Hacker News new | past | comments | ask | show | jobs | submit login
Poll: You and Mobile Development
90 points by acangiano on Sept 19, 2011 | hide | past | favorite | 57 comments
For which of the following mobile platforms/OS do you develop? Please select all that apply to you.
815 points
Mobile-friendly Web development
704 points
669 points
I don't do mobile development in any capacity
309 points
Windows Phone
91 points
BackBerry OS
45 points
Other mobile OS
44 points
33 points
Symbian OS
13 points

For 5 years now I've been occasionally writing more features for Windows CE based barcode scanners. I wouldn't say that qualifies as "Windows Phone", so I didn't vote for that.

Thankfully by now we are more and more replacing these Windows CE things with iphones plus sleeves with barcode scanners which gives us a nicer API and the end user gets a much cooler device to work with.

The .NET Compact Framework is a beast: It's extremely slow and still it happens more often than not that you need something, you check the documentation, you find it, only to read that it's only supported on the desktop.

And don't get me started on the development/run/debug cycle: Both the emulator and the actual device are slow and after the 200st time that crappy assistant offered me to register my barcode scanner or even emulator(!) with Microsoft, I lost all the love that might still have been left.

Oh. And the SDKs for the scanners have this distinct feeling of being software written by hardware people when I look at how "convenient" they are to use. My personal favorite: The driver that hard-crashes the device whenever you terminate the application from the debugger. Yeah. Hit that stop-button you are used to and you'll have to wait through another 3 minute reboot cycle.

You iphone and android people have it so nice </cranky>

As part of my degree I had to work in a team developing on some ancient Dell PDA that ran a crippled version of Java 1.4 so I can feel your pain. My favourite bit was having to pull out the SD card from my machine, insert it into the device, run the code then plug it back into my machine to pull my logs so I could see what just happened!

I did Brew and J2ME device development before Android. Believe me, I _feel_ your pain.

Can you point me towards a barcode scanner for iOS devices?

http://www.koamtac.com/ has a bluetooth scanner that can be accessed via the ExternalAccessory framework (so no crappy HID implementation that hides the keyboard). The build-quality of the hardware is a bit flaky IMHO and the UI on that thing is horrendous. Pairing it with an iphone means visiting two menus with abbreviated texts and bad english.

The SDK is definitely done by hardware-people too.

http://ipclineapro.com/ is much better but also significantly more expensive. It's a sleeve you put your iphone in. It has a cool SDK and the build-quality is very convincing. As this uses the dock connector, there's no pairing issues either.

What ever you do: If you want to publish that app in the app store, be prepared for another hurdle, because if you are using any of these devices, you are not only dealing with apple approval, but also with approval by the device maker.

And you'll sign NDAs with both Apple and the device maker.

In that regards, WindowsCE is much nicer. You buy the device, you buy Visual Studio, you develop. No need to ask anyone for permission.

Great info, thank you for the detailed response.

You missed J2ME (which has certain, but not 100% overlap with Symbian OS development). Now even Bada supports J2ME apps.

Darn, I forgot to list it. I won't add it now though, as it would not be fair. There is an Other option for what is worth. If Other gets a huge number of votes, we'll know that options such as J2ME and others have a somewhat large market share among HNers, and it's not all about Android and iOS.

It's only an hour old - it's not as if this is precise science in any event. Phones with J2ME are very, very widespread. Granted, they don't do much, and are not the future, but it's probably still an interesting space for the right kind of niche product.

Yeah, give it time. It usually takes the BlackBerry people a few years to catch up to the present.

Wrote far to many games for J2ME the best of which was definitely Discworld, some really "great" memories of dealing with the fragmentation that was present. Mobile dev 7 years ago felt like a war against the brick walls that were in the way, no it feels like the handcuffs are off and the only limitation is ourselves and time

+1 for J2ME. Some of us will not let our Nokia 6303's go :)

Not only on old phones, think about all those Chinese ripoffs of the iPhone, Nokia, etc... yeah! running J2ME!

True. Also consider most of the world of Nokia Series 40, which by far destroys any other mobile operating system:

"Analyst estimates place the total number of devices shipped with S40 close to 1.5 billion, as of the end of 2010" [1]

[1] https://secure.wikimedia.org/wikipedia/en/wiki/Series_40

I'm genuinely surprised to see webos people. Are you still developing for webos or did you stop after the announcement?

