Hacker News Reader

But you have to admit, there's a cultural divide between Android and iOS. Android is extremely well suited for working, passing data seamlessly between applications, etc. It makes it easy to get things done in the same way you might interact with a desktop computer.

iOS, however, takes a different tack, focusing on self-contained, bite-sized apps. You hop into something, use it, then leave. And when you're in that app, it completely owns the experience. In that world, things like intents, task killers, and filesystem browsers just don't make sense. Which almost by definition precludes things like Chrome-to-Phone or Swype.

In that way, I see Android and iOS relating very similarly to Windows and OS X. "Beautiful" simply isn't part of the Windows application lexicon, and yet, it's used frequently in reviews of OS X software, along with terms like "polished," "lovely," "stunning," and "gorgeous."

What allows OS X to have beautiful apps? What's preventing Windows from having beautiful applications? And are those same factors at work in Android and iOS?

Absolutely, I'll admit that straight away. It's very much a different expectation of how you use it, and I think that's at the core of Gruber's complaint. He is very used to the iPhone paradigm, where you engage with those bite-sized pieces and then you stop and go do something else. His complaint seems to be that Android doesn't really have bite-sized pieces like that which flatly trump their iOS couterparts, but I think it misses the point because Android isn't just an ugly iPhone, it's a completely different approach to mobile computing.

He's effectively asking "Why isn't the Nexus One a better iPhone than my iPhone?" which...just completely misses the point. The iPhone does the iPhone experience extremely well, and I don't think you'll find many people who will seriously argue that the iPhone UX isn't amazingly polished, but the assumption that it's the only way to use a mobile phone is naturally going to lead you into some completely wrong conclusions. If your whole world is the iPhone, the concept of Android intents and what they mean for how you use your phone is going to be completely lost on you.

I think that "beautiful" apps start with the platform. Most people would agree that OS X has better aesthetic sense than Windows, and most would agree that iOS has better aesthetic sense than Android. Each platform establishes a baseline for what apps should look like, and apps tend to conform to that baseline for the sake of uniformity. On Windows, the worst-looking apps are the ones that diverge from the OS baseline (I'm looking at you, iTunes and anything that uses GTK+). Android's baseline is pretty low - it's highly utilitarian and very typical of Google, which is famous for doing all of its UI design based on how things test out, and while there are some apps that diverge and look good (Twitter, Tweetdeck), there are others that are just terribly bad (Fring, for example).

My hope is that Gingerbread raises that baseline, and sets a new visual target for developers to hit with their apps. We'll see.

> I see Android and iOS relating very similarly to Windows and OS X.

Actually I think the comparison Gruber makes of 'PC vs Console' is more accurate.

I don't understand how Android allows you to pass data between applications while iOS does not. I'm under the impression that I can use copy/paste and custom protocols to pass information between applications on my iPhone. Am I missing something?

If there's any cultural divide Gruber alludes to, it's that iPhone users tend not to know a whole lot about how the phone works deep inside, and whether by necessity or curiosity, Android users do.

You say "iPhone users tend not to know a whole lot about how the phone works deep inside, and whether by necessity or curiosity, Android users do."

I don't think that is true. Most Android users have no idea how their phone works inside. They are not techies, they are just smartphone users. Most people have no idea what a computer does inside. Don't think that all Android users are techies because many on HN are.

I probably know a dozen people in my circle of friends with Android phones at this point. Of those, maybe 3 would give you any more than a blank stare if you asked them what CDMA and GSM are. I'm the only one out of that group that has a rooted device.

I think technically inclined people are more drawn to Android, but I don't think that anything close to the majority of Android's user base is technically inclined by a long shot.

I don't mean hardcore, nuts and bolts stuff, just that they're more involved with managing how the operating system behaves. And I'm not saying this, I was inferring from Gruber, who notes how many task managers, etc., are the most popular apps.

That's the worst kind of user, though. The "power user" is almost the bane of my existence as they do not truly understand anything yet think they understand everything.

These are the same morons that dig into the Windows registry for "super speed optimizations" and end up blowing their installation to hell.

Applications on iOS aren't allowed to touch storage outside of their own data store or the photo roll. Some developers used to backdoor around this by passing data via the photo roll, but Apple started denying app approval for this a while back.

Some developers used to backdoor around this by passing data via the photo roll

I've never heard of an app that did that. Can you name some?

I got it slightly wrong, it seems. Devs were using the photo roll as a way around the lack of USB mass storage support on the iPhone.[1] It's the Address Book that some devs have used to pass data between apps.[2]

[1] http://techcrunch.com/2010/02/02/apple-stanza-usb/

[2] http://stackoverflow.com/questions/220630/how-to-share-custo...

