I think that's the actual problem. Fragmentation wouldn't be a problem if the developers thought that they'd make the money. After all, a fragmented market doesn't appear to have hindered the PC. Admittedly, when the PC took hold of it's market, there wasn't much coherent competition, and IBM was still reaping the benefits of the phrase "No one got fired for buying IBM".

While Android hasn't had that advantage, and has done an absolutely amazing job of building market share, one thing Google don't seem to have done is push the App store. Apple went from almost denying the need for users to write their own apps to being extremely active in pushing the app store as an advantage of iOS, even to the point of featuring several apps on TV.

Google, for their part, push their own services. As do the individual handset manufacturers. No one seems to be actively pushing the App store.

Google, TBH, only seem interested in future opportunities for advertising. Not in pushing developers interests. That may or may not be true, but the recent delay of payments won't have helped that perception. Apple, for their part, don't actively push a lot of apps, but they *do* actively promote the App store and do seem to actively protect their developers as a group. Something which Google don't.

As I say, fragmentation itself isn't a problem. While developing and testing for different OS versions, cpus, memory sizes, screen sizes and graphic capabilities does incur extra cost, if they have the resources, a lot of developers would be happy to allocate them *if* they knew they were going to get a return on their investment.

The problem is, they often don't know.

Another reason Developers may be a little cautious is that iOS users are (in general) more willing to pay for their software. Getting your free app installed on 100,000 phones is a good thing. However, it won't pay the bills nearly as well as getting your paid app installed on 10,000 phones.

Re: "People buy iOS apps - people want Android apps for free"

You're right. Google consumers are accustomed to getting great services for free: Google web search and GMail were the big ones for me. The advertising was non-invasive, non-animated and non-irritating. I don't recall ever clicking one, not even by accident. You can't say that for many other free web-sites run as a commercial enterprise. All it cost the user was some privacy, but so few users have ever really cared about that (lamentably).

By offering these services, which I and many have happily used for years, Google now dominates internet search and takes huge sums of advertising revenue that come from that position. And as far as I'm concerned, it has earned it fair and square.

Unfortunately, as an app developer it is very unlikely that you will ever establish such a commanding position in your target market. You either have to have a great product (that someone doesn't quickly rip-off) or leverage some other advantage you have in some other market (such as media content). And for me, this is where the free advertising based model falls down.

Re: Opportunists

nothing wrong with being the first fish in the pond, Gives you a better chance to become the Big fish,

Re: Here we go...

6 thumbs up for this tosh shows the level of ignorance and hatred of Android here. It might be a paid survey, but the bigger complaint is that the conclusion is wrong.

"Fragmentation" in terms of CPU is about as relevant as fragmentation in terms of the cases people put on their iphones. You don't HAVE to support every android device! If you want to write an app which would need quad-core then you are able to distribute it to those devices and restrict it from others. You can design for one screen size if you want, but you're not forced to, and the tools available make it easy to support multiple sizes.

Can anyone add an OS name to this sentence which will result in a true statement about fragmentation making app development a pain in the ass? Android certainly doesn't fit this sentence.

"There *is* fragmentation in <OS> - not all <OS machines> have the same class of CPU, the same size of screen or the same OS release."

As a rule of thumb in flame wars like this one, if someone uses the word "Fragment" then they know what they're talking about. If someone uses the word "Fragmented" then they don't.

Re: Here we go...

"You can design for one screen size if you want, but you're not forced to, and the tools available make it easy to support multiple sizes."

It's really not that easy. The use of different screen ratios (not resolutions) is the biggest issue for app designers. You may not realise how many applications are designed using nothing more than Phototoshop, but it's very common for the design to be approved purely on the basis of a screenshot-flow. Often, for marketing apps, it's the marketing agency's creative team that designs and lays out the application; not the developer.

The designers often don't know that there's more than one screen-size on Android, and so they will often just go with 800x480. This is a 16:10 aspect ratio. However, many budget Android phones use a 320x240 display, which is a 4:3 ratio and to add to your pain, there are both landscape and portrait versions of this screen size (HTC Chacha = 480x320; Galaxy Y = 320x480). You can't fit the same information into this format of screen as you can on the taller ones. Now, to add to the fun manufacturers are announcing "720p" displays on high-end phones -- 720p is 1280x720, which is 16:9; appreciably narrower than 16:10.

That is what I mean by fragmentation.

Even with good toolkits and APIs to solve the superficial issues (like scaling between different resolutions), only the most trivial UIs will not need to be altered to suit different screen proportions. And excuses like "but those aren't really Android phones" will not wash when your client finds out that a huge chunk of their customers can't use the app they've paid for...

Re: Here we go...

If "the designers often don't know that there's more than one screen-size on Android" then get some proper designers.

To support different screen sizes use fragments. Yes, sizes. It's not just about ratios. On a large screen you might want to display a menu and content next to each other. On a small screen you can display one at a time. If the Size-Ratio combination doesn't allow a menu and content to sit side by side then just show one at a time.

http://www.developer.com/ws/create-flexible-android-uis-with-fragments.html

Check out the images at the bottom of page 2 to get an idea of what's available.

Re: Here we go...