Still doing some maintenance on webOS for now (firesale TouchPads have been a nice shot in the arm...plus the fact that Palm decided to give a bunch of promo codes away for one of my apps!) but I'm working on an Android app for now and probably Windows Phone after that.

If webOS hardware comes back, I'd love to develop more apps for webOS, though.

PalmOS 5! A tad outdated, but still running fine. It's "fun" to write apps in Pascal.

Still works though; a thing that many people don't consider!

Spending all of my programming time these days doing HTML5 Canvas stuff (work, personal projects, writing tutorials, helping people on StackOverflow, and helping those who email me for help)

Canvas is definitely on a slow-but-steady rise in popularity - there are a lot more questions each day on SO than there were a year ago. It's kind of nice to watch it grow, in a way.

I picked "Mobile-friendly Web development" though amusingly Canvas works in strangely different ways on mobile devices/different browsers, so it's more "mobile friendly" than Flash but still not quite the friendliest thing!

Once we start to see hardware accelerated Canvas on mobile devices, I think a lot of the mobile development will move towards HTML5.

<3 canvas. My team and I developed an HTML5 game over the summer, and we were consistently impressed with how capable hardware-accelerated canvas was of answering our needs.

I've only just started working through the big nerd ranch iphone guide and have a few basic, simple ideas for apps that I'll be trying to get onto the app store within the next couple months. No real delusions of grandeur, I'm just curious and have gotten somewhat bored of doing web development, so something new and challenging seemed like a good idea.

I'm blown away by the number of Android developers. I wonder how many of those who voted are just hobby / tinkerers. While looking for experienced mobile devs I have found many great iPhone devs and so few great or even interested Android devs. This poll speaks otherwise.

I do professional Android development. From a Swedish point of view, my observation is that we have more iOS developers than Android developers, but Android is quickly gaining traction. More and more customers ask specifically for Android _and_ iOS versions, instead of saying "we want an app" while thinking of iPhones.

We're about 50/50 with android/iPhone... It used to be everyone wanted iPhone and then eventually some folks would want android... now it's an even break. I think it's just a sign of the times, when you ship more handsets, more folks want software for the platform.

Work as a mobile apps developer fulltime. I mainly do Android and iPhone, but I've done BlackBerry and iPad as well.

For my spare time I work on my 2d android mmorpg: http://developingthedream.blogspot.com/

I develop for iOS and Android using LiveCode ( http://runrev.com ), it is a modern cross-platform HyperCard descendant. It is really powerful and fun to use, not to mention that it can deploy native applications to Mac OS X, iOS, Android, Windows and Linux. Personally I use a MacBook as my main development machine but at home I sometimes use my Fedora 15 desktop. I have LiveCode running in both of them and I deploy to to all systems. Curious people in here might want to check it out.

Since 2003 I've written BREW, J2ME, Blackberry, a smidgen of iOS, and a metric ton of Android. I now write Android software for a living, first for doubleTwist and now for Tumblr.

Thinking about doing Android-apps. And definitely thinking about exploiting the platforms (current) lack of tablet-optimised apps to make my own stand out.

Yeah, that's exactly where I am. The current lack of Android tablet apps is actually kind of exciting - we're in the calm before the storm, and anyone who puts out a kickass Android tablet app now is in a good position.

Great for a personal project. I would be all over that. Looking at it from a business perspective you can see why there are so few tablet apps in the android market. The amount of android tablets in users hands right now is small. You target a much larger percentage of the market by making an app for mobile handsets. While it might initially appear easy to make the app "just work" on both tablet and handset it is actually a considerable amount of development time (not to mention the often missed design and testing time needed as well).

I consider myself a mere tinkerer in mobile development (my day job does not involve such technologies directly). I like to learn new stuff, just out of curiosity. In the past I published an iOS game which netted me a whopping $55. I also have an android game in active (but very very slow) development, as well as a mobile-friendly web app. I don't expect to monetize (or even finish) the latter.

Hypothetically, if I were interested in getting in to Android development, but I didn't want to switch to using an Android phone, what would be the most cost-effective way to get my hands on an Android device that would be good for testing? Or is the emulator reliable enough to release without testing on hardware?

THe emulator is awful, even on really good hardware. I use a nexus one, which was the original developer device from Google, I think they have a new one out. Find out which phone is recommended by google.

