iOS 5 'crashes more apps' than Android

Recent data has shown that iOS apps crash more often than apps running on the Android platform. The data comes from Crittercism – monitoring software that records app crashes as a percentage of app launches and makes money by sending reports and diagnostics to the app's creators. After looking at data taken about app crashes …

[Anonymous Coward]

Mostly useless survey

For apps which use Crittercism's crash reporting component.

Do we know know which apps those are and if they are representative for the whole software catalog for the devices? Only Apple has all the crash reports across all iOS apps (and even then only for people who have switched on the sending of crash reports - as applicable across versions.)

Another thing to consider is that iOS 5 is relatively new and there are going to be crashes due to apps not being updated yet and that Android is likely to have a similar problem as ICS actually starts rolling out.

One more thing to consider is the the number of potential issues with the data it's self

http://junkcharts.typepad.com/junk_charts/2012/02/a-data-mess-outduels-the-pie-chart-disaster-for-our-attention.html

On top of that, this survey was undertaken by a company which receives it's funding from Google Ventures

[Anonymous Coward]

"Mostly useless survey" = Completely useless comment

Since all you did was repeat exactly what was said in the article.

[Anonymous Coward]

Not exactly the same

Come on Stike you're usually better than that.

There was additional info in the comment. A more detailed look at some of the flaws in the data. That definitely wasn't in the article.

Have it your way

[Gordon 10]

What I was going to say - but quicker!

Also the population within each app store using Crittercism would have strong bearing. Eg if its good Apps on Andriod and only badly coded Fart apps on IOS.

[Si 1]

"Crittercism has funding from Google Ventures."

So, no conflict of interest there then.

[jai]

impartial

"Crittercism has funding from Google Ventures"

<AV:PD mode> Oh re-ea-ea-ea-ea-eally? </AV:PD mode>

[geekclick]

AV:PD?

Ace Ventura Pet Detective? Love that film...

Back to the article roll out the fanbois "Oh its funded by google ventures so iOS will come off worse blah, blah, blah"

Reg why do you post these things? Is it to test how restrained we can all be under our new comments system :D

[Anonymous Coward]

Seriously, Ace Ventura???

[Peter 66]

for the Ace Ventura quote

[Nanners]

Ios5 a bomb

Thank you, I've been bitching about this for months. It wasn't a problem until iOS 5. The people pushing the iPad to do something besides surf porn know it's a problem. Though safari is one of the worst offenders at this point. iOS 5 is a completely unstable release, and I can not wait for a new iOS in what ever form. Just for the facts...I have been a Mac user since the mid/ late nineties and I have bitched on Mac forums about this issue. My complaint was promptly removed.

[Anonymous Coward]

" My complaint was promptly removed."

Of course, drones need to shut up and obey!

[Sean Timarco Baggaley]

I see your statistically insignificant anecdote...

... and raise it with an "IOS5 works just fine on MY gear."

Maybe you're running too many crap apps?

[John Moppett]

Rubbish!

I am currently researching the use of tablets in Special Education , and have an iPad2 and a Lenovo A1. I have not had a single app crash on the iPad, which is running IOS5. The Android Lenovo regularly crashes the built in apps, some of them have never worked, crashing and forcing a system restart!

[Martin]

Oh, really?

So your sample of one is proof that this survey is incorrect?

[Anonymous Coward]

Odd.

As were it to suggest the opposite Martin, i.e. that the 'research' showed that iOS 5 'crashed fewer apps', I'm sure you'd be making the very same argument about the data.

[Martin]

@AnotherNetNarcissist - Odd

If the research showed that iOS crashed fewer apps, and someone said "Rubbish! - I've got an iPad and a Android tablet and the Android works much better", I would have made exactly the same comment.

What's your point?

[Anonymous Coward]

@John

Most iOS crashes are silent, you'll be unaware until you check the CrashReporter logs - stability issues aren't limited to 3rd parties either, plenty of crashes here with MobileSafari and AppStore.

[The argument for hiding crashes and auto restarting is good usability - the platform difference is that while Android Apps also autorestart, they inform the user of the crash first].

