Google. None of that makes a difference to Apple’s customer base–both now and in the future.
Disclaimer — I boot to Linux and bought a DroidX over an IPhone. But nobody I know will try Linux. And while they like my phone, they KNOW the IPhone, and want one.
She does, I do, and it’s …uhm… well, let me put it this way: if it were to become fodder for the net, I’d immediately rush out and buy stocks of companies that made insulin. :-)
@wet behind the ears: > Why do people line up to buy Apple’s phone, and not for any Android phone?
@Jessica: > Since when was “lining up” to buy something a good thing?
@tmoney: > So I guess, lining up for any non-necessity is generally a good thing.
@Jessica:
Jessica, while I personally subscribe to your position and absolutely abhor queueing unnecessarily (and a few years ago would have said you are absolutely right) I have to go with tmoney and wet behind the ears on this one.
Why do some reasonably crappy restaurants have lines out the door, when you can get a table immediately at other nearby establishments selling better food for less? I didn’t have a clue about this until a few years ago, when I asked somebody if he wanted to come to a fairly popular place with a group of old curmudgeons who were going there for lunch at 11:40. He said that was too early. I took that to mean that he wanted to go at perhaps 1:00, but then he made it clear he wanted to go at 12:00, just 20 minutes later. The thing is, at that restaurant, going 20 minutes later translates into a wait of 40 minutes, which is just something that I’m not gonna do.
I had seen this sort of thing lots of times in the past, but it hit me just then — some people need the presence of a line to validate their own opinion about the food/venue.
It’s just like shopping on ebay (which I do very occasionally). The only valid strategy is to decide what your price point is, where if you get the item for that amount, you’re happy, and if you lose it for a dollar more, you’re still happy, and then do a manual or automated snipe.
If you don’t snipe, there will be some pathetic soul sitting on the fence, undecided about exactly how much that item is worth, who will come in at the last minute and be ecstatic that they only paid a dollar more than it was worth to you, “you” being some completely random stranger on the net who, simply by dint of being willing to pay something, proved that the item was not completely worthless.
In a way, the people who spoil my fun on ebay are probably acting rationally. Sometimes I have a hard time valuing things too, and if I could buy something I really wanted for only a dollar more than it was worth, I would probably be happy, too. I just have a hard time imagining that how much some things are worth to me has any connection whatsoever to how much the most die-hard ebay fan thinks it’s worth, so for me, waiting to bid until somebody else does it and validates your interest is just not a rational game — it’s frustrating, annoying, and imbecilic. But for many others, it’s apparently liberating to only buy things and do things that are so popular that there is an obvious and sometimes large cost to the popularity. In a very perverse fashion, the higher the cost of being popular, the more sure these people are that they are doing the right thing.
I think a lot of “buyer’s remorse” is attributable to people who rely on external validation for the purchase, simply because their expectations are most apt to be out of line with the actual value they receive.
@tmoney:
I could certainly be persuaded that this is true, but the same evidence that would prove this would also prove that the iPhone is popular because of its popularity, and not because of its utility. In other words, to the extent your statement is true, the iPhone won’t suffer any sort of gentle decline — it will be precipitous, like the decline in the Mac until Apple found its footing there a few years ago.
I’d be very surprised if Nokia adopted WP7. Nokia has rejected Microsoft’s handset OSs repeatedly from the time MS first made one (before Symbian was even founded), and I don’t see any indication of that changing. If the options are Android and WP7, Android is so blindingly obviously a better deal for Nokia that I can’t imagine what sort of a deal MS would have to cut to make them pick WP7. Stephen Elop’s short stint at Microsoft by itself shouldn’t make any difference, unless there really is some sort of an unprecedented deal in the works behind the scenes.
I don’t see what’s in it for Nokia to get WP7 as its main platform. I can see them making some WP7 phones, but a corporate commitment would simply put them in the back seat to Microsoft, and differentiate them from the other manufacturers out there – and not in a good way. There’s just no business case there.
>There’s just no business case there.
Well, I don’t think so either, and I’ve repeatedly said that. But the trade press is sure willing to talk it up.
@ Patrick, and way off topic
>I could certainly be persuaded that this is true, but the same evidence that would prove this would
>also prove that the iPhone is popular because of its popularity
How so, unless you assume the latter to be true? I can see that being true for simple toys (see aforementioned Tickle-Me-Elmo) where aside from the basic toy utility, most of the utility of the product is derived from its popularity and position as a status symbol (both for the parents and the children). While I will gladly concede that there are a significant number of iPhone purchasers for whom the iPhone’s primary utility is in it’s position as a status symbol, there’s nothing in proving the line => popular => good thing that suggests that popularity is the only value. There are a number of people who may simply see the iPhone as the best damn phone for doing what they want with their phone.
@tmoney:
> There are a number of people who may simply see the iPhone as the best damn phone for doing what they want with their phone.
Yes, there are rational actors who will buy an iPhone for its capabilities. But, in general, rational actors are not going to be waiting in line on the first day, because, from a pure utility perspective, waiting in line for hours for something that you could get for exactly the same price a week later without waiting is not rational for most people. (There are exceptions; perhaps someone’s phone just died and they really have to have one, and don’t want to buy a cheap intermediate one.) You’re the one who made a statement that could essentially be read as “the iPhone lives or dies by the line outside the Apple store” and, to the extent that’s true, Apple’s not selling utility.
Ah, I see now. Though at the same time, we do live in an instant gratification culture, and (at least it appears to me) more and more people are lining up just to be the first to have something. That is, the decision to purchase is rational, but the choice of when and how is not.
@tmoney:
I can’t disagree with either of those two statements, yet at the same time, they seem at odds to me. It would be interesting to poll some of the people waiting in line for the latest iPhone, to find out how much planning and execution it took for them to be there.
