The problem with Android is choice

This is the crux of the argument. That choice is bad, and the people who like Android actually like tinkering, and don't really like customization, and it's just wrong.

I personally use my phone more than any other non-work device, and I love having it personalized to make everything I do often easy to access, and put everything else to the background.

I chose the best phone for me out of the very few real contenders on the market, I chose to replace my OS with Cyanogenmod, chose Nova Launcher, chose my layout and shortcut gestures, and chose my default keyboards as Google's English and Japanese keyboards. I haven't touched any of these for a couple of months, with the exception of replacing a couple of the gestures.

Phone usage differs a ton between individuals, and I'm glad I didn't lock myself into a one-size-fits-all solution.

To bring a bit more consumer psychology into the discussion, people love SUVs because they give them the choice to go driving off-road, driving in snow, etc. Those SUV owners aren't then suddenly overwhelmed each morning with the question "Should I commute to work or should I drive to Tahoe?"

Android is a flexible tool that allows you to make choices when you want to make choices.

The problem with usability nuts and Apple nuts is that they fall into the "one perfect solution exists" category. That's fun for them, as they like their uniform devices just fine and get to write whiny blog posts about how all the other phones are Just Wrong. But they've made the mistake of not understanding how their motivation differs from the rest of us who like to have a deeper connection with our tools.

Yawn. Take all the fancy verbiage away and this is just another platform flame. Use what you like. Don't tell me how to use my phone.

You'd jump from a Linux box to a Solaris box, and things wouldn't work the same way? Simple things, like flags on "ps". Or, more seriously, the "shutdown" command (shudder).

That's only the CLI. Then you get into what Window Manager you're using, what desktop environment, etc.

Did any of this benefit the end user? In the end, most of this all ended up coming back to something that looks somewhat like OS X or Windows. A launcher bar and a menu bar, and maybe some status icons. Frequently ones that looked and worked exactly the same as OS X or Windows.

Fundamentally, all that choice didn't help Unix get market share. It hindered it. People didn't know what to expect, and wanted consistency. The same commands doing the same thing for the rest of eternity.

Incidentally, this also explains the Windows 8 backlash. I'm pretty sure that most non-technical people would prefer a innards-updated, security-fixed version of 8 that looks and works identically to XP.

In short, the problem is not choice, it's lack of consistency.

In my opinion though, the solution is not less choice, but better and stronger defaults. There's no reason why android phones couldn't be as simple as iphones out of the box, but have more options for configuration for users who wish to dig deeper. Instead, at the moment we generally have phones full of bloatware, where many of the defaults seem designed to sell products and services of questionable value, rather than to aid the user.

The example of default application bindings in the article is a good one. That system is indeed terrible currently, but there's no reason it needs to be. At a minimum it should be possible to edit the bindings of any installed application, rather than simply clearing them all. (Right now it almost seems like the designers were trying to follow this philosophy of minimizing choice, but ended up with a hybrid that is both more overwhelming and less customizable than it could be.) There is also no reason why users necessarily have to be interrupted by the choice of app to perform a given action. Instead, an app could be chosen based on some predefined hierarchy, such as first installed. Then let the user override that choice if desired. (For example, by pressing the menu button within X seconds of one app launching another, then choosing the appropriate option.) You get the same benefit without overwhelming less knowledgeable users with choice. (And yes, some users might never realize that there was a choice at all, but that's a fair sight better than there indeed not being one.)

Everytime I hear this argument, I can only think that these people have developed a bad taste of Android and haven't used a stock experience for enough time. You don't need to tinker anything. The stock Android experience is a good experience* by itself... but the option to tinker is there if you want it. If Apple made a car, the option to tinker wouldn't exist.

Assuming all Android devices would, could or should have the same interface is like wishing all BSD-derived system were the same. Clearly there's room for NetBSD, FreeBSD, DragonflyBSD as well as OS X and a multitude of embedded systems?

If there weren't any differences between an LG phone and a Samsung phone or a HTC product -- how would they compete? There is no Android without (some) fragmentation.

Does this complexity cost (money, effort, q&a and developer time etc...)? Certainly.

But to me the price is well worth it with the likes of the Cyanogen project -- that allows all kinds of freedoms. For me personally it's good -- but if I were in charge of developing some kind of system (say handing out internal devices at a company) -- it's be much more valuable. I'd be in control of upgrades and features. I could maybe patch some minor issues myself. I could hold off on unimportand updates, and rush those that affected my users. I can't see how you could both have this, and at the same time avoid having a somewhat heterogeneous system. I know Amazon agrees (look at the latest kindle tablet).

I agree that a lot of time has been wasted by Android vendors on differentiation for differentiations sake. It's easier to superficially fuck with the UI than actually create a useful features.

I'm a great believer in Loss Aversion (http://en.wikipedia.org/wiki/Loss_aversion) and Hotelling's law (http://en.wikipedia.org/wiki/Hotelling%27s_law) when it comes to device preferences.

These days I don't think many people are arguing that the iOS UI is significantly better than stock Android, Android has closed the gap or jumped beyond iOS depending on your personal preferences (Hotelling's law). Then consider how many people won't even consider iPhones because the screen size is too small (loss aversion).

2. i like the feeling. that if i ever need to change something, i probably can. its just a search away.

3. i like it that i can change things beyond even what the article was talking about if i wanted. e.g. say the if the silk browser was reviewed to be much faster than webkit/chrome/ff and i wanted to try it out on my lg phone..i can try...

4. 'choice' may even open up new functionality perhaps? does ios have llama/automateit/tasker equivalents?

While setting defaults in stone and never changing them might work if some people like your choices, being flexible has always been the greater path to success.

Witness the car industry, fashion industry, PC industry, etc... For better or worse, we like having alot of choices available.

These choices have to be answered one way or the other. The author of the article suggests that the company is better left answering these choices on behalf of the user.

Why would the company be better qualified to answer these choices over their users? That makes absolutely no sense to me.