.........also I think you'd be better served using a Google Experience device for this kind of research.

[Matt Bryant]

RE: @John

I'm also curious as to how they define "crash" - if the app has to restart, is that counted as a crash? I don't have many problems with Android apps crashing, but I have plenty with ones that I have to force close when I finish using them. Comparing three Android devices (two phones and a tablet), I've had a few crashes on the tablet (2.3 Gingerbread), one or two on one phone (both 3.2 Honeycomb), and I can't remember a single one on the other phone. The difference is the apps - the tablet has a lot of games and similar junk, the first phone is the wife's and is loaded with dross like Facebook and Twitter, whereas the other phone has an almost virgin install and a few, quality apps. Adding apps to any device will increase the chance of crashes, adding junk apps will double that amount of crashes.

[PaulR79]

@Matt Bryant

Erm.. No phone has a version of Honeycomb available to it. I've only seen a tiny few custom ROMs built using it as most ignored it and have gone for ICS instead. Did you get mixed up with your versions?

[Naughtyhorse]

but matt is special ;-) doncher know

[Matt Bryant]

RE: Naughtyhorse

"but matt is special...." We're all special in our own ways, Naughtyhorse, it's just when people talk about you being special, it's a polite way of saying you're a dunce. I'm not surprised that you have nothing of technical merit to add to the conversation.

[Matt Bryant]

RE: @Matt Bryant

Modaco Honeycomb mod, tweaked and loaded onto a pair of HTC Rubys (from the US) by someone that knows a lot more about HTCs than I do. The mod was on Megaupload, no idea where to get it from now but I have a backup saved. Works fine, the only app I've had trouble with display-wise has been the Bubbles screensaver, just displays badly. As I said, loading junk apps seems to be the problem with my wife's phone. I'm not (yet) tempted enough by Ice Cream Sarnie mods to update either phone (or rebuild my mate's NAS system in return). I didn't realise the Honeycomb mod was that rare, a shame 'cos it seems quite stable.

Oh, and the Ruby is MUCH nicer than any iPhone! :P

[Anonymous Coward]

Well...

None have ever crashed on me...

[Simon Wells]

IoS5 crashes more often...

...than any OS I have ever used. And I've still got a Vista laptop.

[Neil Stansbury]

Not my experience

Apps on my Android Samsung 7.7 tab fall over more often than a drunken sailor.

I can't get through a day without at least the GMail app and one of the plethora of awful Android web browsers crashing.

I can't remember the the last time an app crashed on my iPhone - on iOS 4 or 5.

[DaveyDaveDave]

"...one of the plethora of awful Android web browsers crashing."

May I suggest you try one of the good Android web browsers?

Funnily enough, I find that the awful applications I install on any OS crash quite frequently. Some would say that it's one of the defining characteristics of an awful application.

[Anonymous Coward]

These apps aren't crashing, they're trying to escape from the iPhone and the "person" using it!

[Yves Kurisaki]

Not happy

iOS5 (on a 4S) is the first time I've ever been unhappy with my iPhone. Apps most definitely crash way too often. I don't know if it's the OS or badly programmed iOS5 updates, either way, it shouldn't happen on an iPhone.

And to add to that, the 5.0.1 update has left me with a ridiculously fast draining battery (I managed to fix the battery problems in 5 but no such fixes work in 5.0.1) and constant SIM Failure messages.

No good Apple.

[Kirbini]

I hear this all the time too

but never have experienced it.

I bought a 4S on release day and use it heavily. I don't think I've ever even force rebooted it let alone had anything ever crash on me. I've also never experienced any battery draining issues; there's usually 1/4 to 1/2 battery left when I dock it next to the bed at night.

Personally I've never been happier with a phone. So I guess there's that.

[Kirbini]

Then again

Ok, I guess I forgot to knock on wood before I posted. I just had my first crash. In the middle of a stupid Zynga game (Scramble With Friends) so the blame could go to them as well.