I would guess it would depend on what part of the line they were in, with probably less planning / execution the further back you went. Some of those first 10 – 20 people are frighteningly “prepared” to camp out for a night. Although that may not be the planning you’re thinking of. If I had to guess, the first 10 or so are over prepared, these are the ones the camped out. The next 20 – 30 are probably of the “I’m close to the front so it can’t be that bad”, the next 20 are probably there mostly on a whim, but by the time the line starts moving, figure they’ve been there this long, a little while longer won’t hurt. After that I think you get to the crowd for whom “instant gratification” means I decided I wanted one today, so I’m going to wait here until I get one.
Maybe I should have said in the original statement “instant gratification for certain values of instant”?
# tmoney Says:
> That is, the decision to purchase is rational, but the choice of when and how is not.
Which is to say, in reference to your original point, lines outside the Apple store are not a measure of the iPhone’s awesomeness, but a measure of the iPhone users’ vanity. (Which as you correctly point out, does not necessarily diminish the utility.)
For sure, rarity is a feature that is desirable to some, irrespective of utility. However, it is a feature based almost entirely on vanity, which is hardly a good thing. The best example of this is the diamond. The fact is that by almost every measure a CZ is a better gem than a diamond. It has better sparkle, it is usually flawless, it is usually colorless, it is easy to set, cut and design. It is also vastly cheaper and more readily available. It has also been demonstrated numerous times that most jewelers cannot with any degree of reliability tell the difference between the two without the use of microscopes and other gemological equipment. Not to mention that diamonds are dug in the most horrendous conditions, built on the blood of Apartheid, and controlled by De Beers, an extremely nasty, violent bunch of monopolistic thugs.
Why do people still adorn themselves with diamonds? Because they are a vacuous status symbol. Personally, I think people who wear them or buy them are simply victims of a huge con job.
>lines outside the Apple store are not a measure of the iPhone’s awesomeness, but a measure of the iPhone users’ vanity.
Not necessarily. It’s possible to be gadgetophilically fascinated for other reasons.
I’ve been hugging the Android leading edge pretty closely ever since I bought a G-1 just three days after they became available. Since then, I’ve upgraded to a Nexus One and G-2 and am seriously thinking about buying a Nexus S. I could rationally defend those choices – wanted a browser on my hip, wanted hotspot/tethering, wanted faster data speeds and a hardware keyboard – but a dispassionate observer could be forgiven for judging that I just think Android smartphones are mad cool toys and will seek excuses to glom onto new ones as soon as possible. I wouldn’t even argue against that proposition.
Is that “vanity”? Not of the ordinary social-positional kind; flashing a Nexus One ain’t going to score chicks or a good table at Sardi’s. Nor would flashing an iPhone…but both may function as identity goods. That is, an iPhone user waiting in line is telling himself he is more stylin’ than the frumps who settle for Androids, while an Android fan is affirming his own nerd cool and feeling superior to the hipster art fags at the Apple store.
Then you have the really weird singular cases, like me. Android is, among other things, my philosophical/economic ideas made manifest by people who knew they were doing exactly that and meant to. For me, handling a new Android phone is kind of like dandling a bouncing new baby descendant on my knee. Vanity? You could say that, I suppose, but we don’t normally consider people vain for being proud of their grandchildren.
@tmoney: Maybe I should have said in the original statement “instant gratification for certain values of instant”?
Sure, but Hofstadter’s law says most of those people probably spent a lot more time in the line than they expected, which (by my metric) still says they weren’t thinking rationally when they got sucked into the line, and once they had a time investment in the line, they probably rationalized staying there because the time investment would be amortized over the expected life of their new shiny, when in fact, they still could have gotten the thing for less investment the very next day…
“Man is not a rational animal; he is a rationalizing animal.”
ESR:
Re: remote wipe – “https://www.mobiledefense.com/”. Remote wipe, backup, location, etc.
Re: segmentation – I’ve seen what people tend to think of as segmentation on a few apps, but it’s almost always related to system tools or apps that attempt to use hardware-related features (e.g. things that turn on the camera flash light, or use the accelerometer, etc.). I’ve always chalked it up to people not following the spec and taking shortcuts to tweak performance at the cost of compatibility. The other thing I’ve seen is that some apps don’t play well with other apps, this will go away as Google tightens the sandbox a bit on apps.
Re: Enterprise class service and corporate IP – Google is working with VMWare to create a phone-within-a-phone so that the IT dept. can put a corporately approved phone on the same hardware as a personal phone, allowing the user to customize their side of the phone while have the corporate phone locked down. Add to that the “remote wipe” above and you can simply hand you Android phone to the IT droid and 10 minutes later, you have a company phone completely controlled by the BOFH without losing YOUR phone at the same time. Sounds good, not sure if CIOs will buy it though (but if you make the hardware $75 and not contract, they might).
> if it were to become fodder for the net, I’d immediately rush out and buy stocks of companies that made insulin. :-)
She calls him High Fructose Corn Syrup?
HeFsCrS?
Glucose? (dextrose, glukus, …)
The chocolate in her peanut butter?
> For me, handling a new Android phone is kind of like dandling a bouncing new baby descendant on my knee.
You only think you understand having kids. Just ask Russell Nelson or Linus.
Again addressing @esr> “I think Apple’s hopes of retaining market share above 10% will disappear into that chasm, especially since app developers are quite capable of seeing where the price and volume winds are blowing.”
It’s going to take more than sheer volume to get Android developers on-board.
Google is going to need to reboot the marketplace to improve the UI, UX, and do a better job of filtering out spam. (No more instant app approval.) I figure a better than 50% chance of this happening in 2Q11, probably just prior to an iPhone 4S or iPhone 5 announcement.
This should help drive app profitability, though Android apps won’t be as profitable as iOS apps until 4Q11 or early 2012.
esr: The person complaining most notably about platform fragmentation on Android that *I* have heard about is John Carmack of id:
http://mobilegames.about.com/od/gamecompanies/a/intcarmack.htm
In other news, brian’s right about RIM: their target market, all these consumer phones notwithstanding, is IT departments *who do not want you to own your company phone*, because *they* have HIPAA or SOX requirements over the data thereon, and you owning the phone makes them sleep poorly.
