Apple ignores bug report

[alttab]

You're telling me Apple is unresponsive to support requests from their developer ecosystem? Really? Get out of town! I don't believe it.

Despite all of the rightful moaning of iOS developers, for some reason they continue to flock to the Apple platform. Apple will continue to treat their developers like second class citizens until there is a financial incentive to do otherwise. Right now, when one pissed off developer leaves or goes bankrupt because their app was yanked from the store or wasn't approved for some BS reason, 50 developers replace him.

[EnderMB]

The fact that so many of the comments replying to yours are quick to defend Apple says it all really.

Developers will put up with more crap than they should when platform religion gets in the way. Sure, there are arguments against switching platforms to Android or Windows, but most of these mean very little. Those that refuse to ditch iOS aren't doing so because they'll miss out on potential customers, it's because they like the idea of building apps for the most popular phone on the market, despite the crap they have to put up with.

It's the equivalent of dating a bitchy model. You could be a lot happier with the cute girl next door, but you'd rather put up with her shit just because you can say to your mates "I go out with a model".

[technoslut]

>You're telling me Apple is unresponsive to support requests from their developer ecosystem? Really? GET OUT OF TOWN! I don't believe it.

This isn't something that is exclusive to Apple though this has always been an issue with them even regarding security issues with OS X.

>Despite all of the rightful moaning of iOS developers, for some reason they continue to flock to the Apple platform.

They develop for iOS because that is where the money is though this isn't the issue. It isn't as if the Android browser doesn't have its fair share of bugs.

>Apple will continue to treat their developers like second class citizens until there is a financial incentive to do otherwise.

You're right. Apple puts the interests of their customers first (not trying to be smug here). The developers come second. The only time you do hear complaints that go unanswered for devs it is usually only answered when there's a public (consumer) interest regarding it. A good example of this is Phil Schiller replying to customer's email about the Rogue Amoeba situation.

I'm certainly not defending Apple but this is a problem that has existed for as long as I've been interested in tech.

[guscost]

I put up with developing for iOS because it is the only mobile platform that can do low-latency audio on every device. There isn't always a philosophical choice involved.

[WiseWeasel]

It's no secret that third party developers fall pretty low on Apple's list of priorities. I get the sense that Apple's priorities are 1) balancing user interests with Apple's bottom line, 2) satisfying the interests of carrier partners, and finally 3) serving the needs of third party developers.

This doesn't mean that fixing this bug in Mobile Safari should be expected to be unimportant to Apple, since it affects one of their top priorities, the users. It would rather seem to be an indication of either a sub-optimal process for dealing with bug reports, or an issue with the author's submission, which given his liberal use of strong language, doesn't seem to be out of the realm of possibility to me.

[mikeash]

Entirely correct. But what are we supposed to do about it? Switch platforms? As you point out, 50 developers will replace me if I do that.

[neya]

You're talking like an Apple Sales-man, except that we know for sure you aren't getting paid.

[batista]

>Despite all of the rightful moaning of iOS developers, for some reason they continue to flock to the Apple platform. Yes, strange, I wonder why that is, they must be irrational or something, unlike us sane guys.

It's not like there are money to be made there, or that is the most successful mobile app ecosystem in sales volume, returns and customer base. Or that it has a nice platform overall, despite having the occasional 2 year old unfixed bug...

That said I've seen 2 and 5 year old unfixed bugs in open source software too. Come to think of it, Webkit, which powers Mobile Safari, is open source itself. Why not go and submit a patch there?

[franzus]

Dealing with Apple can be frustrating sometimes. The Mac App Store review times are ridiculously long nowadays: https://devforums.apple.com/thread/154948?tstart=0 (developer account needed)

2 - 3 weeks seems to be the norm. That's pretty frustrating if you consider that people's income depends on that. And that developers pay $100 + 30% share for this.

[earl]

Willie, why do you rob banks?

"Because that's where the money is"

[zoop]

Being critical is great. Creating workarounds is even better. Being entitled is not so great.

If you want to talk about how Apple is terrible to developers, fine, talk about the App store double-standards and the developer agreement. But this is hardly an exceptional case of a company being 'anti-developer'.

I can't tell you how many bugs I've filed from OS X, chrome, to various python libraries well the response is pretty much "welp its broke" (if that). If you depend on some sort of functionality that you're not getting, it's time to move on or create your own.

The tone of this post reminds me of 'why I'm not developing for twitter' post ... where the take away for that was be weary of developing on platforms you don't control. I would assume that most of us have learned that the hard way with esoteric libraries with ghost-maintainers. Myself, I'm becoming weary of turning every inconvenience into some sort of political issue. This is _hacker_ news. Can we get back to making clever and disruptive solutions please?

[phoboslab]

I built a lot of workarounds to make my game engine ( http://impactjs.com/ ) work on iOS, Android and other browsers. But I couldn't find a viable workaround for this multitouch bug. The touchstart/move/end events are the only way to detect multiple touches in Mobile Safari - and they don't work as they should.

This bug is affecting many HTML5 games and as long as I'm not allowed to install another browser (engine) on iOS, I will keep complaining.

[drzaiusapelord]

>Can we get back to making clever and disruptive solutions please?

So wait, when MS was abusing its IE product then HN is justified in angry anti-MS rants, but when Apple purposely keeps html5 gimped on iOS and refuses any other browser on iOS suddenly you're all about tolerance and double-standards.

This a very legitimate issue. Apple can't applaud HTML5 in public and piss on it in private. Yes, its a threat to your app ecosystem. Accept it, fix the bugs, and allow other browsers. This is HN worthy, if not extraordinarily HN worthy.

[mootothemax]

Whilst I think Apple took too long to look at this bug, the instructions in the given test case are, for me at least, ambiguous:

1. Hold one finger down in the blue area above

OK, that's easy enough :)

2. Touch straight down with another finger - don't move your fingers on the surface

And here's where the confusion lies; I thought you wanted the tester to slide their finger down the screen or something like that. From the video it appears that you want to ensure that they "hold another finger on the blue area" or something like that.

Now, I confess that I don't own an iPhone, so couldn't test whether that made much of a difference, but that could be one reason why they rejected your test case. If in doubt blame incompetence, not malice :)

[phoboslab]

I'm not a native speaker, so these instructions may very well be ambiguous. Any suggestions on how to clarify it?

Back then, I built the test case and asked on twitter if anybody else could confirm the bug. I got several answers (e.g. one from mrdoob within the minute: https://twitter.com/mrdoob/statuses/6896165488427008 ) so I thought the instructions were fine.

[quonn]

I have filed dozens of OS X and iOS bugs over the years. Almost all of them were closed with the claim that Apple already knew about that bug. However many of them were really subtle and it was unlikely that anybody else had already reported them, especially if a developer preview introducing the bug had just been released. Furthermore some of them are not fixed even now, years after the first report.

My experience is that Apple will almost never admit they don't know about a bug. The only exception are security bugs where one actually gets real feedback.

[Spooky23]

My experience is that Apple isn't that bad. I have filed several bugs with Apple over the years, and IMO their response is pretty good compared to other vendors, particularly given that I have no paid support relationship with them. The bugs ranged from AD integration issues to PDF rendering problems, to an ActiveSync issue specific to our unusual configuration.

I do have an expensive support relationship with Microsoft, and I've gotten similar results for the same types of problems. We log bug reports, and the serious ones with business impact get fixed. UI glitches or corner-case Outlook problems get fixed more slowly, or only via custom hotfix. Serious bugs get fixed more quickly, and typically get rolled into a future patch.

In the past, I've had similar or even worse results from other large companies like IBM, Oracle, Informix and Sun.

Software has bugs, and low-priority bugs persist for a long time, for reasons that may be good or bad. In any case, posting expletive-laden rants railing against a vendor is unprofessional and unlikely to garner a positive reaction.

[bazzargh]

I've only filed a couple (against Quartz), and had a similar experience. They'd maybe ask for a reduced test case, then nothing happens for 3 months; the latest patch of the OS comes out and the next day the bug is closed. I don't mind if they fix the bugs, but the stony silence is a bit frustrating.

One that never got fixed was crashers in Console. That's been going on for years now.

[saurik]

Thers are tons of people using this platform, and many of them are doing very obscure things with it. Meanwhile, that same bug tracking system is used by Apple's Q/A and the very engineers working on the code in the first place. You might thing the bug is something only you would find, but with this many people staring at it, someone else probably came across it. As many ways as I feel Apple is slightly disingenuous, I really can't imagine they have a policy of making up bug numbers in Radar so they can mark things as duplicates.

[vosper]

With all that swearing you could be forgiven for thinking Apple had been caught skinning kittens for iPhone parts. Save the hyperbole, please.

[simonh]

OS X Skinned Kitten. See - I knew they'd think of a new cat name for the next OS release.

[cobrausn]

Two year wait and 40 hours of developer time spent on a simple bug is worth a bit of swearing.

[richardv]

I have so much sympathy for you. (Along with anyone else that tries to produce anything that runs on any iOS device and it's walled garden.)

It's easy to see that when building a platformer game, unimpeded multitouch functionality is pretty essential. Have you been able to identify other games that have the same problem? I would of thought that this would be resolved so much sooner than 2 years. It's a core part of multitouch functionality.

At least you still have your app available on iOS!.. It is total BS how developers are treated. Apple has built up a product which sells hundreds of millions of units per year, largely in part due to the developers who have invested time in their ecosystem, but it's total shit how they return no investment back to the people who have supported this system. (Yes, developers are largely to thank. Look at how RIM and MS are failing, because they don't control nor have the content).

Everyone deserves better.

[crag]

I agree. But, the only reason anyone writes for iOS is money. It's one of the best selling phones on the market. And iPhone uses tend to buy a lot of apps. There's a long list of developers who made fortunes on that platform.

So if you don't want to write for it, then don't. The only force that can change Apple's behavior is the consumer. Apple doesn't care about anything else. And they never will.

So stop complaining. Write your app. And hope it's a hit. Then you can sell it and buy a house on the beach.

[trotsky]

Is there any evidence that Apple wants game developers targeting mobile Safari? It's a classic platform tactic to abuse APIs you don't want used (HTML5) to drive people to the one you do want used (ObjC). Between platform lock in and 30 cents on the dollar it's clear Apple wants developers making native apps.

[thechut]

I think this whats really at work here. For the same reason they won't let Mozilla, Opera, or Google make their own web browsers. Apple is affraid of people building their own HTML5 apps and subsequent app stores. This would cut into Apple's app store monopoly and cost them their 30% on every app that is sold.

Apple only cares about what makes it money, not delivering the best experience to consumers.

And you can be damn sure that they don't give a shit about developers, especially if they aren't making native iOS apps which aren't being sold in the app store.

[jpxxx]

Letting an obscurant input handling bug fester so that they can drive people to a 41% margin product instead of a 40% margin product? Doubtful.

[wavephorm]

Between introducing multi-touch errors, and hobbling the HTML5 Audio API, there is no way to build an engaging game for mobile Safari. Apple has sabotaged their web platform with expert precision.

[Fliko]

Don't worry man, I have this Nexus One phone where the touchscreen registers touch on the side of the phone, and the advertised multitouch doesn't even work, never mind not receiving updates for the rest of my life (my hardware bugs cost me over $600!!). Your game has no chance to work on my phone, why don't I see a "What the Fucking Fuck, Google" post. All systems are plagued with bugs, and a lot of them get neglected forever. Instead of aggressively attacking a company (Apple hateboy?), expecting them to fix something after some f-bombs were dropped, why not write a meaningful blog post.

[Karunamon]

> why don't I see a "What the Fucking Fuck, Google" post.

Because you haven't written it yet?

[wavephorm]

Because it used to work and Apple broke it (possibly on purpose) and refuses to fix it. That's not the same as it never working in the first place.

[jonhendry]

I've only filed a couple of radars, but actually received pretty good response. However, they weren't api-related.

One was about an issue I was having running the Britannica encyclopedia. I don't recall what it was, but Apple even called me about it.

One was about a bug in Lion, where if you pick a large desktop image, and tile it, it gets all corrupted and there are artifacts in the desktop image that respond to movement of windows above the desktop. The corruption of the image looks like what happens when CIAffineClamp is applied to an image: a row of pixels at some place in the image is repeated out to the edge of the screen.

The response was that it was fixed in a point release, but the fix was to disable the Tile option for large images. Except if you have it set to alternate images periodically, with the Tile option selected, a large image can still be tiled and displayed incorrectly.

[TwistedWeasel]

So he let this bug sit in his game for two years without thinking of a work around himself?

The situation where any platform owner has a bug that impacts your software you should try as best you can to work around it because you never know if and when it will get fixed.

[AshleysBrain]

This doesn't sound like the kind of issue that can be worked around. The OS sends you the wrong touch events. How do you work around that?

[Kiro]

What game? I think you missed the point of his blog post.

[SigmundA]

Reviewing the test case code the basic bug seem to be that sometimes a touchend event won't fire when a secondary finger is lifted.

You can make your code much more reliable by handling all touch events and looking that the touches or targetTouches respectively instead of incremental a total touch counter on each start and decrementing on each end. This causes even the slightest movement to trigger a touchmove event which and the entire current touch state is in touches within all the events. So changing your test like so gets close and you can at least see the use more clearly:

var i = 0; var numTouches = 0; var out = document.getElementById('out'); var indicator = document.getElementById('indicator');

function handleTouch( ev ) {

		ev.preventDefault();
	
		i++;

		numTouches = ev.touches.length;
		
		out.value += i + ': ' + ev.type + ', ' + numTouches + '\n';
		
		indicator.style.backgroundColor 
			= (numTouches == 2) ? '#8e6' : '#fb7';
		
		out.scrollTop = out.scrollHeight;

	}
	
	var ta = document.getElementById('ta');
	ta.addEventListener('touchstart', handleTouch, false);
	ta.addEventListener('touchend', handleTouch, false);
	ta.addEventListener('touchmove', handleTouch, false);
	ta.addEventListener('touchcancel', handleTouch, false);

[ottoflux]

1) Yes, Apple is closed and lame. This isn't new.