Completely locked up and had to to a hard reset. Not fun. Not happy. I formally withdraw my previous comment (but will leave it up so all and sundry can witness my humiliation).

[Ian 16]

SIM failure

the sim failure issue is fixed... apple released an update but did not update the revision.

If you check for updates it WONT find one but if you do a restore from a newly downloaded firmware it should fix the problem

I have that issue on my 4S -- but I'm waiting for the next point release

hope that helps

[Bluelight]

Does anyone really care?

Ok so perhaps the open-source zealots are rubbing their hands over another "victory" but to be honest, does the average user really give a toss that sometimes they have to start an app twice - on either platform??

Some people clearly have far too much time on their hands. When your car doesn't start first time off the key every now and again do you just turn then key a second time or do you sit there and wish you'd bought a different model... I doubt it!

[ThomH]

One counter example: I recently had the game, Jetpack Joyride installed on my iPad 2. It's one of those casual games you play for a few minutes every so often so I couldn't tell you my total game time but I had it installed for probably a month or two and had hit all the achievements at some point or another.

There's no reason to assume iOS had any hand in what happened but a few weeks ago it abruptly crashed partway through a game. Upon relaunching I discovered that it had somehow corrupted its state while crashing, with the result that all of my achievements, high scores, etc had been thrown away. I haven't bothered to launch it again since.

That's the sort of thing that I think the average user would care about.

[Field Marshal Von Krakenfart]

does the average user really give a toss that sometimes they have to start an app twice

But why should you accept that behaviour from any application, if your new car kept cutting out you would bring it back to the garage, *and* possibly ask for a refund, so why accept this behaviour from software?

[Charles Manning]

Completely different expectations

People have completely different expectations for software vs cars.

A fart app crash is hardly serious whereas a car malfunction is potentially far more serious.

On the other hand, people expect an Undo button for software actions, but not when they've driven a car off a bridge.

Sure, you could demand high-nines software that crashes once a year or less, but are you willing to pay for it and wait for it to be developed and tested? Nope

[Naughtyhorse]

nice 180 there

bit of a change of tune for the usual jobs megalomania apologists.

the whole point of having steve run everything in your life was that then everything in your life 'just worked'

now you are saying if things don't just work, it's no biggie.

roflmfao

[Anonymous Coward]

Yeah dead happy with an Android app-auto-update to add malware. No thanks.

[Anonymous Coward]

Planning on backing that up with anything or just talking out of your arse in small bitesize pieces?

[Anonymous Coward]

Re AC 'Planning on backin that up...'

And Apple users are fanbois that can't take criticism? You fandroids take the biscuit!

[Anonymous Coward]

"And Apple users are fanbois that can't take criticism?"

That was trolling, not criticism. Trust me, I know trolling. :p

[Anonymous Coward]

@AnotherNetNarcissist

As pointed out above, the OP wasn't providing criticism, hence the request to actually do so - y'know, by explaining how they got from "auto-updating" to "adding malware".

[Anonymous Coward]

"That was trolling, not criticism. Trust me, I know trolling. :p"

;) The 8 downvotes (as of 17.30) say it all really though! Fandroids really need to lighten up because they are a parody of all that they purport to dislike...

[DaveyDaveDave]

So... you always manually update, that way you can be certain you're not getting any malware, right?

[Vic]

> dead happy with an Android app-auto-update to add malware.

You don't have to turn it on, you know.

It's selectable on a per-app basis. IIRC, it's also disabled if the app changes the permissions is requests.

I don't use it myself.

Vic.

[Anonymous Coward]

From what Crittercism says their dataset has three times more app launches recorded for iOS than Android, so I suspect they are looking at a greater number of apps on iOS than Android too.

Now which developers will tend to be using Crittercism's middleware? Those who have problems with crashing apps in the first place! Why else would you add this creepy, CarrierIQ-syle, spyware to your apps when the SDK and app store already gives you analysis and bug reports?

In summary they mostly get the really crappy apps so, if their iOS dataset has more of those, the results are going to be worse.