Still wondering what apps uses that trick, and which ones have gotten rejected for trying to use it ;)

Well, it's pretty clear that Stanza (from my first link) got in hot water with Apple given they pulled the feature at Apple's behest.

As for the address book thing, I can't name any specifics as I don't follow iOS overly closely; I just know it's a hack I've seen mentioned enough times that it seems to be fairly common knowledge that it can be done, just with the caveat that it's also against the rules.

TextExpander uses it for their SDK. Snippets you add in the TextExpander app is available to other apps which integrate with their SDK.

When speaking of passing data between applications, I'm specifically thinking of the use cases enabled by Android's intent system. Applications can register their ability to respond to and fulfill certain interactions, and then the OS mediates between them.

Say, for instance, you're in the Twitter app and click on a link. Instead of embedding a UIWebView, the application fires off a message saying it needs something to display a webpage. The OS then looks at the list of applications that can handle that intent and selects the best one available. If you only have one application that can fulfill that intent, it launches and renders the linked page. If you have multiple browsers, Android prompts you, with an option to remember your preference for that intent.

Which means that I can click a link in Twitter and open it in Opera. Or Firefox. Or any third party browser. And then when I open a link in another application, my preferred browser still gets to handle it. The Twitter app doesn't get to own that experience in the same way that it can on iOS, since it's actually backgrounding itself and letting the other application fully handle the task.

This also applies to getting directions, calling people, sending text messages, etc. The necessary data (an address, phone number, or message) is transparently shunted from the first app to the next. That dance greatly weakens the boundaries between applications, but enables some really great workflows.

Similarly, say I'm looking at my photos and I want to share one with someone. I simply hit the "Share" button, and up pops everything that can handle working with an image. In my case: Bluetooth, Messaging (SMS), Picasa, Twitter, Facebook, Email, and Gmail. Even better, were I to get fed up with the official Twitter client and go install Seesmic, it'd also pop up in that list, automatically, and without any changes to the Photo Gallery itself.

Say I install an FTP client. It'll also appear in that list. Google Goggles, too. Any image I can share, I can also send to Goggles for analysis.

As far as I know, there's nothing like this on iOS, but I'd be happy to learn that I'm mistaken.

Not all applications embed a UIWebView, and many that do also offer opening in Safari as an option. There are also between-app communications like Twitter's "Save Draft to Birdhouse..." or sending addresses to the Maps application. Not as flexible as the workflows you describe in Android, to be sure, but in that data is being passed between its apps, I believe it was overly broad to describe iOS as focusing only on "bite-sized" and "self-contained" apps.

What about opening in Opera? Skyfire?

I really think the Android back button makes all this possible (and perhaps better than on iOS devices. If a link in the iOS twitter app opens Maps, the only way to get back to where you were (pre-iOS4) was to hit home and launch the twitter app... which may have saved its state. In Android, such a workflow would just involve pressing the universal back button. I suppose in iOS4, you can now double tap home and find the twitter app in the list of open apps.

That's not to say Android is better for opening map links from twitter... just that it's a different way of doing things. I would say that most criticisms people have when jumping from android to iOS (or vice versa) stem from trying to use one OS in exactly the same way you used the other OS.

I agree, the back button is one of the best parts of the Android experience. I find myself reaching for it every time I go back to my iPod Touch.

> iOS, however, takes a different tack, focusing on self-contained, bite-sized apps. You hop into something, use it, then leave.

Right, if you think of the user workflow as a graph, the iOS enforces a kind of star topology on the experience. That's the model it's built on. You start from the homescreen, then jump to an app, then back to the home screen, repeat.

Android is more of a web browser stack model, you jump into an app, that app may launch another app, but you can always hit back and slide one step back, if you hit "back" enough, you end up where you start (or you can shortcut it with the home key). The user workflow graph is more random as people bump around between apps.

It would probably be more appropriate to call Android apps "pages" instead.

From what I've seen, this ideal workflow is not near as much of a reality on cheap Android phones with not as great hardware. It is a reality on phones like most of the Droid series (I own a Droid 2) and the Nexus One, but most of the people I know with Droids have the cheaper phones like the LG Ally and Motorola Backflip which just don't work as nicely this way. You end up wanting to go back to the app paradigm on phones with smaller screens and slower processors.

If you read the article, he's not making that assertion about the relative worth of Android. (And opening up with that as the basis of your argument is a disingenuous tactic, if done intentionally.) He's arguing that the third-party software library is weak, especially away from "enhancement/tweak" type software.

