IPhone OS 4.0 unveiled, adds multitasking, shipping this summer
engadget.com
engadget.com
Apple can monitor exactly what resources are being used, have timeouts for api calls, etc. to maintain perf. If you do regular multitasking (just split up cpu time between the top N apps) you have no idea how efficient apps are being, or why they are using cpu. You can't easily prioritize certain activities over others (i.e notifications vs. voip vs location). You can't optimize everyone's calls with core library functions (each app would re-implement their background tasks, perhaps poorly).
When people say they need multitasking they really mean "streaming music/skype/background notifications", like when they say they want flash they really mean they want video. We have to be careful not to just give people the faster horse.
One huge annoyance with the current platform is that remote desktop or ssh apps disconnect when you switch out of the app. It's one of the reasons i'm thinking of holding off on the iPad for another year (another OS release).
Given Scott's comment about how there is no "close" in this pseudo-multitasking world, I'm a little worried that the iPad will stay useless for me for the foreseeable future :-)
I have to imagine (cough) that there are similar facilities for any kind of IP connection.
The cheap way to pay them back for Android would have been to banish Google from the iPhone. Instead, they're attempting to transform mobile advertising into something where Google's technical strengths are worthless, while their weaknesses - user interfaces, platform fragmentation, and apps - are a serious liability.
Worse, Apple is right - mobile ads do suck. Nobody clicks on them, because launching a web browser from an app is incredibly disruptive on a handheld. This is an 'aha' moment akin to when Google launched adwords. If iAd is easy to develop for and deployed across millions of devices, Google's ambitions for mobile are in serious trouble.
This will be big because game makers will buy these by the boatload. I'll go on record right now predicting game ads will be over 80% of what you see on these.
This is why we've never touched it. On Facebook, we just have to make our users want to invite other users. We don't have to constantly release new versions and incentivize users to download them just to get ourselves back in the top 20. There are some tricks, of course, but they're less of a detour from the desired path, if that makes sense.
iAds will make success on the platform a numbers game. Every free app will run it because it can't pay worse than the current options, even with the 40% rake. Paid apps will break out the spreadsheet and almost certainly find a CPA below the app's price for a paid one. My totally off-the-cuff guess is that with a really engaging iAd you'll end up paying $1 to sell a $2 app, at which point it's all about volume. Selling a million like that might be doable, and now we're talking money in line with what Facebook apps can make.
Still, the problem will be that the most successful iPhone apps make about the same amount lifetime that the most successful Facebook apps make in a month. But the odds of succeeding may now be high enough to make it a worthwhile play.
Maybe the App Store needs iAd/GameKit integration of a 'like'/'recommend'-to-friends feature.
Still, I imagine if Apple made an opt-out method of automatically friending all of your email contacts who have iPhones (or at least an opt-out method of seeing them) it would be popular.
Actually, I'd probably be influenced more, since I know some Facebook games force you to recommend it in order to get access. But if this game center, like XBox Live, can just show me what my friends are actually playing... that's the best endorsement they can give me :)
If the iPhone loses its dominance in smart phones, this move will have done nothing but antagonize the most obvious heir to the throne. Like nearly everything else Apple does, iAds will have a monopoly on a small market. Meanwhile, much of the iPhone's greatness depends on Google's own products....
Maps is actually a bigger deal to me --- but maps is also easier to replace.
However I found maps.bing.com to be quite enjoyable and in some aspects superior to maps.google.com -- e.g. bing's "birds eye view".
Let's take the whole mobile phone market as measurement. The iphone is still only used by the vast minority, yet it feels like apple users are caught by all that marketing and buzz around apple they reject reality.
Yes, of course, the iPhone has the appstore, it's good hardware and software and shiny and usable and so on. But the reality is STILL that it's a minority.
The reality is that the big old mobile companies have still enough money, resources and potential customers to get. It's just not that Nokia is making all the fuzz around this as the press does about apple. Steve Jobs farts, press coverage instantly. It's sick.
We don’t know much about the advertising stuff. We’re
learning. We tried to buy a company called AdMob, the
biggest mobile advertising company. But Google came
in and snatched them from us. They didn’t want us to
have them. So we bought another smaller company,
Quattro Wireless. They are teaching us. But we are
babes in the woods. We are learning as fast as we can.
