MattMaroon.com

When I heard that Amazon was building an app store for Android I was skeptical. I don’t know why, I’m about as big an Amazon fanboy as has ever existed. I’ve been buying everything I can from them (which today includes even basic toiletries and food) since probably 1999. I signed up for Prime shortly after it was introduced and have had it ever since. I love my Kindle so much that I’ve thought about arranging a wedding with it in Connecticut, the only state where marriage between a man and a gadget is legal.

Still, something about the idea of a secondary app store that you had to install through the primary app store (or, as it turns out, an even more confusing channel) just sounded a little too goofy to be true. You also have to allow side-loading of apps, which while not a problem for people like me isn’t something 95% of users have done. AT&T even blocks it entirely on their devices.

And, let’s be honest, Amazon doesn’t have a track record of making good-looking, highly functional products. Their website is still an eyesore to this day, even though it’s come a long way in the last few years. It’s got a lot of functional problems too. It often recommends to me things virtually identical to something I just bought. It has a link on the side to filter a search to only items that are eligible for Prime, but when you click it still many items that are ineligible remain. You shop at Amazon because it’s cheaper and easier than going to the store, more reliable than finding things from various merchants through Google Shopping, and has excellent customer service. You don’t shop there because it’s good looking, but good looking sells mobile apps.

Last week it launched and I have to say I was dead wrong. Amazon knocked this one out of the park. It had never occurred to me just how much better than Google’s app market Amazon’s could be. I think Amazon’s is even better than Apple’s. It’s the best looking and most usable Amazon product I’ve ever seen by far.

I’ve written before on my company’s blog about why the Android Market sucks. I don’t really believe that it’s going to improve very much. I think Google has a culture of organizational arrogance and while they admit they could be doing a better job on the app market, they don’t realize how much better and they have no idea why. They think their store is an 8 out of 10 and needs to be a 10, when really it’s a 2. I could go on about this for hours, but there’s no sense flogging a dead horse.

The Amazon market is a clear winner for Android’s many constituents. First there are customers. When I installed the Amazon App Store, which has a small fraction of the apps the Google App Market does, I was immediately struck by how many high production-value games there were. I never knew it because I never saw any of them in the Google store. That store just has the same crappy tower defense game, Fruit Ninja, Angry Birds, and Paper Toss day in and day out.

Carriers too stand to gain. From my conversation with a Google employee at GDC, I’m pretty sure that carriers are getting a chunk of the revenue from app markets on all smartphone OSes, including iOS. A better market means more revenue for them plain and simple, both from apps and handset sales. I expect to see Amazon make long-term deals with carriers to get OEMs to install their market by default on devices.

OEMs will benefit by gaining freedom. While Android is an open source OS, the most important app on it, the App Market, is not. It’s proprietary and owned by Google. The App Market has so far been Google’s method of controlling OEMs. For instance in future versions of Android, Google is believed to be mandating that OEMs use the native Android UI by default. Motorola, Samsung, and HTC all have their own UIs right now that they might not be wanting to part with since Android’s stock one is so poor.

With a viable second app store Google loses a good portion of its hold on OEMs. A smartphone OS without an app market is worthless. But if Amazon’s becomes a viable competitor (and it probably is already) then OEMs can tell Google where to stick their app market.

Moreover, they’re now possibly no longer even stuck with Android at all. If Amazon manages to get all of the big players to participate in their App Store, then all any upstart OS has to do is make a deal with Amazon and ensure that Android apps can run on their ecosystem. The RIM Playbook is proving that this is quite possible technically. You’ll never get Google Market on a Blackberry OS (not without lawsuits flying) but you might get the Amazon App Store full of Android apps.

The biggest winners, though, will be developers. Right now despite the fact that Android is a larger ecosystem than iOS, the poor App Market has held it back from seeing much development. iOS apps are simply outselling Android by an order of magnitude. I saw an article last week about something like 8 apps that have made over a million dollars on Android. There will probably be 8 apps that pull in that much revenue today on iOS.

Even Google wins, though they won’t see it that way. If they’re getting little to no revenue from the App Market themselves, which I suspect is the case, they’d be far better off letting a third party handle it. They’re better off having a vibrant app ecosystem on Android coming from a market they don’t control than a crappy ecosystem from one they do. At the end of the day its really apps that sell these phones. Apple’s slogan isn’t “there’s probably a website that could accomplish that” for a reason.

I’ll go ahead and make the bold claim that the Amazon App Store will be the primary driver of app sales on Android within a year. This is the most important thing to happen to smartphones since the original App Store on iOS.

The tech blogs are all atwitter about Nokia’s move to Windows Phone 7, and the response is overwhelmingly negative. Personally I think they’re all insane. Nokia’s Stephen Elop has sent a bold signal that he’s not going to sit there and shuffle deck chairs on the Hindenburg while Android and Apple eat the rest of their large but rapidly declining market share in smart phones.

Now, I’m not 100% sure Nokia made the right call going with Windows. Android might have been a better alternative. I can see why they’d go with Windows Phone though. Microsoft is the devil you know. They’re honest about their intentions. They want to make money selling you their operating system and selling apps on it. I’d bet they’re even sharing the latter with Nokia. Their interests are aligned with an OEM’s just as they have been in the PC market for decades now.

Google’s somewhat sneaky about the whole thing. They’ll give you the “open source” operating system, but all the good parts of it are closed source and owned by Google. It may be free as in beer, but it isn’t free as in speech, and it isn’t going to get any freer over time either. If you want access to the Android app market (and if you don’t have that, good luck selling units) you have to play by their increasingly stringent rules, which soon will include using Android’s stock UI.  

Moreover Windows Phone 7 is marketable, Android isn’t. Windows is a brand that, no matter how often maligned it may be by people who read tech blogs is still trusted by most people. Android is a commodity. I’ve asked just about everyone I’ve encountered with a smart phone what operating system they’re on. Most of the Android people say “I don’t know.” Windows and iOS don’t have that problem, and if Nokia wants to create value they need to avoid being a commodity.

As unrecognizable as Android is to customers, Symbian is far worse. Samsung, Motorola, LG, and Sony-Ericsson all moved on because of it. Its web browser sucks. Despite having the largest share in the market its app selection is anemic. It is, as Elop said, a burning platform.

A lot of the nutjobs complaining about “Elopocalypse” say Nokia should have just improved Symbian. But then they’d be looking at, at best, 1-2 years before they had a viable competitor to Android and iOS, and another year at least before the apps arrive, if they ever do at all. The smart phone OS market is going to be won in less time than that. Nokia doesn’t have time to wait for Mr. Right (if you even accept the assumption that there’s significant value in owning the software, which itself is ludicrous) because they need Mr. Right now.

It takes a bold leader to ditch a platform with the highest share in its market. There’s a fine line between courage and stupidity, but I think  they’re on the right side of it. They could spend lots of time and money developing an OS that can compete with what’s out there now, then lots more time and money trying to attract developers to it. And if the first iPhone had just launched this year that might be the way to go. But it didn’t, so it isn’t, and I think Nokia should be commended for having the courage to make a bold play to recoup what they’ve lost.

The iPhone is launching on Verizon, and the mobile industry is all atwitter about it. As an iOS developer myself I’m certainly happy to see it. It’s clear that this will expand Apple’s (and therefore my) audience to some extent, and that’s nothing but positive. It’ll be a good thing for everyone except AT&T, and for them I don’t think it will be disastrous.

As I said before the iPhone launched, people largely choose their carrier first and their phone second. While the iPhone has done a better job of getting people to jump ship than any other mobile device, that paradigm still hasn’t really changed.

What this means is that Apple is about to shift a good chunk of units on Verizon. Surveys seem to indicate about 20% of AT&T customers will switch with their next upgrade, which won’t help AT&T with their perennial war for the largest carrier in the US. It also means we could expect to see Apple as much as double their US sales of iPhones in 2011.

But what it also means is that this won’t really be a game changing event. Android is now activating somewhere around three times as many units per day as Apple, and even if 100% of Verizon iPhone owners choose it over Android, it won’t put much of a dent in that. Apple only sells something like 20% of their units in the US anyway, which means even if iPhone sales here double, they’re looking at an overall 20% growth.