I agree that it's difficult to design for all screen sizes. Any company that has a design time draw up a design for one size and then assumes it should work on all flavors did not do their homework. How you go off and NOT know that the target phone (android) has different sizes to be concerned with.. I would not want to buy the app from that company.

If they are targeting the 800x480 screen size, which is the majority of phones on the market these days, then they should either also design a subset for smaller screen size, possibly as a separate app with meta info that allows the market to only show the small screen size app to phones with small screens, or at least draw up the UI for different screen sizes and let the developers build the app to accommodate it. The app can detect the screen size and adjust accordingly.. it's a little more code, but not difficult.

Re: Here we go...

this resolution problem is Googles problem for now proving good UI guidlines until ICS. Now there stuck with millions of devices manufactures wont upgrade.

Re: Here we go...

Yeah, I've seen some UIs designed like that, and they generally work about as well as the rest of the app. If you're writing a form or something similar, use scrolling. If it's a game or something graphical, you pick a distance unit to work with and scale to fit. As for aspect, go look at how film directors deal with distributing a film in different aspects. I'm not an Android developer and I'm not sure of the graphics API, but if it's anything like OpenGL then changing aspects is one call to glViewPort. Not that hard. Someone here already mentioned web pages - that is how your app should work for static forms!

If the UI is a rigidly-designed exercise in corner cutting, I have to wonder what is under the hood?

Re: And now a word from our sponsors...

Indeed. WP 7 hasn't gained market share yet. However, it does seem to be getting good reviews, and in the past, when a product or product line has had a slow start, Microsoft have shown that they can be extremely patient while waiting for it to take off. Not to mention the deep pockets to keep financing it and advertising it.

Look at the xbox. While I had one and loved it, the first Xbox was always a distant competitor to the PS2, but MS stuck in the market, and eventually came up with a console that has equaled or beaten both the Wii and PS3 in sales.

Re: And now a word from our sponsors...

I was working on the assumption that developers are showing an increasing interest in Windows Phone because Microsoft's development tools and languages are very nice. So they're interested in developing for Windows Phone in the sense that they're likely to play around with it because they expect it to be a pleasant experience and because learning additional platforms and languages never hurts.

They're probably also optimistic that a sufficiently large market will appear for it to be worth releasing products. We're probably only talking about something minor like a 10% share for the sort of apps that are not in themselves directly profitable (Facebook, DropBox, other service front ends, anything promotional or sponsored) to be worthwhile to port, and that wouldn't exactly disturb the Android freight train.

So, no, I don't think developer interest need always be a trailing indicator of market share. Indeed it's almost the only way I can make sense of the survey.

Re: Statistics

Actually you can, that's what statistics are all about.

Ask 10% more and the figures could be better or worse.

The proof will materialise in 5 to 10 years and we will see if shit sandwich .10 still has apps

Re: Statistics

Hmm.. replace 0.002% with 0.2% :)

Re: Statistics

Haha..I thought this very same thing! However, 280,000 devs is about 5% of the total android developer community. There are millions of android developers (or wannabe's anyway) trying to break in to the market. Makes me wonder if this is some ploy by an apple employee to make android look bad..but then my long post about why this fragmentation is happening would counter this sentence. ;)

Re: hah hah hah

Although in agreement I have downvoted you for being childish and triumphalist in your tone.

Re: hah hah hah

Well said. I too agree with Mr Hah Hah Hah but still think he's a prat.

As for Android's fragmentation; we code for iOS and Android here, and the main problem isn't that Android's UI or screen size is fragmented, but that some things just don't work properly on certain Android releases, or even different handsets using the same release. A bit like the fragmented mess that is HTML development (shudders...)

Re: hah hah hah

you, sir, are a funny, funny man.

now make sure you've taken your pills next time.

Re: hah hah hah

Why anyone would agree with this moron is beyond me. So what you're saying is you want apple to be the only thing out there.. no competition? Watch how fast apple goes from their minor yearly updates now to something every 3+ years when they have nothing pushing them to be better.

copy? ROFL.. as many already said, apple copied too, on many occasions. Copying is a form of flattery is it not? Besides, there are only so many ways you can make a touch screen do something. So if they had triangular icons that required a hard touch then a swipe to execute, would that make you happy? Oh wait.. apple apparently invented swipe too.. so that's copying as well. I got it.. all android devices should have cameras that know exactly what your looking at..then you blink.. and it runs. MS phones will need to do something else..they already have Kinect technology.. so perhaps they need to invent the brain receiver..you think it, and their phones will execute the app. Now that is fair right? That way nobody copied poor apple.. the company that after years of massive profits is finally giving a little bit back to the stock holders and continues to waste billions on trying to stop competition rather than compete fairly. Yah..that's the company I want to give my money to.

Re: Hmm

on certain models only and on the last two releases of that OS ...

thats my bet sadly :(

Re: I wrote the hah hah hah post

" I wrote the hah hah hah post"

I used to have an incontinent puppy. It knew it wasn't supposed to shit in the house but it did anyway.

I'd get home to find it next to a stinking pile of its own doings looking at it sadly, clearly owning up to me that it was responsible.

And I used to think "It's pretty bloody obvious who did it, you idiot; rather than confessing how about you just stop shitting everywhere in the first place?"

It wouldn't have understood, of course.