To be of any interest they should have picked the same number of apps, ideally apps with both Android and iOS versions. But of course by releasing it this way pleased their sugar daddy Google overlords more and got the Internet flamewar machine going (bonus: Google gets paid in increased ad impressions too)

[ThomH]

Could it be just the differences in how they handle memory?

In iOS objects are referenced by standard C-style pointers and prior to iOS 4.3 and the iOS 5-related development kit you had manually to manage reference counts (well, a little; property syntax made this a lot easier and obviously all the built-in collections get things right automatically, but you still usually need to be explicit about when you want to keep an object and when consider yourself done with it, and dangling pointers are a very real risk).

Android uses a dialect of Java, with a garbage collector that has freed Android developers from having to think about memory management since day one.

One would therefore assume that you can be a worse Android programmer without your lack of skill causing your app to crash.

[aside: I'm aware that the ARC compiler means that developers needn't manually perform these tasks any more on the iOS side, but this survey is talking about historical data]

[Nanners]

Memory problem?

Yes, in my unprofessional opinion it acts like a memory problem. I've noticed that.

[Nanners]

And as a final addendum

Ive often told myself it acts like a memory allocation problem specifically. As I said, I'm not a professional though. I am willing to forgive a little drop in quality control after the death of Jobs, but I also expect them to address those issue as they arrive.

[Anonymous Coward]

Not sure which apps use this Crittercism thing, but for serious developers using the Android NDK - for performance reasons - memory management IS still a big issue and developers have to think about it.

Even in iOS 5 I still prefer manually managing memory. Any automatic garbage collection system is only going to be worse and introduce slow downs at the wrong times.

[MerlynUK]

Having just jumped from iOS to Android, the one thing I have noticed is that when an app crashes on Android it is usually far more visible. On iOS when an app crashes it just closes with no warning message - and it happened fairly regularly on iOS5, more so than iOS4 on my 3GS, as well as being slower and draining the battery quicker - iOS5 on the 3GS was a big mistake in my opinion (my wifes 6 month old 3GS is the same since she upgraded to iOS5 and she doesn't have anywhere near as many apps and services running than I did).

Andoid on the other hand, presents the 'Application is not responding, Force Close?' message so at least you know there is a problem. So far, I haven't seen it too often although I don't have too many apps installed yet!.

I don't use auto-update unless it is a reputable application, but its nice to have the option. On iOS when there were updates I just tended to hit the Update All button in the Appstore whenever there were updates, however with Android maybe you have to a bit more careful!

[Anonymous Coward]

Android

"I don't use auto-update unless it is a reputable application"

Pointless sentiment on Android, auto-updates are disabled for an app when the update changes the required permissions, you have to manually update and re-accept the permissions in the process. Not auto-updating your apps on the other hand means you may well miss out on security fixes for longer than necessary...

[Marcelo Rodrigues]

Android warns You about the updates

I have auto-updates disabled. Color me paranoid, but I have.

I don't risk forgetting some upgrade: the OS nicely warns me, at the status bar, that there are updates - and for which software.

So, yes. I may update a few hours later - when I'm within an AP range - but not such a big delay...

[Anonymous Coward]

I'd rather see a pure Android analysis

Show me a breakdown of crashes by whether or not the bloody thing has Sense on it. I'm sure the crap installed by HTC caused most of my woes before I finally got Cyanogenmod installed.

[DF118]

Cyanogenmod

I've actually had more crashes and hangs since installing cm7 on my previously motoblur-infested Defy. Mostly with stuff that accesses hardware other than the screen. Still a much nicer UX overall.

[h3]

On my Xoom sometimes there is no honeycomb optimized version of a certain app so I have to use a phone app and that generally doesn't work well.

Never experienced anything like that on my brothers ipad (iOS 4.3.5)

[moonoi]

iOS 5 does crash more often

It is true in my experience of one :-) the mail app connected to my companies exchange server is constantly dying, or requires me to manually kill it on occasion to be able to receive push emails. IN addition it is very very slow to retrieve very basic text only emails.

In comparison, my old Android HTC Sensation XE and my new Galaxy Nexus have no such problems.