OK, I had read the emulator wasn't ideal but didn't know if maybe other people had better experiences. Is it necessary to get the newest phone? I've seen an unlocked Nexus One, for about $350, seemed like a viable option. Woot just had the Galaxy S WiFi for $175, which seemed like an even better deal (faster, though I don't know if it'll run 2.3 or have Market access). Anyway, thought I'd ask; thanks for the feedback.

I test on the nexus one no problems. It may be better to test on a device that isn't the latest and greatest, one of the reasons I kept an iPhone 3g around for so long.

I released a first version on Android only using the emulator but of course its never the same as testing on a device. I have a Nexus One and it's a great phone. I think Virgin Mobile would be a good source of a second/development phone as you can get an Android phone without contract for $150 and unlimited data is just $35/month http://www.virginmobileusa.com/cell-phone-plans/beyond-talk-...

(iOS) This has been a question asked of me many times and my main answers are:

1) The other OSs are fragmented. If you develop a phenomenal app on one, it may not function the same on the others.

2) Your customers are self-selected as having a higher discretionary income than the other OSs, thus resulting in you making more money.

3) iPad


Oh! How is it?

It's not very fun, actually. I've been using Mirah (a simple Ruby-esque language with basic type inference that compiles to Java bytecode without a runtime dependency), so at least I don't have to write Java, but the Kindle framework itself feels very beta. My two apps have been bouncing back and forth with QA for several months---Amazon QA does an amazing and through job (including user testing + feedback), but it is frustrating to work around the framework so often in my application code. For instance, it takes on the order of seconds to load and display an image (not the e-ink display, but actually reading the image into memory). The framework requires that an app starts up in roughly 5 seconds, so I've spent a lot of time spinning up threads to load images in the background. But then you need to make sure these threads are stopped when the app exits (or goes into screensaver), and the framework sometimes calls stop() before start(), &c. &c. Oh, and you can only test this on the hardware itself, because the simulator doesn't emulate the Kindle CPU (it just runs using your system's Java)

To get more of a feel of what you're in for, check out this forum post + Java sample code:


A limited version of Java 1.4, the UI set is custom build, Writing a kindlet bit similar to writing an old fashioned applet. Limited memory/processing power though, but you can do fun some stuff, but the eInk slow refresh rates is the largest limitation.

There is no functional app store, right? There is an app I'd make on Kindle if the distribution channel works. Now I'll make it for iPad and maybe Android tablets.

There's the Kindle store.

I think the main problem is getting ahold of the SDK.

I thought it was a variety of J2ME they used.

It's like turning a real page! :)

I posted the 24-hour results in chart form here: http://programmingzen.com/2011/09/20/mobile-development-mark...

I work on Android and develop small games, I also wrote an android program for a programming contest that was an inbox for a workflow management system called Documentum..

Android and iOS consume my day. It is a wonderful time to be programming in these environments - so rich, so easy to use, and so fun.

Android, iOS, and BlackBerry here.

While Android and iOS have great Tools, I still hope RIM is going to release a great Native SDK for QNX.

I am surprised at the amount for Windows Phone, given the age (and adoption) of the platform.

This reflects my experience with my clients. I develop for iOS, Android, and WinPhone7. I've lately gotten quite a robust bit of WinPhone7 contract work coming through my doors. And it's not because there seems to be some huge explosion of demand for it, but rather that nobody else locally seems to be developing for it. So due to the scarcity of local developers, I'm getting a larger proportion of Phone7 work, and I'm having to raise my rates.

J2ME should be an option

No mobile for the moment, used to do J2ME some 5-6 years ago.

iOS, Android, webOS, Blackberry, J2ME

The first two are so much nicer to develop for due to tooling, webOS is pretty nice but perhaps a dead end!

I've been wondering that about webOS. I've started making a few apps that fill my own personal wants for the platform; and it seems that in the short-term it's a platform primed for opportunistic development.

Hundreds of thousands of consumers have recently gotten their hands on a Touchpad, and the HP store has so little new content coming into it that competition is pretty much non-existent. There are also plenty of gaps that are entirely unexploited by existing apps that haven't been found in iOS/Android since they launched.

What's this Backberry OS? :-)

About 6 weeks ago, we launched our new mobile website using jQuery Mobile at www.crutchfield.com (viewable on your smartphone). Fun project and JM is truly impressive!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact