How hard is developing for iPhone?

Answer 1

We have started developing about a year ago and currently have two OpenGL 2D games on the market. My experience so far:

1. Simple application can easily be a one-man show. For a medium-sized application you are likely to manage with just one good programmer, but usually there are other people needed, such as a graphics designer. This highly depends on the nature of your application.
2. A bit steep if you have no experience with Objective-C and Cocoa. C knowledge helps, as does knowledge of some OO and computer language concepts. Even then you’ll spend some time getting used to their way of doing things. (Which is usually well-thought, but often different from what other people/languages/stacks do.)
3. The biggest non-programming issue is the crazy provisioning and review stuff. It takes a while to get used to all the profiles and certificates and signing voodoo. You are going to hate it, but will get used to it.
4. Selling the app is hard. You either have to be one of the lucky ones to make it into the featured apps on the device or you have to be some big title or your application has to be something with a clear audience (like Geocaching) or you will have trouble getting a decent coffee for what you earn. (I am over-simplifying here, but it’s mostly true.) The selling process itself is pretty much painless – $99/year and Apple gets a third of what you earn.

Answer 2

It's rather difficult to answer your question owing to the fact that often this is highly subjective in my previous experience.

1) Generally the effort is much lower than the one required when using a different platform. Those acquainted to software engineering principles including the use of design patterns etc will find that the SDK is built around all of the common abstraction we are used (except a very small part still using C style procedural APIs).

2) The learning curve is steep for people rolling this on their own, is really easy for people being taught on the matter. A fast course style exposure to the SDK and tools (say 40 hours total) it's usually enough for people to become proficient enough.

3) There are no hardware issue to be taken into account, at least in my experience. As already pointed out by Zoul, provisioning the devices takes some time to get used to. The submission/review process is in my opinion a little easier.

4) Selling is as difficult as it is on other platforms. But if you have got a really brilliant idea, then you usually sell many copies of your software. Or, the idea may not be so brilliant, but the software you develop is fundamental for a specific field targeting people always on the move etc. Just developing something without a clear target is the perfect recipe for disaster.

Answer 3

Let me give you some guidance as I have worked on JQTouch. Its a library that build using JQuery and it also provides multi-touch related features too. Basically this is for UI related stuff. Please have a look at JQTOUCH and look at the code samples. The business logic can be done in any server side technology of your choice.

Summing up the things with your relevant questions

1. Effort is not that tough. Easy for developers to develop. Less documentation.
2. Pretty easy
3. Emulator could be downloaded from Emulators
4. Not much knowledge on this.

Answer 4

1. What is your definition of "medium sized app". It could easily be just yourself, or it could be a few people including a designer. Also, to some degree if you have more time you need fewer people.

2. That depends heavily on your experience to date. Many people have come over from .Net and Java development and not found it too hard... you probably need at least a month to be comfortable with a lot of the concepts.

3. You need a Mac, that is it. Any Intel mac with 2GB of memory will do.

4. It's very easy to sell, since all you do is upload a binary and (after a wait for Apple to approve it) Apple puts it up for sale. You need no servers. You do need to pay a yearly $99 fee to develop.

Answer 5

1. Depends what you mean by "medium-sized". Also depends how long you want it to take. In general, to make a decent app, you need a combination of things: programming skill, artistic skill, design skill, and business knowledge. Most people don't like to do all of those things. I'd guess that the majority of iPhone apps only have a single actual programmer, though. You can tell the ones that were written by a programmer who should have gotten some help with the other aspects.

2. Depends what you already know. It took me a month before Objective C stopped seeming really bizarre, and I've used lots of different languages.

3. The hardware isn't a problem unless you don't already have a Mac, an iPhone, and an iPod Touch. The biggest non-programming thing for me is the App Store review process; you have to understand that when you think you're done, you're going to need to wait a couple of weeks, and it's possible that the idea you thought was great falls into some category that will never ever be approved, or that you'll have to change your app's name, etc.

4. It's easy to offer apps for sale on iTunes, once you pay your $99. If your goal is for people all over the world to download your free app, or to put your app on sale and make tens of dollars, the App Store is great. If you're hoping to make millions of dollars, or even thousands, you have to be some combination of competent, persistent, and lucky.

Answer 6

Perhaps an analogy... I want to know how difficult is is to build a house.

1. In terms of builder vs house complexity. To be clear: how many people do I need to build a medium sized house?
2. Power tool learning curve.
3. Permits, plans, and other non-building stuff.
4. How easy is it to sell my house?