Furthermore, Android is now becoming a serious competitor to iOS in the tablet space. The iPhone had no serious competition in the consumer smart phone space for almost three years, the iPad will have it in less than one.

What does all that mean for app developers? Right now not much. The Android app store experience is so poor that even a 3-5x multiple of users still doesn’t make it as compelling of a development environment for someone developing standalone apps. It’s perhaps more compelling for those making apps that they will use as a continuation of their brand (think online banking) but for a developer making an app for the sole purpose of profiting from it directly, the iOS platform is still looking substantially better.

Last year my company raised a Series A funding round. When a company raises money and issues securities, it is required to fill out a form with the SEC called Form D. Form D filings are made public, and intrepid bloggers often use them to confirm funding rounds.

Ours was picked up somehow (perhaps automatically) by TechCrunch and posted on CrunchBase. Our lawyers, who had been using my cell phone to contact me for 3 years, used that number in the Form D. I’d looked over the form in advance and thought nothing of it. There’s no way I could have guessed what would happen.

I quickly discovered that salesman of software and hardware monitor Crunchbase and call everyone who raises money. I started getting multiple calls a day from reps who worked for companies like Salesforce, Dell and CDW. One particularly obnoxious one, Rad Game Tools, still calls me frequently asking for my cofounder.

I had no idea what triggered the sudden deluge, in fact because I’d gotten them on occasion before (who knows how they’d gotten my number) I didn’t even notice for a couple days that something was up. When you’re busy it doesn’t immediately occur to you that something that used to happen once or twice a month is now happening 5 to 10 times a day.

Then I started getting calls from irate Starfleet Commander customers. Starfleet is a very competitive game, and customers’ blood can sometimes run more than a little hot as a result. Customers can spend months (or even years) building up a fleet and then lose it in just one moment of carelessness. Many times when something like this happens, the customer contacts our support to try to convince us to give them their stuff back, which we don’t. Suddenly those people were calling me.

Curious why, I did a little search and found that Google had indexed CrunchBase. Now when someone went searching for our customer support number (which we don’t have, everything is via the net) they saw my cell phone number right there in the result. They didn’t even have to click through!

Between the salesmen and the customers I was quickly getting bombarded. I looked at my phone one night and saw over 50 missed calls, not one of them from a person I knew. I was on the verge of switching my number as I had done once before when it turned out that its previous owner didn’t like to pay his bills. (I not only got calls from debt collectors he owed, but even got hassled at a Wal-Mart when getting an oil change because he owed them too.)

But changing your number is a pain for everyone involved. So I decided to look for a better way, and I found one pretty quickly. I was using WebOS at the time, and it turned out that if you had your phone in developer mode you could, by checking a box in WebOS Quick Install, set up a tweak to block all callers not in your contact list.

A while later I switched to Android and there I discovered a couple apps. The one I’ve been using is made by EasyFilter. It’s missing a whitelist (I’d love to allow all calls from my area code) but other than that it does pretty much everything I want which, really, is just to block every contact number that isn’t in my list from calling or texting.

Unfortunately that’s only step one. While most of the callers will give up as soon as they’re sent to voicemail, enough won’t that you’ll get annoyed. (Well, at least if your number was publicly connected to raising a 7 digit sum.) So the next step is voicemail filtering.

For that I originally used Google Voice. It’s a free service that allows you to have all your voicemails sent to it. You can do that easily, every cell carrier has a number you can dial that will enable you to send your voicemails to a third party service. Why that exists I have no idea, but it’s fully automated and quite simple to use.

Using a third party service like Google Voice you can simply spam filter unwanted calls just like email. Just flag them as spam and future calls will go straight to the junk filter. This is especially nice because the people who leave voicemails tend to be salesman and do so repeatedly. You rarely get someone who does it once.

I’ve since switched to YouMail. Google Voice’s text transcriptions are comically bad, but for a small fee YouMail will give you human edited ones that, while far from perfect, are usually at least good enough for you to figure out who was saying what. Google’s text transcriptions are so far off that they’re largely worthless.

