Android vs. iPhone: How Google is winning hearts of developers | Hacker News Reader

Android vs. iPhone: How Google is winning hearts of developers

techflash.com

83 pointscwan75 comments

I wrote a blog post on a related topic last week.

Here is some more relevant data The folks at http://larvalabs.com/blog/android/android-market-payouts-tot... say that only 13 Android apps have got more than 50,000 downloads. A greater number of iPhone apps have got more than a million downloads.

This should give you a good indication of two things

(a) which platform successful developers focus on and

(b) how many Android users vote with their wallet to indicate that the Android market has good apps.

Edit: It is interesting to see the downvotes canceling out some of the upvotes :) I understand that some people may not like these numbers, but the commenter/reply suggesting that the numbers are "misleading" (because money-making iPhone apps are older) is mistaken.

Consider the fact that the iPhone Angry Birds app has got more than 4 million downloads in less than 4 months. A paid iPhone app in the #1 slot for a month can get more than a million downloads in that month. Now think about how many downloads a #1 or #10 paid Android app can get in one month.

has the Android market opened up paid apps outside the U.S. ?

The Market itself is open to lots of countries, but many of these countries don't have support for paid apps yet (only free) which is very frustrating.

What's many? At least in germany it's open for paid apps for a year or more. Didn't think there are still countries.. that's frustrating indeed. At least there are other markets (slideme)

Australia, Denmark, Russia, etc.

Actually I seem to remember that the only countries in Europe who are allowed to play are Germany, France and the UK.

It is being frustrating because it could be fixed tomorrow - the issue is that Google is using its own payment system (google checkout) instead of a real payment system such as paypal.

If you ask me, it borders on what Microsoft did with IE and netscape in the nineteens.

Mhh, according to Wikipedia buying apps is allowed in a lot more countries: http://en.wikipedia.org/wiki/Android_Market

Maybe, but as a sell that isn't going to do me any damn good.

Isn't this kind of misleading? Haven't the iPhone apps been on the market longer and had more time to gain more downloads?

50x less is kind of crazy though. There are nearly enough android devices out there, but you'll make 50x less money if you're an android developer. So an iPhone developer making $100,000 from Apple would only end up making around $2000 from Google... seems a bit scary to invest effort into a platform that won't return value!

For those not likely destined for the top 10 and that don't have large investments to recoup, these top 10 statistics are fairly irrelevant. A survey among developers who ported their app to both platforms would be far more enlightening.

For a company that spends a lot on development and marketing, and will likely have a top 10 app, those top 10 numbers may certainly mean that their Android investment won't pay off.

An interesting datapoint is this guy:

http://www.reddit.com/r/IAmA/comments/caq8f/iama_developer_w...

Who made the same app for the iPhone and for Android. His ratio was 3:1, which is roughly the ratio of handsets in the US iirc.

I'm not sure why you think the top 10 disparity between iPhone and Android is irrelevant and why that wouldn't have any bearing on the average app's revenue prospects.

In any case, the TechFlash article says "Michael Schneider of HiveBrain Inc., a Seattle-based iPhone app developer who has dabbled in Android apps, said the volume of sales for Android apps has been about one-tenth the size of their iPhone counterparts."

This ratio matches what I've heard from several other people (disclaimer: I know Michael and among other things, we have discussed these topics in the past)

> The folks at http://larvalabs.com/blog/android/android-market-payouts-tot.... say that only 13 Android apps have got more than 50,000 downloads

It's important to note that this number counts only paid games. Is it really that few?

I'd love to see a comparison of ad-supported free applications. iPhone will surely win, but nowhere by that margin.

Also, one of the points of the article is that Android allows for more varied apps to be developed. GOTO (as described in the article) cannot exist on the iPhone. Nor can Swype. Nor any of the the home screen replacements.

Indeed about half of the app ideas I have are not possible to implement on non-jailbroken iPhones.

Is direct comparisons of download numbers the correct measure? The iPhone has nearly three times as much users as of Android and a greater lure of larger returns.

Correction: How Google is trying to win the hearts of developers.

Currently they only deliver a fraction of users to developers compared to the App Store, their "Market" has suffered from recent outages and glitches, they're struggling to cope with device and OS fragmentation, they force developers to remit their own taxes, and there are some glaring omissions in the countries and regions they support.

I think they need to try harder.

Also, their Marketplace is limited to just a couple of countries (which does not include India). I would love to jump into Android development, but then I'll have to manage payments myself.

The Android Marketplace cannot possibly compete with the AppStore. Not yet.

ORLY.

I'm actually building an Android app now instead of iPhone app. Here are the rationals: 1. I don't have an iPhone nor an Android phone so there's no inclination from the device point of view, 2. My last mac was 10 years old and can't be used for IPhone development while Android app can be developed on Windows/Linux machine. 3. The low downloads and low app numbers on Android side indicates that it's a virgin market while the device numbers are huge.

We'll see how it goes.

The big problem I have with Android apps is just how terrible the GUIs are. They range from mediocre to unusable. For example:

K-9 Mail on Android: http://tiny.cc/l965n

Mail on iPhone: http://tiny.cc/ss4e9

HTC Mail on Android: http://tiny.cc/b2f6m

I love Android and use a Droid as my phone but I totally agree with you, theu really need to step up with their UI.

It seems that they know this. They hired the lead designer for WebOS a while back and according to TechCrunch the next release after Froyo is all UI stuff: http://techcrunch.com/2010/06/16/android-team-laser-focused-...

One can write good looking apps for Android. It's just not as easy as it is for iPhone. Having done both myself: on iPhone you can just throw together a bunch of default widgets to get the job done and your app will look pretty good. Doing the same on android.. not so much.

However, the latest apps out of Google (IO Schedule, Twitter) are using some of android's new UI patters, and they look good. Hopefully these new designs will be reflected more in the widgets shipped with 2.3.

I heard that the gmail client was much better than the android general mail client. How do you feel about it?

(btw mail for iOS is not a very good mail client either, though it looks pretty nice and has been improved a bit in iOS4)

That's true. Hopefully I have only gmail account that I use frequently, I also have me/mac.com one, but I rarely use it.

Google shines in the web also with gmail.com, while somehow Mac's me.com feels slow, also their Mail client is not that good either, but the rest of the Apps are top-notch.

And Mail for iPhone rocks! Too bad I don't use my iPhone as phone, just for developing (T-Mobile customer with G1).

The mail application on Android feels very clunky, but I think the gmail application on both Android and iPhone are pretty comparable in terms of features/flow.

The GMail client is OK but it's lacking in the feature department. No option for a 2-3 line preview of the message body, when composing a message you can't pick contacts from your GMail contacts or local address book -- you have to type in the addresses each time, as far as I can tell it doesn't support copy & paste.

No multiline is a fair point. Message composition auto-fills the recipient when you start typing - I can get to my intended recipient in 1-2 taps, usually. I prefer it that way, though I can see how others might now. You can, though, just tap a contact's email to start a new email to them. It certainly does support copy-paste, though. I'm on 2.2, so that may be part of it, perhaps?

Every input field supports "select all", cut, copy & paste at least since 1.5.

Doesn't work on 2.1 Seems to be a known issue:

http://bit.ly/bQUdvU

The general client doesn't do search of any kind, not with IMAP at least. Pretty sad for the number one search company in the world.

A huge issue with the gmail client is that you cant send-mail-as. I have my school account hooked up to gmail using my schools SMTP server but I cant't send mail as that email address from the phone.

If any Google engineers are reading this. Please fix this. This is a huge issue.

Btw this double post wasn't intentional.

Might as well throw the normal GMail app into the mix: http://smarterware.org/wp-content/uploads/2010/01/android21-... (via http://smarterware.org/4308/android-2-1s-best-features-in-sc...).

It seems odd that GMail's interface is so horrible to use when Google Reader's is awesome. I actually like the mobile version of Reader better than the desktop version, currently. I wonder if they're made by the same team...

I should add that I mean the mobile versions of both. GMail's desktop interface is great.

I'm using K-9 on my HTC Desire and Apple mail on my iPad and I like both. Note that the GUI of my K-9 application is different than the one you show. Keep in mind that they may iterate, which BTW is apparently more easy on Andoid than on AppleStore.

Anyway, generalizing on your personal experience and taste is not very instructive.

As I live in Canada, I don't even have the option of selling apps via the Android Market.

Yea, wtf! And there's basically no word on wether they're even working on in.

They are not.

But it is a general problem with Google, they don't give a shit about you if you aren't American (take google voice for example, they also only work in the US).

In Hong Kong, we don't even have the option to BUY apps in Android Market.

The cutting edge apps will be developed for the iPhone first, and then ported to Android (if it's even possible).

Reasons behind my claim:

Device fragmentation: If an app makes heavy use of a built-in sensor (eg. accelerometer), it's much harder to test the app with all Android devices. Some of the Android devices might not even have the sensor, that the app uses, which will result in an angry customer. And angry customers are bad news, when you aim for word-of-mouth marketing.

Low-level coding: With the iPhone, it's possible to implement cpu intensive code blocks in assembly optimized for each generation iPhone, and thus optimize the code, so that you make the most of your resources. If I'm not mistaken Android only supports Java.

Android has supported C/asm for a while, but your other points stand.

>Some of the Android devices might not even have the sensor, that the app uses

Same goes for iPhone (e.g. the new gyro, the front facing camera).

Don’t forget the iPod touch. I’m really annoyed whenever some app doesn’t check whether there is a camera and just displays a camera button, no matter what (prompting the standard “no camera” dialog).

Apple will generally not approve an app that does this if they notice

Fragmentation: iPad, iPhone 4, iPhone 3GS, iPhone 3G, iPod Touch 2G, iPod Touch... Each with different capabilities and hardware specs. I've shipped titles to the App Store where I had to tip toe through the minefield of all those permutations, making sure the app worked correctly and/or similarly on all. Sometimes pretty hard to do.

Android more fragmented? Surely.

Android is more fragmented in a very fundamental way. If one ignores the iPad and just look at the phones they all have basically the same physical screen size. Android phones come I sizes from 2.5" 320x240 up to 5" 480x800. Designing a UI that works well at all those sizes is a challenge far beyond the more superficial differences of the iPhones.

Device fragmentation: There's certainly fragmentation between generations of the iPhone, too. Gyroscope, front-facing camera (or any camera at all), different amounts of RAM, different processors, different screen sizes, etc. It's a problem that has to be dealt with on both platforms.

Low-level work: http://developer.android.com/sdk/ndk/index.html

If your app requires a particular hardware feature, you can list that in the app's manifest file, and it won't be installable (or even visible in the Market) on devices that don't have that hardware feature.

And the Android NDK lets you write native code for the stuff that needs it.

So one problem I have with developing for the Android platform, is that I don't know how to go about it and I don't understand the different Android versions and which phones have which version of the OS and how many people have each version.

Starting development for the iPhone was really simple, I went to one place on Apples site, downloaded xcode and started building.

While I don't think it will have hard to figure out all the answers to my questions once I get started, there are a lot of unknowns for me in terms of the Android platform. My main point is this, it is much more clear how to get started building for the iPhone, which I think gives Apple a large advantage.

http://developer.android.com/index.html should get you started. It even has a link to "Target Devices" that shows you what percentage of active devices run each version of the Android OS.

For me, I just develop for the lowest common version, 1.5. Don't need the latest feature in my app. The important thing is to figure out what's interesting and useful for the end user, not the latest and greatest OS feature.

Cool everyone, thanks for the info. This makes sense that the software would be forward compatible (I guess I should have assumed this, but I didn't even know this was the case).

The only significant thing that we need are push notifications; in which version did push notifications start, did Google change how they were done?

Maybe this site helps with the fragmentation issues:

http://androidfragmentation.com/

It's a strange title... It doesn't seem to be much about Google winning the hearts of anyone, even though this particular developer chose Android because he was familiar with Java from before. Seems more like brain-winning to me.

It seems like a moderately good piece on someone who has chosen to develop for Android though, even though I would have liked to see the conclusion (that the choice to develop for Android was looking good) supported by some kind of metric, I didn't get any feel for the success of his company so far.

In the places where Google Checkout supports seller accounts, perhaps. The rest of us are mostly giving Android development a miss because we can't sell apps.

This seems like a serious shortcoming to me. I can't sell from Canada, for example. Presumably this will improve with time.

I'm also really scared of dealing with Google's customer service if anything goes wrong with my merchant account.

This must be some sort of joke... Google certainly hasn't won my heart. Working with Eclipse and their SDK makes me want to throw my computer out the window, along with their phone.

The title of the article was "Google winning the hearts of developers" not "Google winning the heart of thinkcomp from Hacker News".

Well, perhaps this was my error, but I considered myself a developer as someone who had written Java code using the Android SDK for later deployment on Android devices.

Who says you have to use Eclipse? Just use your favorite editor, and run "adb install" to install it to the device or emulator for testing.

You can use any IDE you want (I use IntelliJ IDEA), and you do not have to use Java either. You can develop in Scala, or via the NDK in about any language you want. Then you can do a lot without using the SDK. But AFAIK you need a small GUI-wrapper around you native code to interact with the user.

After using the iphone for 2 years, I decided to try the nexus one Google gave us at googleio. What I really like about the nexus one is the seamless integration with google apps, gmail, running gtalk in the background etc., but thats it. Most of the apps I have downloaded really sucks, their UI is terrible and they crash often. The multitouch response on google maps is not that good,the pinch doesn't respond that well. Google needs to polish a lot of things, and we are still waiting for Froyo to come out.

The multitouch issues on the N1 are actually because the touch screen on the N1 is (when compared to the Droid) crap. Google has acknowledged this (sorta) but say they can't do anything about it. There's actually an app called "Multitouch Vis Test" that you can download and play with that easily demonstrates how easily the touch screen gets confused when you have two fingers on it.

I thought the article would be all about how Google gives developers free phones at all their conferences.

The huge quotation mark in that pull quote looks ridiculous. (Well, the small one does, too, but it's slightly less noticeable.)

A huge impediment to developing apps for Android seem to be the absolutely incredible shortcomings of Google Checkout. Most countries cannot even buy paid apps, even fewer countries can publish paid apps. That kills pretty much any incentive for many developers, especially with a new "iPhone devs make bajillions" article every other day. The saddest thing about it is that Google do not seem to be moving about this, at all. Android Apps, E-Commerce sites, etc., lots of people are waiting for worldwide Google Checkout. I don't get why Google don't get it.

Let's not forget the awesome easter eggs thrown throughout- I've never been so happy reading docs.

In Log... http://developer.android.com/reference/android/util/Log.html...

and SensorManager. http://developer.android.com/reference/android/hardware/Sens...

True. Considering that one can develop android app on any machine but iPhone SDK requires Mac OS (else they can't be published to the app store).

I suppose, Apple has lost a lot of hearts after the series of close-structured and seemingly unfair app store policies. Though the fairness could be debated and a different issue altogether.

I don't think Google is winning hearts - rather, they are losing hearts more slowly than Apple.

It doesn't matter, what matters is the consumer.

The consumers will go where the software is, in some cases. Take a look at the Sega vs. Nintendo market of the 8-bit 80s in the US - Nintendo offered a larger market, and locked up all the developers in exclusivity contracts. Even with superior hardware, Sega couldn't compete because they couldn't get many third party developers to develope for the Master System.

reads like advertising. where's the technical meat?

If Google would only buy Titanium or Phonegap to overnight make it easier to build Android apps. But they won't. Because they don't care about making better tools.

Google get rid of Google Checkout and allow users to pay via their cellphone bills for apps.

Also my assessment of Android users vs. iPhone users is the latter buys their music from iTunes, while a big portion of the former download from p2p and the alike.

Reply on news.ycombinator.com