That said, iAds will be successful just because its Apples advertising platform and very high profile.
from Gizmodo live coverage:
Q: And can people use stuff other than iAd and Game Center?
Steve: Yes, you can use any service.
Everything Google's done in the last few years has seemed bent on extending their search ads monopoly to mobile devices. If success was counted based on devices shipped, Android was doing pretty well. iAd is a serious setback if it turns out to be more profitable than mobile browser ads.
It'll be harder to implement things like keeping the user in the app while viewing the ad or adding video/photos.
What if using Apple's ad network is few lines of code? No library to import, or have a 3rd party dependency on your app.
Most mobile ad networks doing in-app advertising switched to a web view within the application months ago. In addtion, clickthrough rates are generally pretty high - although I blame 'butterfingers' for a lot of this. (Tracking post-click conversions is hard to do on the iPhone platform but when we've done it I haven't been horribly impressed.)
Also, I now love my blackberry much more than I ever did the iPhone. Apps I use are available (maps, yelp, foursquare, pandora), and the multitasking and the advanced features of the OS just keep surprising me. I can use my blackberry to connect to a VPN (it's built in). Coming from the iPhone, I feel like my handcuffs have just been taken off and I'm now free to use my phone the way I want to.
Android's browser uses a long tap for cut and paste, so never had this problem.
The only way to bring up that dialog accidentally is to tap on the screen, keep your finger still for more than one second and then try to start scrolling. Seems odd for anybody to try doing that.
It's kind of the way of the world.
Apple is looking to establish and to extend their beachheads in both the smartphone and tablet spaces, and this is how you do that. You invest design and development and debug and test. Big-time.
Being a response of marketing and features for the Microsoft Windows Phone 7 and the HP Slate announcements, this looks quite potent.
And interestingly, this approach looks quite reminiscent of how Microsoft slid in a new (Windows) computing platform from up underneath its competitors way back in the 1990s, too.
"Now we weren't the first to this party, but we're gonna be the best. Just like cut and paste."
I think it's a fantastic idea to add Background * API if you can't provide full blown multitasking. Covers 80% of the use cases.
I'm not sure if I'm more amazed at how something as inane as this basic operation is touted as a huge feature differentiator, or that the CEO of a multi-billion dollar a year computing company, who's been in the business literally from the start, talks about it with breathless urgency like it's the second coming.
Copy/cut and paste is not a major feature, it should be buried among a list of minor enhancements. It's like the equivalent and releasing a new keyboard with a shift key on either side!
No, I'm not saying that it doesn't have certain considerations necessary to make it work well. We can't even get proper clipboard support in Ubuntu these days. So I'm not saying it's really that simple.
Yes, I am saying it's not a big feature. Again, it's like VW coming out with a new Jetta, and one of the specs they touted was "Now with tires!!!"
You are right, in a multitouch world, I have far more options for how to do this.
1) Has trouble finding optimal solutions given numerous choices?
2) Has found the most optimal solution?
1. It's unlikely that anybody has found an optimal cut+paste system for multi-touch devices.
2. The broader range of imaginable ways of implementing cut+paste means that the first solutions to market are relatively unlikely to be considered "good" once the various options have been more fully explored.
3. The large number of ways to implement cut+paste means that there will almost certainly be a significant difference in the quality/usability of different solutions. These disparities mean that the mere presence or absence of a cut+paste solution is not enough to judge whether a product does a good job of fulfilling the need for cut+paste. Simply asking whether a product has a cut+paste system is not helpful for comparing multitouch devices, because not all methods are equal, and some may be virtually unusable.
Apple has sold 50M iPhones. Microsoft has sold roughly 30M Xbox 360s. This summer, before the launch of Windows Phone 7, Apple will have the world's largest gaming network.....overnight.
How is that not a bigger deal than multi-tasking?
I agree - the gaming network will be a big deal, especially since it will provide an alternative channel for developers to market their games. It gives the iPhone the social marketing aspect of Facebook games.
"Social gaming network" doesn't mean anything to me. What am I supposed to do with it? How will it change ... anything ... for me?
Being able to keep the user in the app while showing them high-quality ads is a huge advantage. And most likely the source of the recent, supposed kerfuffles between Apple and Google executives.