Flexibility is useful only insofar as it translates into capabilities of high value to end users. It's not a feature, especially if you're an end user and not a programmer/hacker. In most product domains, users have clearly shown that as the category matures, they are more than willing to sacrifice flexibility, the ability to modify and even the ability to repair at home in favor of a more integrated, reliable and cheaper product. (See: cars, home electronics, home appliances, clothing, food.)

Consumers evaluate product purchases based on what the device can do for them out of the box, not after hours of tweaking and configuration. Frankly, I don't think _either_ platform truly has "killer apps" - I don't think people are going out and buying iPhones so they can play Star Wars Arcade: Falcon Gunner or something. A "killer app" is an exclusive app that drives customers to buy that platform just so they can run it. Arguably at present no such app exists on either mobile platform.

The point you make about the killer app is very true. Halo on the XBox came to mind. I think Google hoped turn-by-turn navigation would be it initially for Android, but I agree that (for the average user) there aren't any killer apps. If you get into domain niches (such as medical), that might be different.

Google Nav is pretty friggin awesome. I had a TomTom for the first year I lived in NYC and the last year I owned a car, and Google Nav was comparable on all fronts, as demoed in rental cars up and down the Eastern Seaboard since then.

Notable downsides include battery life and built-in mounting options. Notable upsides include no upfront purchase, free updates, constant updates, better information density (pixels per meatspace volume, roughly), ease of ad-hoc "mounting" (e.g. next to the speedometer, behind the steering wheel), and automatic "straight-to-pocket-on-exit" theft protection.

It's free, it works well, and it integrates nicely with other Android apps (notably Google Maps).

The problem with Gruber's argument though is there are no killer apps on the iPhone either. He lists a bunch of iOS apps in his story that I've never heard of. There's no Netscape. There's no Excel, Word. There's no Lotus 1-2-3.

Today I've only really seen four killer apps on a phone... email, SMS, mobile web browsers, and GPS. Facebook would be a marginal fifth. Everything else is stuff that frankly most people would barely notice if it was missing.

The killer app for both platforms is the web. This is fairly obvious if you remember the original iPhone had few apps and was widely considered a poor phone (or cameraphone).

The iPod was the third leg of the original announcement but it's hard to see that as a killer app when an iPod was available stand alone.

I'be never heard exclusive be a necessary feature of a killer app before this. It's more like a console fanboy bragging point and for technically comparable platforms it often seems driven by kickbacks from the platform owner. Even then it's usually just time limited exclusivity.

What constituted a Killer app 3-4 years ago, when the iPhone hit the market, probably isn't one any more. At the time, decent web browsing on a mobile truly was a killer app because none of the alternatives (RIM, Palm, etc) had decent web browsers. So it was essentially exclusive - if you wanted a great web browser on a phone, the iPhone was your choice. Now it's basically a commodity feature, since all of the major mobile platforms have a workable web browser.

"Exclusive" is maybe too strong a word, since it implies that no other platform will ever have that Killer Feature / App, e.g. Halo. But if only one platform has it when you need to buy, it's for all intents and purposes an exclusive.

Gruber seems to make the assertion that a platform only has worth based on its exclusive "killer apps".

This is a straw man. Android's lack of much in the way of exclusive native software is novel among the few mainstream software platforms over the last few decades.

Maybe this state of affairs is interesting, and maybe it isn't — we're dealing with a small sample of comparisons, the web is an extraordinary leveler, and to the extent Android has a shared lineage with Linux and its developer community, exclusivity is an alien concept anyway.

But pointing out this historical anomaly isn't tantamount to claiming Android has no worth, as Gruber takes pains to mention. There's no need to assume a defensive crouch.

"Android's strength is not its suite apps, it's its flexibility. It runs on an enormous range of hardware, and doesn't require that you buy into a single vendor's toolchain to use it."

Gruber's point is that the average user doesn't care about this. They don't care that the OS that runs on their phone can be used on a robot or a microwave. They just care that there is some awesome app that will motivate them to buy a certain phone, and he's saying Android doesn't have that awesome app.

But, like you, if I wanted to use a phone as the brains for an autonomous drone, I would use an Android phone. But this is an edge-case.

How about things like Launcher Pro, Swype, SwiftKey, and AppBrain?

None of those apps actually do anything. None of those are apps that help me cook, or tell me where the bike shop is, or let me send a photo of my dog to my parents, or give me something fun to do for 15 minutes on the bus.

I think that's Gruber's point. Android is full of apps that either...