If I disable the exchange mail sync, it becomes more stable, but as its a company phone that defeats the object for which it was provided!

My wife refuses to upgrade to iOS 5 simply because of the constant app crashing I have experienced on my iPhone 4. I wish I had stuck with iOS 4 now....

[MoT]

Is it not possible to roll back? I haven't an iPhone so I'm only guessing.

[Naughtyhorse]

lol

option to... anything on an iphone

brilliant line, you owe me a new keyboard

[Giles Jones]

Language problem perhaps?

I would imagine it can be put down to the fact that Android uses a familiar widespread syntax (Java style) and iOS uses Objective C which is around for other platforms but is almost exclusively used on OSX and iOS.

It's quite a different syntax than Java and C++, so I can only imagine there are a lot of inexperienced iOS app developers around.

[Mike Bell]

Not 100% Convinced

My experience of iOS 5, for what it's worth, is that it's pretty damned stable (for me). With a notable exception of BBC iPlayer, which would crash on start-up on about 25% of occasions. That particular app was updated recently and the release notes acknowledge that it includes a fix for crash-on-startup. I wonder how many millions of recorded crashes were accounted for by this one hugely popular app?

My experience of Android, for what it's worth, is that it's also pretty damned stable. Unless you are one of the mindless numpties who gets clever and thinks it's a good idea to install a Task Killer on the device. In which case you don't understand how Android works, and you are simply begging for your device to crap out on you.

I am a little intrigued by the stats in the article, though. The trend seems to imply that it won't be very long before every app launch results in a crash (look at the way the graph is headed), and all that Android is doing is better keeping pace with iOS in this regard. Such convergence looks a little suspicious to me.

[Anonymous Coward]

I run iOS5 and apart from the previously noted iPlayer app (which looks like an bug with the app not iOS) it's been very stable - of course YMMV. Apps crashing seem more likely to be problems with the app rather than the OS?

[ratfox]

"Why didn't they give data for Windows phone?"

"My WinPhone never crashes ever ever it is the best thing I ever had and I just signed up to tell you how great it is and you should try it"

C'mon, astroturfers, do your job...

[Nordrick Framelhammer]

Could it be that there are too few WinTards to make a statistically significant bump in the data?

[Great Smell of Brut]

My WinPhone never crashes ever ever it is the best thing I ever had and I just signed up to tell you how great it is and you should try it.....

and you don't need to install completely different firmware or some kind of quad-cored monstrosity to stop it running like a dog

(PS will this do?)

[Anonymous Coward]

"My WinPhone never crashes..."

'coz you have no apps for it.

[Lord Lien]

Mail app on IOS 5...

... is really bad for crashing when you reply to emails. I've worked out if you delete the entire email thread before you reply its fine. Very strange bug in the application. You have thought this would have been found in testing.

https://discussions.apple.com/thread/3394146?start=0&tstart=0

Paris because.... well because ;)

[Toothpick]

O bugger...

I was reading this article on my iPad and it cra

[eforce]

Maybe iOS developers are just more lazy?

[Anonymous Coward]

Hang On

So an app crashes and it is the fault of the OS???? Nothing to do with the hack that wrote the app then?

[Ken Hagan]

Re: the hack that wrote the app

Not the OS, no. But the vendor? Certainly.

Apple claim that their walled-garden approach to apps is to ensure that the badly written apps don't make it as far as the end-users. They are quick to point out that there's no quality control at the Android store.

On the strength of this (partial, in all senses of the word) evidence, it doesn't make any difference.

[Field Marshal Von Krakenfart]

May I be the first to enquire...

...are the iPhoney users holding it the right way?

[Anonymous Coward]

I got nothing out of the Forbes article...

... other than that there are some people who should never be allowed to make pie charts. Ever. I've seen explosions in abattoirs that weren't as disgusting as those pie charts.

[Gil Grissum]

The only app on my iPhone that crashes in Google+. The rest work just fine, so I have to wonder where all this Apple crash data is coming from, considering that this think tank is Google Funded. of course they couldn't go against the org who is funding them. I call BS on this "crash study".