(edit, the ads are basically mini-apps, apparently in HTML5)
Admob, the iPhone advertising platform that Google bought had the same split as iAd, 60/40 for CPC and 50/50 for CPM on banners. There are others that pay higher, more here: http://mobiforge.com/running/story/mobile-advertising-how-mo...
Now if they can figure out how to target this stuff AND deliver high quality... damn.
>Q: Will there be an ad approval process? A: (Jobs... super long pause) Um... well there's going to be some process... but these people are paying to run ads. So, I'm not sure it'll be anything other than a light touch.
So that implies that they won't be too selective about the advertisers, though they might just charge enough to ensure that most advertisers are high quality.
"There's got to be a developer kit for making these HTML5 iAds, right?" (http://twitter.com/gruber/statuses/11835821330)
Even before this I started thinking that an online app that made HTML5 banner creation cheap & easy would be huge. Now, jeez, seems like a no-brainer.
We brainstormed this idea a couple months ago but put it on the backburner because of still poor HTML5 support across browsers, etc.
Fall is a ridiculous deadline... but, admittedly, also an inspiring one ;)
It all depends on how Apple structures the ad submission process. From Steve's pause before responding to the "will ads need to be approved?" question, I am going to guess they haven't entirely fleshed it all out, yet.
They're now competing in Google's main cash cow market.
Mobile advertising is growing, but that doesn't make it a big part.
The future of small, handheld computing devices is coming.
It's not whether or not you are tethered to a desk, but what OS you run and what your relationship to your computer is. I have been using notebooks for the past 5 years and I would not call me mobile in any but the most literal sense.
Desktop computers will remain popular for work in information based economies like ours, but there are billions of people out there who will access the Internet soley from mobile devices.
(I'm currently sitting in front of my computer but am writing this on a mobile device.)
Perhaps Apple is aiming to be a boutique ad net like the Deck or Federated.
Also - The FTC seems poised to block the Google/AdMob deal (http://news.google.com/news/search?aq=0&pz=1&um=1...). Conjecture on conjecture, but if this happens, it would indicate a much more aggressive regulatory environment (makes sense given changing administration), so Apple might face issues of its own. As they said during their presentation, they control ~60% of the mobile browser market....
10:11AM "iPhone has a 64% mobile browser use share... everything else added together is half the iPhone."
Now, we shouldn't need a stylus or task manager: http://www.engadget.com/2010/04/08/jobs-if-you-see-a-stylus-... The problem is, I want a stylus on something like the iPad - for some operations, writing is a more familiar metaphor than finger-painting. Likewise, their app-switcher is 90% of a task manager - all they'd need to do is add closing background apps (a feature we don't need?).
I know Apple is the master of opinionated design / "not listening to the customer" in order to provide something better than what the customer thinks they want. However, as a technical user, sometimes they outsmart me and sometimes they fail. In the next year, I'll purchase a tablet device (most likely Android-based, maybe the Notion Ink Adam... if it lives up to its hype). I don't expect for it to be as elegant as an Apple product, but it will do more of the things I want (and not insult me).
Likewise, they never said users didn't need copy/paste, they just shipped the first few versions of iPhone OS without it while they were still working on it.
It's a bit of a stretch to say that AdMob was a "horrible" purchase, though. They had to do something to move into the space. The only question is whether the iPhone is dominant enough in the market to make Apple's strategy successful over the long term -- if Android takes over the mobile market, Apple will only have a monopoly on a niche ad network.
All I can say is LOL! sorry.
edit: forgot to say it's a quote I saw on arstechnica live coverage.
Hopefully this isn't going to allow apps to just run any random equivalent of a tight loop of their own (memory chewing CPU and battery consuming) design.
I'd also like a reverse chronology notification history that pops up from tapping on the click.
To even call it multi-tasking is a joke.
What's so bad about that?
No, you don't. Just because an app shows up in a process list doesn't mean it's consuming resources. (Other than memory, and background processes are automatically killed as needed to free up RAM).
The reason is that intermediate users who know just enough to be dangerous think that they're necessary, when they aren't: http://droidtalk.net/should-i-install-a-task-managerkiller-o...