A) are for people who think fucking around with their computer/phone is a worthwhile activity

B) are essentially ports from iPhone

or C) suck

"Gruber seems to make the assertion that a platform only has worth based on its exclusive 'killer apps'. He goes wrong there and never really recovers. Android's strength is not its suite apps, it's its flexibility."

It is baffling how you managed to read the article and still evade his point even though he addresses your rejoinder directly. Gruber says,

"At this point, I’m guessing, Android fans are ready to exclaim that the fact that Android supports things like home screen replacements (or other system-level tools, such as touchscreen keyboard replacements) — and that iOS does not — is precisely why they prefer Android, and/or consider iOS to be an unacceptable toy, or what have you. But, again, that’s not the argument I’m making. I’m talking about third-party developer exclusives — and the only ones Android has are ones that Apple doesn’t want."

Your opening salvo is baffling, because Gruber says, "I’m not saying Android is in trouble. The opposite, in fact: I think it’s going to continue growing — in terms of handset sales — despite this. And maybe as Android handset sales grow, this situation will change, and developers will start creating exclusive killer apps for the platform, drawn by the size of the market."

His main point is that Android is currently weak as a software platform. People don't choose Android because of its superior applications but because of its perceived flexibility/hackability. Developers don't see the Android app market as particularly profitable. The opposite holds true for iOS. He describes this as being a new phenomenon, the "app console market" which is a market that Android currently doesn't feature in.

We can extrapolate that if the situation with Android doesn't change, then Android could become the new Symbian, dominating a quasi-category of feature-smart-phones but with iOS and WP7 taking the more profitable segment of the market.

>People don't choose Android because of its superior applications but because of its perceived flexibility/hackability. Developers don't see the Android app market as particularly profitable.

I think you've hit the crux of the flaw with Gruber's argument: developer profitability does not have much bearing on UX as those in the proprietary app developer community might like it to.

It's not developer profitability per se but the appearance of developer profitability (among other things) that I think he was trying to get at. He says,

"Developers complain, not without merit, that the iTunes App Store is rigged toward low-priced apps. But the Android Market seems rigged toward no-price apps."

He posits that there is a significant qualitative difference between the most popular, exclusive, third-party applications in the iOS market and those in the Android market with the former providing a superior experience. Are you saying that this isn't true? If we admit it is true, then how do you account for it?

"Turn the table and we could be here all day running down the list of high-quality, interesting apps which are exclusive to iOS."

Thing is, the "Cupertino Grand Unified Vision" usually syncs up better with the needs and desires or the general population.

Some people don't want to have to configure anything to start using it or to start being productive with it. Apple's desire to focus on these types of people is obvious when not only analyze the type of features they add, but how they're implemented.

> the "Cupertino Grand Unified Vision" usually syncs up better with the needs and desires or the general population Well, hackers are hardly "the general population".

I think a meaningful comparison would be to the situation of home computers (C64, CPC, later Amiga and Atari) and consoles (NES, later SNES). If you wanted to play games, consoles offered a much better experience. In contrast, home computers could be programmed using BASIC and Assembler (and, somewhat later, C), which made them much more appealing for people who wanted to explore the possibilities for themselves.

If the general population is able to feel in control of their equipment without opening a shell on it and exploring what's in there, or being able to scratch little itches they have, that's totally fine with me.

excuse-me. last time i checked i couldn't even use the ssh client that is installed by default in my phone.

pretty much all the good things you mention are for rooted phones. The stock one is just as dumb as an iphone.

Android could be the PC-revolution all over the mobile industry. but it choose to start with baby steps :(

The only thing I mentioned that requires root was the FM Radio. I intentionally didn't mention root-only apps, given that your average user won't be running a rooted phone, but if we want to get into that territory, SetCPU, JuiceDefender, and Titanium Backup are all incredible apps that have no peer in the iOS ecosystem.

Hmm, which things are for rooted phones? Mine's unrooted and I have everything he mentioned except FM listener (I don't know what it is).

It's a part of the Cyanogen 6.1 ROM which enables me to listen to FM radio via the dormant-on-stock FM transceiver in my Nexus One. I specifically denoted that it was available via Cyanogen because it is the exception.

Ah, my Desire has a Radio app, so I don't need to root it for that either... Do you know if there is a compelling reason to root an Android phone? I haven't found anything I can't do yet, and the phone is close to supplanting my computer (especially with Swype/Swiftkey, K9 email and Skype/Meebo).

Complex SSH tunnels.

Better 2g fallback.

Touch screen fix, for nexus.

Hosts file editing.