Forgot your password?
typodupeerror
IOS Apple

Apple Handcuffs Web Apps On iPhone Home Screen 298

Posted by CmdrTaco
from the evil-or-oops dept.
SF Polack writes "On Apple's iOS 4.3, HTML5 and JavaScript apps are running significantly slower when they're run from the iPhone or iPad home screen rather than Safari, and the OS is hindering the performance of these apps in other ways. The end result is that it's harder for web apps to compete with native iOS app sold through the App Store, where Apple takes a 30 per cent of sales."
This discussion has been archived. No new comments can be posted.

Apple Handcuffs Web Apps On iPhone Home Screen

Comments Filter:
  • by Shikaku (1129753)

    Doesn't iOS have backends that needs tending, like I don't know, being able to receive calls and mail?

    • Re:Uh. (Score:5, Funny)

      by russlar (1122455) on Tuesday March 15, 2011 @01:07PM (#35493432)
      Try holding it differently.
      • by recoiledsnake (879048) on Tuesday March 15, 2011 @01:30PM (#35493768)

        This is the opinion(not mine, I know this will be downvoted regardless) of this very interesting and detailed article which I wanted to post.

        http://blog.millermedeiros.com/2011/01/ipad-is-the-new-ie6/ [millermedeiros.com]

    • by technomom (444378)
      ....and handle seasonal time changes?
    • This has to do with if you go to the page from the home screen (effectively clicking a shortcut that takes you to safari) versus going to safari, then the page. Nothing to do with the available capability.

      In both situations, you end up with the same result (page running in safari). When you use the home screen shortcut, you get less performance.

      As to backends, if you're phone app requires CPU power to *wait* for calls, you're doing it wrong. (memory I can understand. actively polling a hardware si
    • Doesn't iOS have backends that needs tending, like I don't know, being able to receive calls and mail?

      If that's a gay Apple joke you're making, I am sure most are not getting it. /joke.

  • by linuxci (3530) on Tuesday March 15, 2011 @01:07PM (#35493442)

    Why would it be about 30%, most web apps are free and 30% of zero is zero. Apple allow free apps in their store.

    This bug only occurs when you launch a web app that contains a meta tag of name="apple-mobile-web-app-capable" content="yes"

    If your 'web app' is just a shortcut to Safari on your homescreen then you won't see this bug.

    Basically this web app meta tag launches the app fullscreen without any Safari chrome. To the user it looks like a separate app rather than it's running in the browser.

    The slow behaviour is just using the iOS 4.2 JavaScript engine. It's possible that this is either an oversight or that Apple deliberately kept the old JavaScript engine for web apps in case it broke functionality that the app was depending on.

    We'll see in the coming weeks I'm sure.

    • by interkin3tic (1469267) on Tuesday March 15, 2011 @01:17PM (#35493598)

      most web apps are free and 30% of zero is zero.

      You must be new here. The correct way of saying that would be "Apple takes 100% of the royalties of most apps!"

      Write that up, post it on a blog somewhere, and submit it, quick!

      • by MightyYar (622222)

        The correct way of saying that would be "Apple takes 100% of the royalties of most apps!"

        It's worse than that... they take an arbitrary amount and they never inform the developer.

    • Apple? Do something architecturally inelegant for the sake of backwards compatibility? Are you sure that you aren't mixing up your Steves here?
      • I guess you've never heard of Carbon, Rosetta, or the Mac 68K emulator? (Although admittedly that last one is non-Jobsy.) Apple has quite a history of subsystem ghettos.

        <sarcasm>What this really means is that we can expect web apps to be phased out in two to three years.</sarcasm>
    • Not anymore.... (Score:4, Interesting)

      by recoiledsnake (879048) on Tuesday March 15, 2011 @01:18PM (#35493608)

      >Why would it be about 30%, most web apps are free and 30% of zero is zero. Apple allow free apps in their store.

      Not anymore if it involves any money exchanged between the user and the app provider. Now Apple is forcing (users of) subscription services like Amazon and Netflix to pay up 30%. ( an extra 43% to the user). It's curtains from June.

      http://www.washingtonpost.com/wp-dyn/content/article/2011/02/19/AR2011021902399.html [washingtonpost.com]

      Free app Readability already got banned for this.

      http://blog.readability.com/2011/02/an-open-letter-to-apple/ [readability.com]

      Free Sony e-reader app banned:

      http://voices.washingtonpost.com/fasterforward/2011/02/apple_bans_sony_e-reader_app_a.html [washingtonpost.com]

      • by linuxci (3530)

        >Why would it be about 30%, most web apps are free and 30% of zero is zero. Apple allow free apps in their store.