"Is it really necessary to be notified about every new tweet or status update? Reduce the refresh times on your social networking applications and reduce the number of items you’re notified about. Reduce the number of widgets you have on your home-screen, instead replace them with program shortcuts, you’ll still have quick access to your most used programs without the constant battery drain caused by continuous data pull."
Update — services offered:
Background audio
VOIP
Background location
Push notifications
Local notifications (don't require the server)
Task completion (e.g. Flickr's upload can continue
in the background)
Fast app switching (storing/restoring app's state)I agree. I'm referring to the defenders. The people that were actively pointing to reasons why multitasking for the iPhone/iPad were bad for non-technical reasons (screen size, the UI, they didn't need it, etc.) There were lots. Every time multitasking was brought up, they would offer countless reasons why multitasking was bad.
Which was all just a cheap way to defend it's lack of multitasking. Multitasking was never a bad idea. It was always a missing feature. Regardless of the reasons for it, it's clear that they've finally overcome the problems.
Their inability to do so until now shouldn't have shielded them from criticism early on.
But it's not. And you can disagree, but you are wrong. The defenders I'm referring to were supporting not having multitasking; or, better said, were touting lack of multitasking as a feature. Their were other defenders who were for multitasking, but against doing it poorly. I'm referring to the former, not the later.
The lack of jailbreak/Android-style multitasking is still a feature. The fact that they took the time to implement it as carefully-designed background services means I don't have to deal with process killers and don't lose my battery to moronic apps that busypoll on GPS.
Or, as they said: "In multitasking, if you see a task manager, they blew it. Users shouldn’t have to ever, ever, ever think about that stuff."
"The lack of multitasking is still a feature."
You added qualifications for a reason. Because they say completely different things. You seem to want to suggest I'm arguing against the first. I'm not. I'm arguing against the second.
The first feature is an app ecosystem that doesn't require me to babysit processes.
The second feature is the ability to do things in the background (or, more precisely, do things with more than one application at a time).
I don't care what you call them, but that the iPhone has always had the first feature, and talking it up isn't crazy. I don't think that it makes up for not having the second feature, but I still recognize it as something that makes my iPhone experience better.
I equate it to the result, not the implementation. I don't care how it gets done, as long as it gets done without ruining the rest of the experience.
Not too long ago, we were reading iPad reviews disregarding the lack of multitasking as unneeded. Comments would fill up with mac-addicts defending the lack of multitasking suggesting things like "How can you see two windows on an iPhone" and other crazy things like that. People questioning the needed for multitasking, as well as suggesting that no multitasking was a feature, not a missing feature.
So now everyone is looking at Apple's MT and saying "WoW! It's so amazing!" and the rest of us are saying "It's MT done for the iPhone, okay." It's what we've been asking for, and what so many people were saying to us "It's bad" or "It will hurt the experience" or "MT on the iPhone is a waste."
Maybe you weren't one of those people, but I had enough respond to me saying these very things. And it's just nice to know that Apple, in the end, agreed with me and people like me.
One: battery. They solved that pretty well by not doing true preemptive multitasking (I assume that it isn't preemptive, except maybe for background music). This is likely the right call too. Give apps access to the services they need to do what they want in the background, and keep them suspended the rest of the time.
The second is user interface. I thnk that their task switching solution is probably good too. It looks a lot like palm's, but with an apple flair.
You're right though... any previous defenders who gave other reasons were just deluding themselves. I think that battery life was the main factor. Once they figured that out, the rest became how to make it easy.
There were, however, people arguing against multitasking as if their was only one way to implement it.
And just like there were lots of people saying multitasking wasn't needed at all, there were lots of people saying they wanted multitasking even if it killed their battery life, they wanted the choice. I'm glad Apple didn't listen to either of them.
I guess you'd say I was one of them, but I'm still going to say I was right:
http://news.ycombinator.com/item?id=1243742
There are both technical and UI reasons why multitasking as we've known it is hard to do on a mobile phone; those issues constrain what you can do when implementing it, and it certainly looks like Apple's done the best it can within those constraints (which looks pretty good at first glance, by the way).
Do you think the lack of multitasking is a feature? Apples introduction of multitasking... harms the platform?
"but I'm still going to say I was right."
Apple disagrees. Multitasking is a feature. The lack of multitasking wasn't.
You seem to be caught up on the implementation, not the results.