I’ve been told that Android over VMware has been demonstrated already, allowing your phone to simultaneously have a personal and a business personality, so IT can shoot their magic bullet at it if necessary, without wiping out the pictures of your 2 year old. (Yes, that’s happened).
And to Jay: my former employer’s exex all switched from various Blackberries to various iPhones, and it was a frickin’ nightmare for me; Apple has never understood enterprise requirements (the Xserve notwithstanding), and never really cared.
@wet behind the ears:
I’m not an app developer, and don’t have any visibility into what goes on at google, but I find it interesting that you believe that the problem is so bad that google knows they have to fix it and is working diligently on it, yet you can’t imagine them coming up with a fix that keeps instant app approval in place. If anybody knows how to deal with spam, in mail, on websites, whereever, it’s got to be google. If anybody wants to design their system so that no costly human intervention is required, well, that’s got to be google too. I just can’t imagine them adding too much friction to the process if it’s at all avoidable.
@AlexK> Secondly, you imply that makers will “start paying carriers premiums to take their phones”; however, the day the carriers decline unlocked phones from their networks, I expect the class action lawsuits to fly and the FCC to step in.
You confuse unlocked phones with the economics of the carrier industry. First, the carriers loathe unlocked phones. Second, you’re going to want carrier services from *some* carrier, no? LIke it or not, they collude on price.
As for Microsoft not offering bribes^Wsubsidies, if Microsoft want to catch Google, they’re going to have to do it at a scale disadvantage, meaning that Google is going to make more off of the same searches than Microsoft will simply because they have a bigger marketplace.
To beat that, Microsoft has to commit to spending lots of money to try and close that scale gap by buying share through distribution deals and spending a ton on technology to differentiate the WM7 and search products while accepting that they don’t monetize either WM7 or search as well as Google.
If they can eventually build a product in either space that will pull enough marketshare from Google to be roughly equal, then they should start to see better monetization.
And that’s the way that cookie crumbles.
@Patrick> I’m not an app developer, and don’t have any visibility into what goes on at google, but I find it interesting that you believe that the problem is so bad that google knows they have to fix it and is working diligently on it, yet you can’t imagine them coming up with a fix that keeps instant app approval in place.
I suppose they could police the app store after the fact, but that still leads to a spammy marketplace. You say you’re not an (app) developer, shall I assume that you’re not a developer, and that you therefore don’t understand the challenges of the problem?
I don’t think the coming struggle is between Apple and Google. Rather, it will be between Google and Microsoft.
As business models go, there are currently two dominant ones: either people like your product enough to purchase it or they don’t care enough to buy it but will overlook its deficiencies if it’s “free” in exchange for their personal browsing and purchasing info sold to advertisers. The former model is Apple’s, the latter is Google’s.
Apple sells emotional experiences. The price is what users pay to be delighted by Apple’s stream of innovations and to be free of the lowest common denominator burdens and the pervasive harvesting of their personal info.
Google sells eyeballs. To be more precise, the clickstream attached to those eyeballs. Thus scale, indeed dominance, is absolutely crucial to Google’s model.
Android may be a lackluster clone of iOS in terms of UI and fluidity, but as an economic proposition it’s nothing short of an extension of Google’s desktop/online business model. Google’s model wouldn’t work with something like 20% market share. If a market is highly fractured among smaller players, business models like Google’s that rely on massive scale wouldn’t work well. As with Microsoft’s Win32 API or Office formats, scale is erected to beget inevitability. Inevitability becomes its own marketing engine. Windows had virtually no security architecture by design for so many years, even long after its costly effects became obvious globally, but because it was ubiquitous, thought to be irreplaceable and thus inevitable, it has continued to net Microsoft billions year after year. Likewise, MS Word could get away with some of the most insane formatting problems ever invented by man only because it has so dominated “desktop productivity apps” that it’s become inevitable. If anyone, even Microsoft, were to design a modern word processor today, it sure wouldn’t be Word. And yet everyone else designing a better Word has had a very difficult time of competing with the inevitable.
Like Microsoft, Google doesn’t sell best-of-class user experiences to paying customers. It sells their eyeballs to advertisers. The more eyeballs, the better. The most, the best. If it can dominate a market and thus make its products and platforms inevitable, it wouldn’t even have to care about user experience at all. Google Buzz didn’t have to have good user experience because Google management thought if they could just bolt it on top of the very dominant Gmail it would make Buzz…inevitable.
Look at the undeniable similarities between the two companies’ willingness to raise their paranoia to a level of corporate survival strategy.
During its growth period, Microsoft entered into one risky bet after another, from cable TV to office equipment automation to Dick Tracy watches. It saw threats to its core revenue base from every new development, every new player to come along. And expand and spend it did. It did, mostly because its management thought it could.
So too, Google has to be everywhere software could possibly run: wikis, cars, windmills, electric meters, audio ads, location-based services, microblogging, catalogs, print ads, web page layout apps, online answers, social networks…even when, as you may have noticed from this list, it fails to get any traction.
For Google, nearly all of whose profits depend on advertising revenue, dominance expressed as clickstream traffic is the currency. To maintain that dominance the “Don’t Be Evil” company has been willing to go into business in China despite all evidence of rampant human rights violations, get into bed with the worst phone carrier to rape net neutrality, let its “walled backlot” search become a cesspool of SEO swindlers, collect unauthorized data via illegal WiFi mapping all over the globe, risk exposing private email account data in hopes of capturing social graph info by default, favor its own properties in search results in surreptitious ways and so on.
Whether it’s on the desktop, mobile or TV, the ability to sell advertising by maintaining market dominance is everything to Google. But then what’s in it for Google’s Android hardware “partners”?
What happens when one company ties its market destiny to another’s rate of innovation? The movie “One OS, Many Partners” that we’ve seen before in Wintel theaters didn’t have a happy ending. Having secured a very fat market dominance, Microsoft displayed an embarrassing level of paternal indifference and inability to innovate.
Even Microsoft’s biggest partners complained: Acer about lack of proper tablet OS support, Dell about better server support against Linux, HP about media center innovation and nearly everyone about getting burned by the WMP/PlaysForSure/Zune debacle. At the end of its inevitability run, most of the Microsoft “partners” were left holding the bag…of stalled innovation, disappearing margins and market irrelevance. That’s the leitmotiv of the “One OS, Many Partners” screenplay.
It’s a classic dominance play, and Google is perfecting it in its rerun. For years, Google played deaf to complaints from publishers and studios about its copyright violations of their books, news and video. Until, of course, its own operations scaled enough to dominate those distribution channels to then dictate terms to content owners: “You can’t live without our traffic to your website, so let Google commoditize and leverage your properties for next to nothing.” Just like the Wintel hardware manufacturers who had no OS of their own and were thus at the mercy of Microsoft, content providers that stood by and never developed their own digital platforms find themselves now at the mercy of a dominant Google. This inevitability is worth so much more to Google that the several hundred million dollars it has already spent on Android to give it away for “free” remains a rounding error on its balance sheet.
Between Android’s market dominance and overwhelming commoditization of mobile content, stand Apple’s iOS devices and Facebook (and perhaps to a lesser extent Microsoft and Twitter). On these platforms, Google search – the key to dominance and inevitability – is either absent, highly mediated, in decline or mostly obviated. That’s why Google’s most belligerent words and actions have recently been directed towards those two companies. In a reversed mirror-effect, Microsoft used to call open source an anti-capitalist “cancer” then, Google’s Android head likens “un-open” Apple to North Korea today. Google loves to index Facebook social graph data, but won’t let Facebook access Gmail relationship graph – of course, all in the name of “openness”.
So the Android crusaders like Eric will be assulting us in 2011, swinging their $85 smartswords to demand our capitulation in a rapture of inevitability. Inevitable like Knoll, Orkut, Froogle, Lively, Health, NoteBook, SideWiki, Answers, Wave, Buzz, Nexus…like an army of 41 shades of blue. No matter. Resistance is futile.
Curiously, even the most successful Android hardware manufacturers like Samsung and HTC are hedging their bets on Google’s mobile platform either with their own OS (Bada) or Microsoft’s (WP7).
Why would experienced OEMs hedge their bets on Android if it were so open, so free and so benevolent? Let’s hope they too have seen the “One OS, Many Partners” movie and still remember the OEM extras with un-speaking roles in the “Razor Thin Margins” and “Race to the Bottom” scenes…when everything exploded.
Potentially, most aspects of what makes a phone is an unknown variable in the Android world: screen size/resolution, CPU speed, memory, storage, battery life, and myriad interface options like trackballs, buttons, stylus, touch, multi-touch, physical and virtual keyboards…and any permutation thereof.
To open source advocates this, of course, is a triumph of options. The Wintel world was anchored around the notion that a user could get its OS from Microsoft; PCs from many different manufacturers; video, audio or network cards from yet other sources; and pretty much each app came with its own UI and print driver. While this wild competition expanded markets and drove prices down, it also increased complexity dramatically. So much so that even Microsoft and Intel had to get into the business of reigning in manufacturers with reference designs and compatibility requirements.
The bankruptcy of this aggregation approach was finally exposed in the sad saga of Microsoft’s Zune. Microsoft, the principal promoter of horizontally aggregated product design, finally admitted the vertically integrated model of tightly coupling hardware, software and service long championed by Apple was inevitable for product innovation. Microsoft dumped its hardware “partners” and went solo with integrated Zune hardware, software, DRM and online store. Here’s how Steve Ballmer explained Microsoft’s shift two years ago in an internal memo:
For its own business rationale of securing unfettered and increasing access to its online advertising and services, Google is trying to organize what is clearly a disjointed cellphone market under an open source umbrella. From an architectural perspective this is a desirable approach, after all the iPhone has demonstrated the value of integrating hardware, software and services into one coherent offering that is easy to acquire, use and extend.
Unfortunately for Android, this coherence will prove to be difficult to achieve in an open source market. At every step, the Android team has to worry about accommodating disparate interests, and in turn, participants in the Android ecosystem have to worry about the complexity and variability of the platform. Such variability is necessary to attract diversity of participants but each participant dilutes the coherency of the platform.
iPhone developers do not have to worry about differing UIs or device configurations. They don’t have to accommodate all kinds of input devices from trackballs to multi-touch to stylus. They don’t have to invent their own syncing or notification systems. They don’t have to negotiate for different app stores. And as Kogan found out too late, they don’t have to worry about “compatibility and interoperability in the near future”, in the form of varying screen sizes and resolutions.
This coherence is regarded by many open source advocates as “lock-in.” The open source community seems to value openness more than innovation. In the post-iPhone era, for example, Android (while open source and backed by the largest Internet company) isn’t leading the innovation charge.
Similarly, Win32 API was proprietary but resulted in the largest app platform ever. Apple’s FairPlay DRM is proprietary but created the largest legal media ecosystem to date. So while the power of proprietary platforms to create large markets has been demonstrated, the ability of open source to create large and lucrative markets coherent enough to attract commercial developers in the consumer markets has yet to be proven.
esr Says:
> Is that “vanity”?
Yes, perhaps vanity is not exactly the right word. The social dynamics in those Apple lines is complicated. Partly it is religion (I believe so much in Steve I’m going to be first in line), partly it is an identity good thing, partly it is one up manship, partly it is a kind of “party to celebrate the new phone” kind of deal, a kind of social gathering almost. However, for sure, none if it is much to do with the original contention. It is still a deliberate marketing ploy of fake scarcity to generate buzz. Hardly a real measure of objective superiority.
BTW, I have to give you props on the word “dandling”. It is rare that such an uncommon word fits so absolutely perfectly.
@wet behind the ears:
While there is no doubt that the carriers signal to each other and reduce prices as slowly as possible, prices for voice and data on cellphones are actually dropping fairly quickly, compared to, say, cable, where prices are still inexplicably (unless you believe in the power of competition) going up for most of the country.
But, I find the juxtaposition of these two statements interesting. I could believe the carriers are in collusion, or that they hate unlocked phones, but I’m having a hard time understanding fear and loathing of unlocked phones if they are in collusion.
No, I don’t think they are in collusion (or rather, to the extent that they are, I think that consumers will still be able to fix it), and I think a lot more people are going off-contract, and whether the carriers like it or not, they have to go along and actually compete on things that people care about. Here’s an article about the trend to prepaid. I think that’s borne out, for example, by AT&T’s most recent quarterly report, which showed net retail adds of 745K postpaid and 321K prepaid customers. And AT&T isn’t the most prepaid-friendly company out there by a long shot. When you add the prepaid customers to off-contract postpaid customers, and look at the amount of cellular capacity that has been added in the last decade, I think there is a marketplace ripe for competition. Sure, the carriers may not like that cell phones are cheap for the customers now, but they’re going to have to live with it.
If Microsoft is going to try to compete head-to-head with Google on search, more power to them. But once Microsoft thinks they have to subsidize WP7 to increase their search revenue, they’ve lost. The dirty little open (source) secret of Android is that google is heavily subsidizing a loss of control — by the carriers, by the handset makers, and by the handset maker software suppliers — because then people are free to do search how they want to, and that is providing (so far!) excellent results for google. Google’s not in this to own the space; google’s in this to keep anybody else from owning the space.
You could assume that, and it probably wouldn’t be the first bad assumption you’ve made.
> You only think you understand having kids. Just ask Russell Nelson or Linus.
Hehe.
@Last Word:
They need a high market share for search and other net services, but they just need to make sure that open platforms allow customers to get to their services. They don’t actually need a huge share with Android; they just need to make sure that nobody else has that huge share.
Getting best-of-class search and email for free in exchange for seeing a few ads is apparently considered by many to be a bargain. Not sure how it’s at all like microsoft, though.
Although I have been disappointed by some of google’s actions, anybody who claims that collecting wifi data from open access points is an unethical breach of privacy needs to be taken out and shot. Possibly repeatedly, if only to impress upon the growing crowd the seriousness of such a baseless accusation, and the importance of securing wifi access points properly. This argument alone destroys whatever credibility you thought you were throwing around, even before you lamely attempt to argue that they favor their own results in search — something that Microsoft has been proven to do, but to my knowledge, google has not been shown seriously to do.
What part of “open source” did you miss?
OK, let me get this right. Google is evil because they index whatever website owners say they can access, but they won’t show facebook my contact list, but a minute ago google was evil because they might accidentally leak my contact list? My head hurts.
Any big company with a big budget is going to have multiple baskets for its eggs. Nothing to see here.
Yes it is going to explode. Yes there will be true competition for handset makers and for carriers. Yes, google will have helped. And yes, google had a profit motive all along. We’re proud of you for noticing.
That’s the stupidest thing I’ve seen written today. Nobody regards the coherence as “lock-in.”
I wasn’t sure if you worked for FaceBook, Microsoft, Apple, or some other huge conglomerate. Thanks for clarifying that it’s Microsoft.
Whoa, let’s drill down on this one a bit.
What Google did: they captured packets from open wifi networks. I think we need to be really careful to be precise; “wifi data” is a loose phrase, and the first time I read that I assumed it just meant they were collecting wifi network names, MAC addresses, and so on. I was wrong. They were capturing the full packets going across those networks, and that absolutely is a privacy violation. It is equally true that you’re dumb if you leave your wifi open like that, but the ability to access a network does not imply permission to access that network.
That said, Google’s provided a very clear explanation of what happened and they’ve taken great pains to make things right. I believe that it was an accident, and I think any reasonable person would concur. They have been very open about what happened. I don’t think it’s reasonable to criticize Google on this point.
Argh, lack of edit. Also, Google didn’t capture full sessions or anything like that — they captured whatever fragmentary bits of data were available as the cars drove by. This is, again, evidence that it wasn’t a malicious act.
@Bryant:
> I don’t think it’s reasonable to criticize Google on this point.
Me neither; I just worded it a bit more strongly :-)
>> For me, handling a new Android phone is kind of like dandling a bouncing new baby descendant on my knee.
>You only think you understand having kids. Just ask Russell Nelson or Linus.
*Grandbaby* — You get to play with them, admire them, then let mom or dad change the diaper.
(Yes, I am very well aware that grandparent isn’t that simple either, but it makes a real good story — and has quite a lot of truth)
I started laughing hysterically when the (current) minister for communications in Australia perked up a full head of steam about the wifi data issue. It was billed by the news as a “Conroy says google has a poor privacy policy” and i’m thinking “woah… something he may have actually gotten right” and then it was that…
I sent an email to their office (which i know will be ignored but still) describing all the many ways that Google unnecessarily tries to track you and other methods that could constitute a privacy breach(found by typing google wikipedia privacy no less) but seriously? Trying to nail them for collecting data that is loudly broadcast into the street in unencrypted form is like trying to nail someone for espionage for hearing something that some four star general is screaming from the top of his lungs on the nearest building.
It’s retarded and if I didn’t have such a low opinion of him already after the netfilter debacle that would have clinched the deal. Of all the politicians to slip the speech writer muzzle, why did it have to be him.
> but the ability to access a network does not imply permission to access that network.
In an environment where the top 4 or 5 wifi clients are all promiscuous about offering to connect you to any noticed open wifi net, no, I don’t think that’s at all a reasonable assertion, practically.
I fail to see how capturing unencrypted broadcasts is in any way a privacy violation. If you want your radio transmissions to be private, don’t transmit them in the clear into the space occupied by my antenna.
Similarly, if your WAP is unencrypted, and provides a DHCP lease offer to my laptop, you can’t complain that I’ve “gained unauthorized access” to your network by accepting that offer. If you wish to exclude anyone from your network, that’s between you and your WAP.
Huh, OK. So the general principle here is that if I fail to take appropriate security measures, you have the right to do whatever you like with whatever access I’ve accidentally given you?
>Huh, OK. So the general principle here is that if I fail to take appropriate security measures, you have the right to do whatever you like with whatever access I’ve accidentally given you?
Cease this trolling now, Bryant. We know where it’s going – burglars and unlocked houses – and we already know know why the analogy and the inference is bogus. So do you, unless you’re much stupider than I think you are. You should be ashamed of yourself.
@Bryant
“Huh, OK. So the general principle here is that if I fail to take appropriate security measures, you have the right to do whatever you like with whatever access I’ve accidentally given you?”
People can look into your porch from the street. So they can take a photograph of you sitting their.
They invade your privacy because they should not look your way? I agree that they might not be allowed to post those photographs with name and address on the WWW. But not taking a photograph of your garden and porch?
There is a difference between Google capturing radio transmissions, and them posting your WiFi address and location on-line. The first might be perfectly acceptable, the second would most certainly be unacceptable.
Generally speaking, at least in the US, you forfeit much of your right to privacy vis a vis photographs when you open yourself or a scene to public view. You could argue that by failing to take appropriate security measures, you’re opening your network to the equivalent of public view. I just want to understand if that’s the argument you’re making.
@ Last Word
Are you also Kontra from Counternotions?
http://counternotions.com/2010/12/28/the-unbearable-inevitability-of-being-android-1995/
Er, yes, we do. That’s not where I was going. I was going to the creeps who took advantage of your security error (one that I’ve made myself, btw) to inject spam into your blog.
Look, this isn’t a hugely controversial issue unless you’re fixated on making excuses for Google. The EFF concurs with me, and while they’re not magically always right or anything, it’s a relevant data point. They also said that you should encrypt your wireless, which is equally true. Collecting network traffic from open wireless access points is a mistake. Google said it was a mistake and apologized.
@Bryant
“Collecting network traffic from open wireless access points is a mistake.”
There is a boundary crossed when you inject traffic into an open network that is not clearly labeled as being intended for that purpose. Just as an open front garden does not mean I can picknick in it. Also, you “infringe” both copyright and privacy if you republish anything private that was in the packets. In this last case, it is simply much better not to collect the stuff in the first place.
But there is no reason to condemn anyone mapping the WiFi access points for, eg, location mapping (where am I? without GPS).
That’s an appeal to authority.
Only because of the bad PR they received for it.
I’m not making excuses for Google here at all. But I 100% with agree the The Monster here in that if your wireless access point is spewing broadcast packets over the free and clear airspace, you have absolutely zero expectation of privacy there. This is, quite literally, the equivalent of shouting “Hey! Here I am! Over here!” I suggest you read and understand the definition of broadcast traffic before writing your next reply.
Morgan Greywolf Says:
> if your wireless access point is spewing broadcast packets over the free and clear airspace, you have absolutely zero expectation of privacy there.
As a general rule I agree that whatever you do in public is pretty much in the public domain, and you have no expectation of privacy. However, what constitutes “in public” is not necessarily 100% clear. I think WiFi is a case of this. Consider this. Lets say I am Natalie Portman and I don’t want the press or scuzzy bloggers putting pictures of me naked on the Internet.
Can I strip naked at a public beach and have an expectation of privacy? Probably not. When I am at home in my bedroom getting into the shower, do I have an expectation of privacy? Yes. Do I have an obligation to close the drapes, to protect against telephoto lenses. Probably. Do I have an obligation to install a Faraday cage in my walls, to protect against back scatter X Ray radars? Probably not.
The point is that you have to take the sort of reasonable precautions that reasonable people would expect you to take, and you do not need to take extreme precautions.
Now lets say Natalie installs WiFi in her house. She is no computer nerd, and doesn’t really fully understand the technology, but she figures it is a home wireless. Can the press grab her unencrypted packets? What if she puts in MAC filtering, can the press spoof a mac address, and use that to defeat her security? Can they still grab her packets? What if the press guy has software installed that does that automatically? What if she puts on weak encryption, can they still grab and unencrypt her packets? What if she puts on stronger encryption, say DES? Can the press still grab her packets and run a decrypter for a few days and grab her packets? What if she puts on AES encryption, but chooses her dog’s name as password? Can the press still grab and hack her password?
Neither Natalie, nor the majority of the public have any understanding of the security settings on a WiFi router. They sure as heck have no idea of the difference between DES and AES. Does that lack of highly technical knowledge exclude them from a right to privacy on their in home networks? How much expertise and skill are you required to have before you are allowed to keep your email private?
I don’t think it is black and white.
Jessica Boxer Says:
> Nonsense. Lining up for all these things is bad.
> Since when did people go to Disneyland because they liked hanging out for two hours in the line for the Haunted Mansion?
Disney Tackles Major Theme Park Problem: Lines
http://www.nytimes.com/2010/12/28/business/media/28disney.html
@ Marco,
Like I said, it’s all about perspective. The lines suck for the people in them, but if they weren’t there (baring a deliberate attempt to reduce or eliminate them), it would suck even more for Disney. It would mean no one thought Space Mountain was worth waiting in line for.
@Jessica Boxer
Shouldn’t that be naked and petrified? With hot grits? Oh, wait, wrong site.
Okay, okay! Seriously, you have a lot of interesting edge cases, but we’re not talking about edge cases here. We’re talking about packets sent en clair over the public airwaves. Every consumer WiFi device sold today comes with a cute little installation CD that automagically enables WPA-PSK and/or packet filtering. The instructions in or on the box tell you to pop the CD in your Windows or Macintosh PC and run the setup, which does backflips to prevent you from not enabling security. It’s perfectly reasonable to expect the consumer to do that. If you can’t read and follow simple directions, you have no business having a computer in the first place.
@Jessica:
> Lets say I am Natalie Portman and I don’t want the press or scuzzy bloggers putting pictures of me naked on the Internet.
A bit late for that, isn’t it?
Anyway, while I could agree in principle that, based on intent, there are shades of gray involved, I don’t think google got anywhere near the darker shades, or that the governments which are effectively saying to google “you need to preserve the data and show it all to us so we can determine for ourselves just exactly how bad you were” are anywhere near the lighter shades. Let ’em do their own damn snooping.
@tmoney:
> It would mean no one thought Space Mountain was worth waiting in line for.
Brings back fond memories. When Apple announced the Apple III, they rented Disneyland for a night, and gave out a limited number of tickets to attendees at the National Computer Conference.
I think I rode Space Mountain something like 13 times. It was awesome.
Peter Vesterbacka from Rovio (Angry Birds devs) on developer ecosystems and fragmentation:
I hear sentiments similar to Vesterbacka’s on developer forums all the time when the topic of mobile development comes up. Mobile devs overwhelmingly prefer the iPhone. Apple’s tight-fisted end-to-end control of the device, the software, the development tools and the ecosystem has made iOS the best mobile platform for third-party devs to target. Android has a long way to go to catch up, and millions of cheap smartphones won’t fix this; in fact they’ll make it worse.
@Patrick Maupin:
I worked at Cedar Point for a summer in my youth. I remember doing the same for the Mean Streak and the Magnum XL-200. It was cool riding those without the lines. :)
@Jeff Read:
At this point, I doubt if the developer experience is a significant priority for google. Oh, sure, they’ll keep working on it, but google’s model is about providing eyeballs to advertisers. They need enough developers to help lure the requisite number of eyeballs, and frankly, they’re already there. How many smartphone apps do most people need, anyway?
Unless all the developers exit the platform en masse, google probably couldn’t care if some percentage of them abandon ship.
Patrick, indeed. To Apple the end user is the customer; to Google the end user is the product.
@ Jeff Read
> Apple’s tight-fisted end-to-end control of the device, the software, the development tools and the ecosystem has made iOS the best mobile platform for third-party devs to target. Android has a long way to go to catch up, and millions of cheap smartphones won’t fix this; in fact they’ll make it worse.
Huh?
I’ll be forever grateful to Apple for introducing the AppStore. Our entire industry (mobile app development) changed thanks to it, and it’s been pretty much all for the better.
That said: the amount of control Apple excerts over iOS development and AppStore deployment is not a core strength of the platform.
Waiting for approval for a week or more for a minor app update and being failed for some trivial detail that has been in the app since 3 revisions back is not a good thing. Being forced to rely on a specific piece of hardware to deploy your app is not a good thing. Being restricted in what kind of apps you can and can’t sell is not a good thing. Having one single device configuration to consider might be a good thing in that it allows you to cut some corners, but not a good thing in that you might end up a lazy and careless developer.
Apple are top of the class when it comes to end-user experience and interface design, as well as when it comes to the number and range of applications on offer. The first of these they’ve achieved by recognizing that user experience is a crucial factor and employing good designers. The second by offering developers access to a large installed base of users aching for content, and a way to reach those users that did not involve carriers (who would skim much more than 30% off your profit, limit you in many more ways than Apple and were very difficult to approach for anyone but a major publisher).
Developers will go where the money is. If “millions of cheap smartphones” turns out to make Android development more difficult (and I’m far from convinced that it will, to any great extent) it’ll also bring “millions of potential new customers” for your apps.
Again: Android, as an operating system and as an ecosystem, is far from perfect – but platform fragmentation is not one of its significant faults. Differation does not automatically equal fragmentation. If your only experience of mobile development is iOS I can see how all these different-looking devices might seem scary, but the effort involved in supporting multiple resolutions etc is minimal. I’m not saying there isn’t any work at all, but it isn’t much and it pays off many, many times over.
But in the same article he says :-
Reading between the lines I see that as “we’ve hired N+1 ObjC developers for iPhone games and none of them like programming in Java”. (Yes I accept that is an exaggeration, but if you extend “Java” out to toolsets and other things that aren’t necessarily built to work together…)
With one exception (that of debugging specific issues) I don’t think the plethora of new devices is going to mean anything… every source i’ve seen quoted here is saying that outside of some special cases, device fragmentation is a dead issue. What we’re waiting for is the million and one “phone app” shops that have geared themselves towards developing in the walled garden with 1 to 2 year development cycles (common in gaming) to change lanes to Java. The other possible issue is the perception (rightly or wrongly) that google market users don’t pay money for things.
I do think the “release the same game free with ads” model is going to wither and die (unless ad revenue is comparable on a long enough timeline to rival the ROI from a single sale… to which history says no) since imo the ads are fairly unobtrusive and thus removing them isn’t as high a value as you might expect. I’d like to see games with additional paid content (e.g. episodal stories)
>I do think the “release the same game free with ads” model is going to wither and die
I don’t. Especially not in the variation Alchemy uses. “Here’s free game with advertisements. You can buy a mod that turns off the ads.”
This is very clever. It turns the negative value of ads into a revenue generator, so Alchemy wins coming and going.
Funnily enough, in one interpretation actually the opposite is the case.