[Cameron Colley]

@ Those running IOS5 with no problems.

Just because your device is OK with the apps you run doesn't mean there aren't problems.

Personally I've found this iPad crashes more whilst using Safari tha it did before the upgrade to IOS5 (I recall it crashing a few times before the upgrade but now it crashes on Some sites every time). Other apps seem to crash more too but, to be honest, I think they were fairly crash-prone anyhow so Apple can't be blamed for them, or perhaps only blamed a little since the point of walled-garden app stores and limited configurabilitynis supposed to be increased stability.

Then again, this is just the model of device and apps I use, so perhaps the survey is wrong after all?

[Archivist]

Another sample of 1

The only crashes I've experienced are with the Guardian app. and it's happened less on IOS5. Mind you, I don't play games or connect to Exchange servers (same thing really ;-)

I've plenty to criticise Apple for, but this isn't one of them.

[scarshapedstar]

My anecdote

My iPad 1 crashes on Safari pretty frequently, but it's very elegant about it - you just get booted back to the desktop, which gives the impression that it wasn't a crash, more of a miscommunication. Android gives an honest "App dead; force-closed" popup. If anything, I would think iOS crashes are underreported for this reason.

[jayess]

Windows.. reliable???

I've not had much exposure to Android but I have been surprised to find Windows Phone much more reliable than iOS - seriously!

[MoT]

I wonder...

All this discussion reminds me of my days supporting clients in the field. Every now and then they'd pay me to make a house call about a PC problem. Usually something about it "crashing or going so sloooooowwwww". Once I arrived and took a look at things I'd notice all the file sharing applications and instant messenger clients sitting active or waiting to pounce in the task bar. I would then ask, "Do you have teenagers?" Then I'd get the sheepish acknowledgements. Then look at the browser with the countless tacked on "tool bars" and other worthless spyware ridden cruft to understand where the problems lay. Now that's not to say there aren't problems in the base OS, there always are, but to compound said weaknesses with unstable crapware is only to beg for a justifiable beating.

[Cameron Colley]

So you're saying that Safari is unstable crapware?

Under IOS you only run a few applications simultaneously and even then some are just running cuncurrently, there should be no crapware services running because it's not possible to start any as a user. If software is somehow staying resident after the user has closed it then it's the OS that's at fault because Apple own the API and have full control over applications -- users have no control. Also, each app is supposedly sandboxed and, therefore, should have no impact on any other.

As I mentioned in my previous post, the whole point of locked-down walled-garden OSs is that they are stable and the user can't mess them up. So crashes under IOS are either Apple's fault because they allow apps to crash or Apple's fault because their walled garden is broken.

Android I have less experience with but left unaltered (i.e. not rooted) it ought to be prett much the same situation.

[pcsupport]

I don't have a problem with crashing apps...

... but then I use my smartphone as a phone, I leave the computing stuff to a computer and for taking photos I use a real camera.

I also have a life....

[Anonymous Coward]

Once an iPod always an iPod?

Obviously it was originally designed as a clever mp3/video player, not a multi tasking computer.

Whilst it's never worked very well since it became a phone it still looks nice.

[k9gardner]

Sad but true

The problem with iOS 5 is definitely there, but it's not like it wasn't there with 4 as well. I'm a die-hard Apple fan for many, many years, but I certainly noticed the difference between the "get it to market as fast as possible" mindset of the phone/pad world and the more considered approach to product releases in the computer division. Particularly in consideration of Apple's walled garden, where nothing gets released without first having been at least tried out and looked at, the number of issues that cause things to break is a little disappointing. In fact, unless there's some very real truth behind the allegations that Crittercism is not quite looking at a level playing field, it's pretty inexplicable.

I'm sure it'll get fixed in due course, but the fact that everyone is constantly in a mad rush to offer features over finesse will always mean that products are not as polished as they could be. The extent to which we all just accept it because of a nice shiny presentation is the extent to which we'll have to deal with the failure of things we need to work.