The "discussions", if I can be generous and call them that, have been ridiculous and transparent and usually runs like this:
Normal Person: The i<product> has no multitasking, this is a sorely missing feature
Rabid Fanboy: This omission is actually a "feature". Since Apple never does anything wrong, it's clear that this is actually just better for us. Besides nobody uses multitasking anyways, what could you possibly want to use it for?! Only crazy people want a computing platform that can run more than one thing. Besides Apple needs to keep the user experience perfect like it is (oh it's so orgasmic). And double besides, it already does do multitasking, Apple just uses it for their own things and keeps outside developers (who don't know what they are doing anyways, those idiots) from mucking up the precious user experience and draining the whole battery down in 5 minutes.
Normal Person: That actually doesn't even make any sense. There are plenty of examples of mobile platforms with multitasking that don't have any of the feared problems you are talking about, Palm, Android,...
Rabid Fanboy: THOSE SUCK! SUCK! <hits own head with closed fist over and over> <pause and takes a deep breath as if trying to explain some critical thing to a 5 year old> None of those platforms are the iPhone. NONE OF THEM.
Normal Person: Again, that makes no sense. By all appearances, the fact that this feature is supported just fine elsewhere, and not on the iPhone actually makes Apple look a little inept.
Rabid Fanoy: <head explodes>
None of the "rabid fanboys" have been claiming that multitasking will reduce battery life down to 5 minutes. Most haven't claimed that nobody uses multitasking; they just say that most normal people can get along without it.
The "normal people"/people who love to troll Apple users/you like to claim that Palm, Android, etc. have great multitasking systems, while ignoring the fact that those platforms do have noticeable UI lag, and generally worse battery life or performance. If you're trying to make a sound argument against Apple's stance of offering limited or no multitasking, you need to acknowledge the tradeoffs and then show that the stuttering, longer page loading times, or lower battery life are worth the overall productivity increase of multitasking. You can't just claim the downsides don't exist.
Do you even read this site?
Go look back at the last 6 months of posts on the iPhone and iPad. Nearly every single thread brings up this issue, and the fanboys come out in droves to try and explain away this lack of functionality -- most typically with nonsense arguments like I posted above:
- GUIs for multitasking are too hard!
- The battery will drain!
- The GPS will be pinging all the time!
- Nobody uses multitasking!
- I don't want a phone that needs a task manager to kill background apps! (I bet a whole Nigerian Dollar that we'll see task manager apps within 6 months in the app store)
- etc. see http://news.ycombinator.com/item?id=1240366
Now Apple comes out with multitasking (limited or not, it doesn't matter, just because Pandora can now play music in the background doesn't mean that Apple pumped their API full of magic pixi dusts where suddenly that won't eat away at the battery.) So now what do we see fanboys carrying on about?
<Backtrack backtrack backtrack, revision, revision, revision> Fanboys: We never said that multitasking wasn't important! We just said that mumble mumble. Besides it's not lack these are just background apps, there's a special magic API that developers can use that's so carefully designed and built by dwarves in their mountain code smithery out of mithril and fairy dust that it doesn't fall afoul of any of the problems we see everywhere else, like UI LAG!
> The "normal people"/people who love to troll Apple users/you like to claim that Palm, Android, etc. have great multitasking systems, while ignoring the fact that those platforms do have noticeable UI lag, and generally worse battery life or performance. If you're trying to make a sound argument against Apple's stance of offering limited or no multitasking, you need to acknowledge the tradeoffs and then show that the stuttering, longer page loading times, or lower battery life are worth the overall productivity increase of multitasking. You can't just claim the downsides don't exist.
My point is now proven. See your own comment. The OS upgrade and the availability of an API for background services doesn't suddenly free Apple from these constraints either. I think somebody here even made some claim about how other phone makers had a problem with location based apps draining battery by constantly pinging the GPS. Apparently he didn't read the press announcement where Apple allows for just exactly that function in background services.
Either fanboys said multitasking was a bad idea on a phone or they didn't.
And if they did they listed a bunch of reasons why.
None of those reasons are eliminated with this variant of multi-tasking.
sigh It's like fighting with people who suffer from continuous short-term memory loss regarding things they just said.
The fact that Apple now has an implementation does not invalidate them.