Answer 7

1. One programmer (a good all-rounder) can easily do it. Once you have done your first app you will be surprised to see what sells and how little actual programming there is in some apps. The reason you have to be a good all-rounder is that the apps that succeed have it all - design, inspiration, execution. Consider involving a designer if your taste doesn't match what seems to be popular. Don't expect to ship anything with standard UIButtons, on the store it stands out like the proverbials. Normal practices are essential, e.g. revision control, issue tracking and all that good stuff. It seems to matter more the higher level an API you work to.

2. SDK learning curve - not so bad. Initially you struggle with why a NSArray can't take another value, but inside 1-2 months you'll be subclassing things all over the place. However read below, don't try to do too much custom stuff...

3. You need an Intel mac, you need at minimum an iPod touch to submit an app - try to submit without testing on actual hardware and you will miss something, and it will be rejected. You don't have to have Snow Leopard but if you do the static checking feature is pretty neat. If configuring choose more RAM over more processor speed. BIG (or multiple) screens are, as with any coding task, a big advantage. The new 27" iMac would be a great development machine.

4. Selling is not so hard. Provided your app isn't complete rubbish and doesn't get 10 1-star reviews right away sheer numbers will get you some sales. To make it big is hard, and you will need to investigate marketing, review sites, twitter, youtube, in fact to your all-rounder programmer skills you can add marketing director. The noise on the store (sheer number of applications) means only a truly stellar application (i.e. featured by Apple) will stand out in the absence of any other effort. There are probably plenty of apps on the store than in 2008 would have made their developers rich, these days they are lucky to sell 1000. The cost is $99 to join and after that you get 70% of sales revenue while Apple keeps 30%.

Additionally... With the context that I am a C/C++ programmer who has spent most time programming embedded devices and handsets, with almost no C#/STL/Java...

Here's what I found easy/good:

• Xcode (although I admit getting started was jarring coming from Visual Studio)

• brevity - what you can do in just a few lines of code is amazing

• Stanford CS193P iPhone programming class on iTunes University - great intro, free!

• WWDC 2009 video sessions. Not cheap but probably worth more than what you pay in terms of in-depth knowledge. I've been to similar developer conferences that were more of an excuse to stay in a nice hotel and do some duty-free shopping but if I'm not at WWDC 2010 I will feel like I'm at a severe disadvantage.

Here's what I found hard:

• knowing just what storage classes to use in a certain situation. My first huge performance problem came from using indexForObject on NSArray with hundreds of thousands of objects. Obvious now but who knows this the first time it happens to them?

• "letting go" of preconceived ideas about what a UI should do. Don't go laying out a .xib until you have used at least 20 iPhone applications and have some idea of how things are usually done. Doing things otherwise is not only likely to be harder, if your idea is too far against Human Interface Guidelines chances are it will never be accepted to the store anyway.

• Xcode debugging messages - do google these because they are cryptic at first but when you find other people explaining them they start to make sense after a while

Here's what I found completely perplexing and got working through trial and effort:

• Apple's on-device provisioning process
• actual submission to the App Store

So far I have one small game on the store. It's not a particularly good game unless you really like that kind of thing, and only scrabble nerds do, but it still has 10 sales after 1 week and that's with no publicity at all. I did it to get experience with how the store works and by that measure it was a success. In learning curve terms it took me probably six, seven weeks full time from opening the first Apple doc to submitting the game, but today I could do it in about two days.

Answer 8

Very subjective. A one person app developer can develop a medium sized app. How long will it take? Depends on how much free time the developer has and how much experience with Obj-C.

I learned some parts of the SDK in less than a day. I still don't know the entire SDK, as I haven't needed to. I doubt that any one programmer would want to spend the time to learn the entire SDK. For example, if you aren't doing anything with the accelerometer, why study it?

You need to roll up your sleeves and delve into it yourself to see how long it takes you. If you are asking for your team, then you will have to judge how well their expertise will apply.

As for selling on the iPhone, there are some easy aspects, like not having to worry about packaging or salespeople, but you still have to sink money into marketing or no one will find your app in the almost 95,000 apps on the app store today.

If you're asking because you keep reading that it's an easy "get rich scheme", then I'd say you're in for a surprise. Despite the reduced overhead in some areas, and low start-up capital, it's as much work as any other software venture, since the ratio of team members to work to be done stays about the same (the economics of a $2.99 or $4.99 or $9.99 apps forces you to have a smaller team).