(I’m thinking about switching back to Google because though I’ll miss the text, I’m finding YouMail sometimes takes over an hour to notify me of my voicemails. That may be slightly more annoying than transcriptions are helpful.)

Either way, this two step solution has saved me from having to get a new number, and if you’re struggling with unwanted calls and have a modern smart phone it can work for you. (Unless you have an iPhone, since Apple almost certainly would reject such an app, though even then if you’re willing to jailbreak I wouldn’t be surprised if one exists.) It’s mildly annoying at times, when someone new is calling me they simply have to leave a voicemail once so I can add them to my contacts or I’ll never hear from them. But unlike email, where this sort of approach would fail miserably, I can block out everyone who isn’t a contact and miss very little.

Fred Wilson wrote, a few days back:

I saw two HTML5 apps yesterday. One running in my Android browser. The other running in the iPad browser. They looked and worked exactly like their mobile app counterparts. It was a mind opening moment.

…

I’ve always disliked the idea that we have to download apps on our phones when the apps we use on the web are loaded in the browser on demand. But I’ve accepted the mobile app paradigm as something we will be living with for the next five years.

I’m not sure it’s five years anymore.

As an app developer I hope he is correct but I think he’s perhaps misunderstanding why native apps are so appealing. As far as I see it native applications have three primary benefits: technology, distribution, and monetization.

As far as technology goes, Fred may be right. The number of things you can do in HTML5 will continue to increase. HTML5’s geolocation functionality (already supported on many devices) takes care of one of the big ones.

Unfortunately the one thing you can’t make in HTML5 right now are rich games, which is exactly what people are using. You can make pong, but you can’t make Angry Birds. If and when WebGL is ready for prime time the number of apps you’ll actually need to make native will decrease significantly. I’m always hesitant to put a timeline on standards being finalized and adopted (remember the 802.11n debacle) but 5 years might be realistic for usable WebGL on most major smart phones. I wouldn’t bet on much sooner though, and as long as games are driving what people use the platforms for, which you can always safely bet on, native apps aren’t going to disappear.

However I’m far less hopeful that the other two advantages will disappear in five years. Right now Apple has a huge lead as far as monetization goes, even over Android. The iOS advantage comes from the network of credit cards Apple has stored on their servers. Almost everyone who owns an iOS device has a payment method already in Apple’s database that enables 1-click purchasing. Nothing quite like it exists on the internet in terms of ease of use. PayPal is possibly a viable competitor for Android, Google Checkout seemingly has not been.

And then there’s distribution. Right now a top free app in the app store gets hundreds of thousands of downloads a day. Multiple apps achieve this milestone every week. It’s extremely hard to do this on the web, and it would probably be even harder on the mobile web. There’s no one central place where people can see all of the cool new websites every day, sorted by category, and ranked and reviewed by other users. The App Store is the Amazon of native apps, and there’s just nothing like that for web apps.

Web apps have other distribution methods, of course, that are a mixed bag. Search, recommendations, Facebook/Twitter connect, etc., offer unique advantages and disadvantages over the centralized app store model. But I can’t think of one mobile web app that’s achieved mass market success yet the old-fashioned way. Most mobile web apps are simply extensions of already existing desktop web apps. If anyone is having any success yet trying to change this I have somehow missed hearing about it. That doesn’t mean it won’t happen either, but it’s going to be a long hard slog from here to a world where mobile web apps can compete with native ones for users.

If anything what we’re seeing right now is the opposite of what Fred is predicting, with web apps going native. Apple’s new Mac App Store is trying very hard to push things in the opposite direction. Magazines like Time, Wired, The Economist, and new ones like Project are launching iPad apps that could easily, from a technological perspective have been web apps. They’re doing it for the same advantages of monetization and distribution that native apps enjoy. I wouldn’t be surprised at all to see Microsoft do something similar with Windows soon, and if they don’t OEMs will.

As long as Apple remains an influential player, they’ll keep abusing and improving those advantages, especially the last two, and maybe even restricting the technology available to HTML5 to keep control over what’s running on their devices. All Apple has to do to ensure that the majority of apps remain native is not support WebGL and/or Flash.

Apple has strong financial incentives to see a healthy app store. It monetizes directly and it also drives hardware sales. Try as hard as you might, you can’t hammer “there’s an HTML5 app on the web that runs on any phone including ours for that” into a catchy slogan.

One thing I always hear when reading about mobile development is fragmentation. Steve Jobs, who talks about Android so much because he isn’t worried about it, harps on it relentlessly. Just today I read, on an Android blog, the following quote:

In an iOS world, you only have to write code once and know it is optimized for every phone that’s been sold.

Bahahahahaha. Yeah right. I do iOS development, and let me tell you, fragmentation there is a huge problem. First off, there are different OSes on different devices. Here’s a chart of that:

Compare that to Android:

Pretty similar. Android in this case has the benefit of having fewer OSes in play.

Our newest app (not yet released) uses Gamecenter, which runs on iOS 4.1 or later only. Which means that when we move to the iPhone we’ll have to do two apps or simply be unavailable to half of the population. Even on the iPad, where we are developing now, it remains to be seen how many people will upgrade to 4.2.

If I make a website and buy ads for it on Google, I don’t have to filter my users out by operating system. If I buy ads for my iPhone app I’m doing it somewhere that will let me choose to show my ads only to people on 4.1 or better.

Then there’s  the hardware. Good luck developing anything for any iPhone before the 3Gs. I can’t find how many of those are left on the market (thankfully probably not too many, at least in the US) but if you want to support them you’re going to be adding a decent amount of extra work. Same for the first couple generations of iPod Touch which, by the way, don’t get upgraded nearly as quickly as phones.

And then there’s the display. We’re making a universal binary of our game that we want to work on the iPad, the iPhone 3GS, and of course the iPhone 4. As a result we have to do a bunch of extra work to get things to display properly on both handsets.

There’s plenty of fragmentation on iOS. It might not be as bad as Android. It just depends on what you’re doing. For some apps it won’t matter much at all on either platform. For some it will be impossible to support all devices. For many it will fall somewhere in between.

The post on the Android blog I mentioned earlier was inspired by Angry Birds, which has had some performance issues on older Android devices. But a little Googling shows it has issues on even the iPhone 3G.

Mobile devices will always have fragmentation because they’re evolving so quickly. They are right where PCs were 10 years ago, where hardware is improving quickly, developers are finding new ways to take advantage of it, and as a result users are upgrading frequently.

Nowadays you buy a PC and it lasts forever. I have a laptop that’s about 4 years old and it still feels more than good enough. But I’ll upgrade my phone every year or so for as long as Sprint lets me.

As a developer fragmentation isn’t a deal-breaker, it’s just another line in your cost-benefit analysis. And odds are it isn’t nearly as important as unit sales (where Android is now crushing) or ability to monetize users (where iOS has a healthy lead, though I suspect Android will close the gap significantly with PayPal) or distribution (also in iOS’s favor at the moment for most apps, though not all).

In the end developers will go where they earn the most money, and betting on fragmentation to stop that from being your opponent’s platform you’re making a big mistake.

Mark Zuckerberg said yesterday, in Facebook’s mobile announcement, that the iPad is not mobile. For some reason this has been controversial, but I think he’s right in principle. Clearly semantically he’s incorrect, but then a desktop computer is mobile too because I can pick it up and carry it, so the semantic argument is useless in the context of mobile app development.

The real question is how do people use their iPad, and it’s important to keep this in mind when developing for it. You don’t develop the same things for a device that is used largely in the living room that you do for a device that people carry with them 24/7. Of course there’s a lot of crossover, just as people use Facebook both on their home PC and their cell phone, but there are differences too.

It’s a mistake to think of the iPad as just a larger iPhone, it’s not. For one, it is substantially less mobile. You don’t carry it with you everywhere you go. You won’t use it, for instance, to occupy time spent riding in a cab, in the line at the supermarket, on the john at work, etc. Since iPad sales are largely at the low-end, Wi-Fi only model, it probably spends a relatively decent amount of time offline entirely, whereas an iPhone rarely is.

And all this isn’t to say, of course, that people won’t use Facebook on the iPad. They’ll use Facebook on their napkins if someone makes one with a Wi-Fi connection. But he’s right that tablets have to be treated much differently than phones. What people will do on Facebook on the iPad will substantially differ from what people will do on their mobile.

If I had to guess, I’d say the mobile experience will lean heavily toward pictures, status updates, and things you do on the go. It will be more about content creation. The iPad, on the other hand, is a content consumption machine. It will be more about browsing profiles, looking at others’ pictures, etc.

It’s not a mistake for Facebook to attack tablets differently than mobile devices, even if tablets are technically mobile.

I bought my Palm Pre on launch day, in June of 2009. I remember the day well, since my wife and I were going to Detroit to watch game 5 of the Stanley Cup finals at the Joe Louis Arena. I got up early, called around, found the one Sprint store in the area that had a few units left, and rushed up there to pick it up. I set it to charge on the drive and played with it incessantly for the rest of the day.

At the time, I had already written my prediction that Android would become the #1 selling smartphone operating system, which just recently came true, but at that point it was still too rough around the edges. After using WebOS a bit, you start to feel like even iOS is unpolished. WebOS is just that much better than everything else out there.

The Pre has served me well over that time, nearly a year and a half. Unfortunately for Palm it didn’t serve them quite as well. Their team built the best mobile operating system of all time, then struggled to cash in on it. They aired one bad commercial after another, didn’t update their hardware in a timely fashion, didn’t roll out their developer program quickly enough, waited far too long to get onto Verizon (by which time they launched their first big Android phone, the Droid) and didn’t respond well enough to a couple build quality problems.

They chose Sprint as a launch partner, rather than Verizon, which was dumb for them. In hindsight I’m glad they did. I switched from Verizon to Sprint to get the Pre, and I’m now dumbfounded that everyone isn’t on Sprint. Their coverage is best in class, especially since you can roam (even for data) on Verizon’s network. Their prices are substantially lower than Verizon’s. They’ve now got arguably the two best Android units on the market in the Evo and the Epic 4g. Their customer service is pretty good, at least at the stores by my house. They’ve replaced two Pres for me at no charge (one of which they really were not obligated too). I’m not surprised that their network is finally increasing in popularity again, they’ve just executed phenomenally well over the last couple years. I’m probably going to pick up a few shares of S this year given what I’ve seen.

And now, with the Pre 2 launching first on Verizon, I’m giving up on Palm. Don’t get me wrong, the new Pre looks good. WebOS 2.0 looks great, in fact it seems they’ve extended their lead over the other major OSes in terms of both usability and functionality. They fixed two of my three main hardware complaints, by going to a glass screen and fixing the USB door. Having double the processing horsepower and a lot more RAM looks great too. And you have no idea how much I will miss the Touchstone, unless you have one, in which case you probably have a hard time imagining having to plug a phone in like I do. Had Palm improved the keyboard (my one major complaint with the hardware, and probably the biggest) and made a 4g model I might have hung in there another 6 months.

But they didn’t, and it’s time for a change. Besides, I have lots of experience with iOS, obviously also with WebOS, and am, amongst people I know, pretty much the only who can really compare the two. I’d like to be able to say the same about Android. Most people you hear talk about the merits of one or the other OS or phone have only a passing familiarity with the one they don’t own, and swear by the one they do. I’d like to be able to actually experience all three as a user. And if anything I’m giving iOS a favorable bias since I’m using it on the iPad and iPod Touch and not as a phone, which has always been its Achilles heel.

I haven’t decided which model to get yet. I’m thinking the Epic 4g. The screen on that bad boy is fantastic. The keyboard is spacious, and it’s shockingly thin and light for a slider keyboard. It’s got 4g data, which I’ve heard has significantly lower latency (the true problem with 3g data) and should be rolled out in my area this year. The Evo looks great too, though I’m still not ready to go without a keyboard, even with Swype. I’ve used the keyboard on the new iPod Touch and iPad, which have to be at least as good as Android’s, and I’m just constantly frustrated by it.

So we’ll see. I’ll probably get one in the next month or two, and will of course post thoughts.