Does anyone know if this means any O/S object that broadcasts notifications via NSDistributedNotificationCenter (http://developer.apple.com/iphone/library/documentation/Coco...) will now work?
Previously there was no way to receive notifications from any other app or O/S process.
It was never a feature, it was a choice to prolong battery life. We'll see if the other side of the tradeoff is worse or better on the whole once it is put in everyone's devices.
Also, I already have double-home button mapped to bring up the iPod controls and go to search. I don't like that that'll be going away.
Though I'm not surprised that the implementation they're going with is allowing apps to have smaller background services instead of full on multitasking.
I'm gonna let them finish before saying whether it's the best multitasking phone of all time though.
edit: That is if you believe the real reason was for performance. Personally I think it was half that, and half that they hadn't come up with a good way to present it to the user.
edit 2: I think this sums it up exactly: "(...) in multitasking, if you see a task manager, they blew it" - Jobs in Q&A.
http://arstechnica.com/apple/news/2010/04/iphone-os-4-specia...
I've always been disappointed in iPhone bluetooth support. When we finally got A2DP they only implemented a half-assed AVRCP profile so we couldn't skip tracks with our A2DP headsets. Frustrating. Hopefully version 4.0 of the iPhone OS finally implements this, but I've yet to see confirmation.
in the new license agreement, did they just ban apps generated by flash cs5?
This is just a way of forcing people to upgrade to the latest non-jailbreakable hardware.
Either way, it's irrelevant. I didn't buy my 3G in the anticipation of having multitasking. I bought it because I wanted what it was then. I'm upgrade-eligible soon enough, and happy to have a reason to buy a 4th-gen iPhone.
I also want a 4th Gen iPhone but not if its not jailbreak-able, there are too many missing features that can only be enabled by jailbreaking.
I'm just reacting to the implication that there is something shady about them not giving the 3G a feature.
Someone else pointed out that in the case of multitasking they cannot guarantee that a 3G can handle the general case, that may be true, we shall have to wait and see.
Imagine having to reboot your phone when you dont have access to your laptop ==> No phone till you get home.
If an un-tethered jailbreak becomes avaiable I would upgrade in a heart-beat.
"Forcing" only if those users actually want multi-tasking. I'm guessing a large percentage of the user base won't covet this feature.
http://www.blogcdn.com/www.engadget.com/media/2010/04/iphone...
http://live.gdgt.com/2010/04/08/live-iphone-os-4-0-event-cov...
My impression is that the kernel just suspends the process and keeps it in RAM. I'd expect there to be a notification that the app gets before it's suspended so that it can free any memory that's not necessary for preserving its state.
It will be rare that you're not running two things.
Other than pandora, which I agree will be awesome in the background, what in that use case requires multitasking?
Open Maps, see a place, click the link to switch to Safari to see it's website, check reviews in yelp. Nowhere is more than one thing happening here. Ditto with googling something, answering the phone, switching to email, switching back to Safari. Everything here is saving state so when you launch it it comes back to where it was.
THIS is what people mean when they say that 90% of use cases are covered currently. The background services are great for pandora, etc. but let's not make them out to be more than they are.
foursquare app open to auto-checkin (or a Waze/OpenStreetMap editor recording GPS traces), IM app open to receive Skype calls, SSH connection open for quick access to top on your new widget-server.
I admit that none of those things need to be running simultaneously, but the fact that they are running is pretty nice.
> "but there are some thigns they will not run because hardware doesn't support them, like multitasking"
(via Ars live blog)
...so the 1G phone would presumably be out as well.
edit: sorry Only 3GS gets multitasking, you probably were asking about other OS4 features. no idea.
Wait, multitasking? I thought we've just been having the discussion about how Apple is awesome for not having multitasking followed by a long bulleted list of apologists talking points?
- You can't change the battery
- It doesn't support 3D glasses and 4K cinema resolution
- No 20MP camera with flash and replacable lens
- There is no noticeable lag when running applications in the background, so it's actually not "true multitasking". You can't run 100 apps in the background.
- Can't install apps without jailbreak (or dev license)
- Doesn't have a nanoSD card
- No USB portWe were having the discussion "plausible reasons why Apple might not have implemented multitasking yet, and a pragmatic understanding and acceptance based on the wonderful nature of the iPhone anyway and hope for future multitasking done in a way that wont ruin that".