        Not anymore if it involves any money exchanged between the user and the app provider. Now Apple is forcing (users of) subscription services like Amazon and Netflix to pay up 30%. ( an extra 43% to the user). It's curtains from June.

        http://www.washingtonpost.com/wp-dyn/content/article/2011/02/19/AR2011021902399.html [washingtonpost.com]

        Free app Readability already got banned for this.

        http://blog.readability.com/2011/02/an-open-letter-to-apple/ [readability.com]

        Free Sony e-reader app banned:

        http://voices.washingtonpost.com/fasterforward/2011/02/apple_bans_sony_e-reader_app_a.html [washingtonpost.com]

        We're not talking about these apps which were native apps and not web apps.

        I'm just saying that most webapps are free and so it doesn't affect apples profits whether they're distributed as webapps or through the app store.

        Remember, when the iPhone was launched web apps were the only way to get your app on the phone. The app store came later.

        • I think that was a counterpoint to this line:

          Apple allow free apps in their store.

          Readability was a free app before it was pulled with the new rules.

          Anyway, Readability made a HTML5 app after they got rejected.

          http://techcrunch.com/2011/03/09/readability-html5/ [techcrunch.com]

          They can't be too happy with this news and might be thinking it is intentional to close the HTML5 loophole for subscription apps.

          • by s73v3r (963317)

            Its not really free if they need a subscription.

          • by Karlt1 (231423)

            Readability was a free app before it was pulled with the new rules.

            You mean the free app that charges authors 30% for doing nothing but allowing them to be viewed on the Readability platform.....that business tactic seems awfully familiar....

            They can't be too happy with this news and might be thinking it is intentional to close the HTML5 loophole for subscription apps.

            Oh the solution that Apple gave developers before the app store came on-line and that everyone cried foul about?

            • Readability was a free app before it was pulled with the new rules.

              You mean the free app that charges authors 30% for doing nothing but allowing them to be viewed on the Readability platform.....that business tactic seems awfully familiar....

              They can't be too happy with this news and might be thinking it is intentional to close the HTML5 loophole for subscription apps.

              Oh the solution that Apple gave developers before the app store came on-line and that everyone cried foul about?

              Umm, users can opt to pay for ad free content, 30% of which Readability keeps. Perhaps you can provide a service like this if you think it's easy and Readability shoudn't charge or charges too much. The content authors are utterly free to offer this kind of service through any other app or their own app or website.Where do iUsers and Readability go if the App Store doesn't carry it? There are no ways to offer apps directly(due to strict DRM) or through another store like the Amazon store like in Android.

              A

        • Re:Not anymore.... (Score:5, Informative)

          by Dahamma (304068) on Tuesday March 15, 2011 @02:08PM (#35494302)

          We're not talking about these apps which were native apps and not web apps.

          Yes, that's exactly what we are talking about. The whole point of the article is that if these services now want to be on the iPhone, etc without paying that cut to Apple they will need to create a web app, and when they attempt to integrate them seamlessly using Apple's recommended method, they will run more slowly for no apparent reason.

          • We're not talking about these apps which were native apps and not web apps.

            Yes, that's exactly what we are talking about. The whole point of the article is that if these services now want to be on the iPhone, etc without paying that cut to Apple they will need to create a web app, and when they attempt to integrate them seamlessly using Apple's recommended method, they will run more slowly for no apparent reason.

            No dumbass we're not. Subscription services now HAVE to offer in app purchase as well as web store purchases at the same price. There is no reason to force them to be slower when using a web store.

    • Why would it be about 30%, most web apps are free and 30% of zero is zero. Apple allow free apps in their store.

      It also cost $100 or so a year to be in the app store.

    • by prockcore (543967) on Tuesday March 15, 2011 @01:41PM (#35493922)

      > Apple allow free apps in their store.

      Apple still gets money for that. $99/year to host a free app. If you stop paying the $99/year, Apple removes the app from the store.

    • by Snocone (158524)

      "It's possible that this is either an oversight or that Apple deliberately kept the old JavaScript engine for web apps in case it broke functionality that the app was depending on."

      No, it's a security thing. This was noted when people first found out about Nitro.

      "apparently iOS 4.3 features JS JIT. did they lift restriction from the kernel that prevented mmap-ing rwx memory pages? hmm."

      http://twitter.com/mraleph/status/43030240175468544

      So in 4.3 they've lifted it for Safari.app and only Safari.app. Presumab

    • by Dahamma (304068)

      The big problem is with web apps that allow purchases or subscription services.

      If you want to create an iOS app (even if the app is free) and it includes a subscription service or sells in-app items, Apple now wants a 30% cut of ANYTHING you buy from the device. Buy an online magazine subscription, or sign up for an MMORPG through your iPad? Apple wants 30% of the fees ONGOING. Want to buy eBooks through Amazon or Sony? 30% goes to them. I mean, come on, these days even the ultra-competitive brick and

    • by linuxci (3530)

      Note this bit in the same Reg article:
      All three issues also affect native iOS web applications that uses Apple's UIWebView API – i.e., native applications that tap the web in a big way.