To Google all that matters is bums on seats. They don’t care if your phone is a 6″ cubed brick that displays in monochrome so long as people buy it and look at their adverts. All they care about is the end customer. In a way this is partly why android(or something resembling it) will take majority share. For people who feel that physical keyboards are essential, an iPhone is a complete non-starter. End of Story. (or people who like flip phones or )
To Apple the phone is supposed to represent unparalleled style and design. If you don’t like the way it looks, if you don’t like the way it works. Tough. Live with it or go somewhere else, we don’t want your business here. This was best highlighted in the initial reaction to complaints about the antenna. “There’s nothing wrong with the phone… you just need to hold it like this”.
P.S. i’m not saying that Apple’s UI design is crap or not user focused. What i’m saying is that their UI design is good because you can’t claim a phone that is the paragon of design and style without having a well designed UI.
@esr
>This is not a fragmentation problem. Rather, it’s all about the DRM and the platform being so open that the DRM is critically exposed to attack.
Agreed. I almost added that I considered it a special case and that I am rooting for Android. Instead I just opted to relay the info devoid of cheering one way or another. Besides, I detest DRM and it’s stench of a general lack of imagination.
Perhaps we should call such telephony implosions the Mustakrakish Effect.
From the sounds of it however it’s not working. The negative value of ads is not enough to motivating buying the same game you’ve already gotten for free. So in my mind the model lives and dies on whether the ad revenue is enough to sustain the company and i’m just natively dubious about sustaining a business on revenue from selling clicks.
# JonB Says:
> i’m just natively dubious about sustaining a business on revenue from selling clicks.
And yet my Google stock went up $25 this week.
I probably have terminology issues but (in my head at least) Google doesn’t (just) “sell clicks”, they “on-sell clicks”. On-selling clicks has been a moderately successful business if you can keep it sustainable by maintaining ROI to the purchasers. (Which is another issue I have with ad revenue and games, i’m not likely to be interested in clicking a link or even moving my attention to that black box at the top of the screen when i’m focusing on a game)
And while i’m willing to list Google as an exception, that doesn’t stop me from being natively dubious in the general case.
Concerning Android fragmentation:
I’ve had more fragmentation problems with iOS/iPhone than with Android
and a discussion of the essay on HN.
@william
So the interesting comment for me is the one that points out that any credible review process should have picked that up at minimum.
@William B. Swift. To be fair, the guy could have installed an older version of Xcode in a directory different from the default. This is more of a cultural problem than a technical one.
Note that with the Android SDK, emulators for all the different versions of Android are included in the box and the differences between different Android versions are highlighted clearly in the docs.
Nice article at the register on this, including a discussion about how Android is moer developer friendly
http://www.theregister.co.uk/2011/01/05/android_beats_apple/
Sorry, the Android explosion is largely an artifact of carrier lock-in, Android being an adequate booby prize for when your carrier does not (yet) offer the iPhone.
>He points out that in countries where the iPhone is on multiple carriers and competes with Android, the iPhone outsells Android phones.
On current market-share trends, this won’t continue for long. Android has been gaining share at Apple’s expense worldwide since early 2009. Whether you interpret this as head-to-head competition or a fight over dumb-phone conversions (and I think it’s mostly the latter) the fact is that Android uptake is faster than iPhone uptake, which is eventually going to result in Android having a larger aggregate userbase. In some important markets (notably East Asia) this is almost certainly already true, though we won’t know it until ComScore and similar outfits release their 2010 summaries.
UPDATE: I actually punctured the Apple-outsells-Android-overseas myth on this blog months ago.
This is talking about australia only but something else to bear in mind is that Android seemed to be very US centric until relatively recently. The first Android advert I saw in Australia was Q3 last year. Whereas IPhone has been apparantly changing everything again (and again) very loudly for the past 3 years (minimum) according to Telstra.
In half that time, the house went from 3 HTC WM6 phones, to 2 WM/1 Iphone, 1 WM/2 Iphones, 1 Android/2 Iphones and very recently 2 Android/2 Iphones (with high potential for 3 Android/2 Iphones once the contract on a Nokia dumb phone ends).
I’ll be very interested to see numbers for Australia since IPhone has been multi-carrier here from the beginning.
Just to make things a little clearer… by “the house” I mean “the housemates i live with”. None of them (bar me) would be considered tech savvy and the two iphone users are probably the most savvy of the bunch. So my personal anecdote is that Android is appealing to the masses more than iPhone with iPhone taking the “upper middle”(?) class of tech users.
Of course i wouldn’t claim that anecdote is worth the paper its written on.
And to the extent that Android suffers from fragmentation and preinstalled crapware, that’s also an artifact of carrier lock-in. In an ideal world, the carriers would be dumb pipes, the iPhone would be available everywhere, and all Android phones would be like the Nexus One and S. As it is, both Apple and Google have to fight with one arm tied behind their backs.
@Jeff Read:
Maybe you missed it when I wrote this, although you were active in that thread:
So, yes, Apple on Verizon will be good for both Apple and Verizon, and might hurt AT&T a bit, but doesn’t really matter much to Android adoption.
BTW, I think that article is completely wrong — I think that starting off on a single carrier gave Apple the kind of control that they need for Jobs to work his spin magic correctly. There is no way they would have been able to keep the carriers from loading the iPhone up with crap if they had attempted to roll it out to multiple networks simultaneously. The control gained from the exclusive was gold from Apple’s perspective.
Also, when I wrote that, there had been speculation that the reason AT&T was urging people to upgrade early was because of the Verizon rollout. This speculation has gotten a lot more traction lately, and it makes sense — AT&T was letting people upgrade to iPhone 4 for $19 or whatever, because it locks them up for another two years.
Apple will sell lots of phones and make lots of money, but that’s not where the volume play is. Samsung sells more total phones in a month than Apple sells in a year.
iphone games outsells android ones in quality and game experience, android will be more techy and for developing counties
Note that there’s going to be a minimum of a 1-2 year lag between popularity and A rating game company response. The reason? A non-trivial game will generally take 1-2 years to design, build and polish to the point of any form of quality.