2) I almost fell over laughing when you mentioned a whole "40" hours. I think if I went through that whole process with Apple, I'd be happy if I got it out of the way in 40 hours.

Apple "used to" (as in, they cast a damn good image of) following Dieter Rams type of design philosophy, but that's long gone in the freak show app-orgy that exists within iOS now. It's a mess, it's terribly disorganized, it's not "less, but better" it's "more, and worse".

Personally, I'm surprised that they let iOS get this stale. But apparently as long as the penumbra of Steve's[RIP] reality distortion field exists, they will blindly march on.

What Apple needs is a new visionary. Not someone to try to emulate Steve Jobs. And they need to open the hell out of their products.

If you believe your hardware and OS are best, then open the hardware to letting people install Android on it if they don't agree, and write a version of iOS to run on other hardware.

It's not going to happen (and I hope they prove me wrong). Apple is designed as a coupled HW-SW ship, and I don't see them changing course any time soon.

My advice: Pay attention to your perspective bed partners before you commit and get into bed with them.

[astrodust]

It's not like Android isn't without it's infuriating problems. Worse, most people are running several versions behind and the uptake on new ones is slowing, so some problems may never be fixed.

[timkeller]

Apple are way more focussed on UIKit and the rest of the iOS frameworks, than they are in fixing a minor Mobile Safari glitch.

Two years spent building a work-around would be a far better use of OP's time.

[kalleboo]

Going to a WWDC lab is basically the only way to guarantee an acknowledgement of a bug from Apple. Too bad it costs thousands of dollars and you have a 2 hour window in which to seize your chance.

[nixle]

There is no need to get emotional. It won't help anything and it just clutters the issue. Just mind you language and stick to the facts, we are all grownups here.

[ahoge]

This was 100% deliberate. It's just some spin. The post gets more attention this way, which makes it more likely that the underlying problem will be fixed somewhat sooner.

Besides, with a headline like this it will also get automatic upvotes by everyone who dislikes Apple.

[johnohara]

phoboslab:

I filed all the bug reports in a professional manner, without insults or swearing. I have a lot of respect for the engineers working at Apple, ...

Why are we excluded?

[andrewfelix]

There are more accurate ways of registering the second touch. Have you tried using 'event.targetTouches'?

[phoboslab]

The touchstart event isn't fired. There's no 'event' object to access.

[NinjaWarior]

This is not a only blocking problem for HTML5 gaming. Chrome Canvas latency problem, Firefox canvas performance bug, HTML5 "poor" Audio, HTML5 "poor" video, Canvas "poor" API, SVG never works correctly, underestimated fragmentation cost, poor programming language (why the hell did they abandon ECMAScript 4?), when the hell can we use WebGL? etc...

People was complaining Flash was vendor lock-in. So kicked out Flash and now we are prisoners of fucking browsers. Kicked out poor Adobe, and Apple, Google and MS rules! Yay! Do you enjoy this?

I think plug-ins were clever, liberal and democratic software designing. We were able to choice the best technology we want to use (including Java Applet, Unity Web Player, Silverlight and even Shockwave). We were even able to develop our plug-ins freely. We had a freedom of programming language other than fucking JavaScript.

I feel people's enthusiasm for HTML5 is completely wrong. At least they should stop saying "HTML5 is a cool technology" "HTML5 is an open technology". COMPLETELY WRONG.

[Iv]

What the fuck, developer ?

how could you not know that iOS is a closed garden ? Stop being complicit of this dangerous ecosystem. I understand that a lot of people have spent a decade or more in a very good relationship with Apple, swimming in wonderful hardware, nice frameworks and a helpful and friendly community, but it is finished now. This is not the Apple of today anymore.

[ahoge]

Don't put spaces in front of question marks. If you're French, okay, you're forgiven. However, a regular space is still completely wrong and it will totally break word-wrapping.

[ademuk]

I wouldn't be at all surprised if this was intentional in order prevent devs creating HTML5 games on par with those of the AppStore.

[wavephorm]

It absolutely is intentional. And with that same intention Apple forbids other companies from writing web browsers that fix these and other issues with mobile Safari. Auto-playing audio and video in the browser is also forbidden, which prevents any sort of online music, or video app from being created, among other things.

So games, music, and video apps in mobile Safari are all hopelessly hobbled, and if you build a browser that fixes these issues you'll be banned.

Happy developing!