      So it's saying apps available through the app store are also affected if they use the built in HTML renderer. Doesn't this show that it wasn't an intentional way to make apps look better, otherwise apps would have access to the new faster engine too,

      The Reg is just trolling for pageviews.

    • by Tharsman (1364603)

      FACTS?! How dare you!? This is SLASHDOT!!! Take your facts somewhere else, Mr.!!!!

  • After Jobs said in public Apple was committed to supporting HTML5?

  • I smell troll bait (Score:3, Insightful)

    by neosar82 (792049) on Tuesday March 15, 2011 @01:08PM (#35493460)
    I'm sure this is a bug and not by design as the OP's argument doesn't make much sense. Most native app versions of services that also offer webapps are free anyways. Apple gets to eat the distribution overhead for no 30% cut. Just sayin.
    • Re: (Score:2, Informative)

      by Anonymous Coward

      RTFA.

      It says (near the end) that the Apple mobile team has confirmed the issue and Apple engineers also said that it will not be fixed.

      It doesn't matter whether the new behavior was introduced deliberately or not, the negative effects are there. And Apple is saying they won't be fixing it. Therefore the article is not a troll.

  • I have not RTFA but I think the summary got it the wrong way round: If apps are running slower when run from the home screen compared to Safari, this would be an advantage of web apps as opposed to native apps. So if the article makes any sense, apps should run slower in Safari.
  • What? You can't install "web apps" on your home screen. Except for shortcuts that launch Safari, which would run with the newer, faster engine.

    AFAIK those "native iOS app sold through the App Store, where Apple takes a 30 per cent of sales." are precisely those that get launched from the home screen, which will be slow when displaying an embedded "WebView".

    Did I understand this the other way around or did the summary just twist facts about 540 degrees? Please correct me if I'm wrong.

    • You can save a bookmark to a web page on your home screen. If that web page has a <meta name="apple-mobile-web-app-capable" content="yes"> then it launches in a full screen (no status bar) chromeless window without the safari interface.

      Apparently it also runs slower if you do that.

  • by dlsmith (993896) on Tuesday March 15, 2011 @01:12PM (#35493514)
    The article suggests two explanations:
    • Apple can't stand to "lose" money to Web-based apps that it wishes would be sold in the store, so it is going out of its way to cripple them.
    • Apple introduced some new features in the latest Safari version, and didn't manage to get around to integrating those improvements into its web-based app launcher yet.

    Given that Web-based apps are typically free, I'm a bit skeptical about #1. But guess which explanation made the headline?

    • by microbee (682094)

      Its more about control than money.

      Web-based apps == less control for Steve == bad for Apple

  • by GFLPraxis (745118) on Tuesday March 15, 2011 @01:17PM (#35493590) Homepage Journal
    The cause of this has been discovered already; it's a software bug. iOS 4.3 has a new JavaScript engine. Websites launched from the home screen seem to be reverting to and using the old JavaScript engine from iOS 4.2. The article makes it sound like a conspiracy. I'm sure it'll be patched soon; I can think of no obvious reason to do this but give the same apps full speed if bookmarked within the web browser.
    • by H0p313ss (811249)

      The cause of this has been discovered already; it's a software bug. iOS 4.3 has a new JavaScript engine. Websites launched from the home screen seem to be reverting to and using the old JavaScript engine from iOS 4.2.

      And as a software professional that would have been my first guess, "Sounds like a bug to me"

      The article makes it sound like a conspiracy. I'm sure it'll be patched soon; I can think of no obvious reason to do this but give the same apps full speed if bookmarked within the web browser.

      It would appear that the anything iOS related brings out the tinfoil hat crowd. Time to up the meds guys.

  • by alvinrod (889928) on Tuesday March 15, 2011 @01:17PM (#35493604)
    From TFA:

    It's unclear whether these are accidental bugs or issues consciously introduced by Apple.

    So, they have no idea whether or not it's actually malicious, but they've decided to run with the story using an inflammatory headline anyway.

    According to Apple developers posting to the web, the speed issue has been discussed in the company's developer support forums, and one developer – the same unnamed developer quoted above – confirms with The Reg that multiple bugs have been filed on the issue.

    Developers are using proper channels to report what's most likely a bug and this is most likely a non-story as of the next minor update, but they've still decided to run with it anyway.

    Apple isn't degrading the speed of home screen web apps. It's boosting the speed of web apps in the browser. But in the long run, the effect is the same. And if this is a bug, Apple has yet to fix it.

    So, in fact, Apple hasn't intentionally hobbled anything, it's just that they haven't sped them up, possibly due to a bug, yet they're still going to run this story.

    This developer reiterates that if Apple didn't specifically introduce these problems in iOS, it's aware of them now. And he says that the Mobile Safari team has indicated the issues will not be fixed.

    You'd think that such damning evidence would be posted, but it isn't. Complete hearsay, but they've decided to run the story, inflammatory headline and all, regardless.

    • by jdgeorge (18767)

      Your post is so far down the slashdot page that few people will be distracted from their rants against the imagined Apple monopolistic conspiracy.

      Let the mindless flames continue!

  • I am so much happier with my Droid X than I ever was with my iPhone. And the Droid still needs bugs worked out in it. But it just isn't putting me into an ecosystem with a situation like that.
    • by MightyYar (622222)

      But it just isn't putting me into an ecosystem with a situation like that.

      A situation like what? The apps still work just as well as they did prior to an update - they just didn't speed UP. Nothing was "lost".

  • They didn't cripple, handcuff, or kneecap anything.

    They just didn't UPGRADE the web-app-run Safari to the new Javascript engine.

    web-app-run websites will run at the same speed as in 4.2, they just won't run FASTER, as a 'Safari-run' website would.

    Still not great, but not what people are calling it out as.

  • Not using the new JavaScript engine in UIWebView (the component used to render web like content) is a good thing in my eyes. It is pretty hard to debug JavaScript apps in UIWebView, since you don't have things like a DOM inspector or a JavaScript debugger. By keeping the engine the same, they have guaranteed that the existing UIWebView based apps will continue to function after the upgrade to 4.3. Imagine if every app making heavy use of UIWebView had subtle errors after the upgrade... I *would* like to se
  • by lee1 (219161) <lee.lee-phillips@org> on Tuesday March 15, 2011 @02:38PM (#35494698) Homepage

    I wrote one of these that caches some graphics and the single html page the comprises the app, which is mostly javascript. Using 4.3 it still runs fine offline, so whatever cashing bug exists does not affect everything. Running the app through iphone's Safari might be a little faster. I didn't time it, but the difference, if there is one, is not dramatic.

    I see dozens of ignorant comments here claiming that webapps on the home screen are just Safari bookmarks. What makes them real apps is that they can store themselves and the resources they need on the device and work offline.

  • I'm not sure I understand the distinction between web apps launched from the home screen and web apps launched in Safari. I have an Android so maybe I'm just unfamiliar but it sounds like home screen apps are just bookmarks to websites. When you open the bookmark wouldn't it launch in Safari? Is this only an issue in iOS 4.3 or was it in previous versions?
  • by leamanc (961376) on Tuesday March 15, 2011 @03:41PM (#35495372) Homepage Journal

    When Safari is up front and center, let it have the majority of the CPU time. When a "web app" is on the home screen, let it compete for clock cycles with the rest of the "web apps" on the home screen and the main functions of the home screen (and the phone in general).

    Maybe I am not qualified to speak on this topic, because I don't use any "web apps" to speak of on my iPhone 4, as I vastly prefer native apps. I have no web pages pinned to my home screen at all.

    And stop bitching about the 30 percent that Apple takes for App Store sales. It's right in line with what Kagi, et. al., take for selling software. I fail to see how it is unfair in the least, as Apple is bringing developers a huge audience, hosting the files, footing the bandwidth, managing the update system, etc. In an ideal world, the standard percentage that a distributor takes would be less, but 30 percent is perfectly average in not only the software distribution world, but in other areas too (e.g., 20th Century Fox generally gets a lot more than 30 percent on the first run of the movies it distributes for other producers--it's more like 50 or 60 percent; maybe it will get down to 30 percent at second-run movie houses and for DVD sales).

  • by nickovs (115935) on Tuesday March 15, 2011 @04:46PM (#35496266)

    So the real story here is "iOS Safari has got lots better in 4.3 and iOS web apps run just like they used to." Doesn't exactly sound like Apple adding handcuffs to me. Sadly that didn't make a good enough headline for The Register and anyway El Reg loves a conspiracy theory.

    The fact is that this is almost certainly just a versioning issue with the UIWebView code. When you launch a Web App in iOS what actually happens is that it a launches a dummy app that just consists of a full screen UIWebView pointed at the web page. Other than when Apple releases new full version (e.g. x.0 release) the UIWebView code has typically trailed the code in Safari by at least one point release. This "bug" quite possibly "won't be fixed" as a bug because it will just come out in the wash. If there is ever an iOS 4.4 I'd expect to see it resolved there, otherwise I'll put money on it being fixed in iOS 5.0 come the summer. Apple aren't going to put up with on-going maintenance of diverging WebKit code bases just to make sure that Objective-C applications stay running even faster than JavaScript than they would with Nitro.

How much net work could a network work, if a network could net work?

Working...