If they're going to pitch HTML5 as an alternative to Flash (and, implicitly, the iPhoneOS API), then why not let them compete directly with Cocoa Touch apps? Why not offer HTML5 developers a way to take advantage of the App Store's monetization and discoverability?
Apple already provides some amazing documentation and tutorials for making offline-capable mobile web apps, but they're still treated by Apple as the bastard cousins of "real" iPhone apps. They should buy or provide an officially-supported analogue to a tool like PhoneGap[1] or Titanium[2], allowing HTML5 apps to appear in the App Store, and perhaps provide an Apple-supported framework such as their mysterious PastryKit[3] or contribute to SproutCore[4], giving HTML5 developers better tools to conform to Apple's HIG.
HTML5 apps built on open standards should be on an equal footing with apps built in Cocoa Touch. That would be walking the walk.
I get this argument, but couldn't you say Apple gives the customer the best of both worlds?
Web & HTML 5 support: developers can make or do anything they want to on the platform.
App Store: Cultivated catalog of high quality apps for customers.
At the end of the day, we don't need PhoneGaps or Titaniums-- we need MobileSafari to have the same hooks into the hardware (gps, compass, accelerometer, camera, etc) as Cocoa Touch. Then it's a level playing field.
If you're saying that that, too, ought to be in the App Store, well-- then what you're really arguing for isn't technology but that the App Store is a better distribution channel to mobile devices than the Web. It certainly is-- I have millions of competitors for the word "Congress" on Google, and about 12 in the App Store.
To solve the distribution channel problem, I don't understand why some enterprising YC person doesn't simply start making a web App Store app for the iPhone that allows people to sell web applications to iPhone users, handle fulfillment just like the App Store does, and handle curation, just like the App Store does.
What I'm arguing for is not technology (which already exists), but technology support in the form of official frameworks, which subverts the "3rd party middleware" argument, and for equal access to distribution and payment channels. I have no motivation to create iPhone web apps if I can't easily get them in front of iPhone users monetize them.
A few other points:
- MobileSafari already has access to many hardware functions: gps, accelerometer, maybe compass? There are some great iTunes U videos on this if you're a member of the iPhone dev program. Obviously it would be good to have access to all the hardware features.
- Every time I hear about the App Store being "curated" for "high quality" I choke a little (iFart?). There's no reason why html5 apps in the app store wouldn't go through the same review process.
- The Google results for "congress" have no relationship whatsoever to "congress" in the App Store. Web pages != web applications.
"I don't understand why some enterprising YC person doesn't simply start making a web App Store app for the iPhone that allows people to sell web applications to iPhone users, handle fulfillment just like the App Store does, and handle curation, just like the App Store does."
Not a bad idea, in fact. A few big hurdles, though: 1) you can't install apps on the iPhone except via the App Store (or asking a mobilesafari user to create a home screen icon themselves) 2) payment system inertia-- who's going to sign up for another micropayment account when they're already set up with Apple?
EDIT: while I was writing this, nailer said it better and more succinctly.
So I think we're in (nailer) agreement, though I don't see why anybody has to wait on Apple to make a web app store. I think there's money to be made even with all the marketing hurdles.
I want to explain myself in regards to your response here:
"- The Google results for "congress" have no relationship whatsoever to "congress" in the App Store. Web pages != web applications."
In a lot of cases this is true. But I, for instance, make web apps about members of congress. My mission is to get people information about members of congress wherever they're at. In order for those web apps to get discovered, I have to rely on Google for people to discover them (as well as good marketing and PR and all sorts of other things). I'm simply point out that the App Store is, in fact, a less competitive environment than the web for the dissemination of information, which a lot of web applications, and iphone applications do.
Type "Weather" into the App Store, and there's a few hundred applications. Type it into google? 319,000,000. I'd wager to say that 99% of "Weather" applications on the iphone are about the dissemination of weather information.
So while, yes, web pages != web applications, many web applications are about the aggregation and dissemination of information. And search-- whether it be through Google or through the App Store-- is an important factor to their discovery and use. Being in the "top 25" apps for a search result in the App store is valuable, just like being in the top of Google search results. And right now, the App Store is far less competitive than Google search results.
I don't see any reason why HTML5/Javascript apps couldn't be in the App Store as some sort of "featured" section, with a means to install them as icons on the home screen, except that Apple would be diluting the market for the native apps. That's not in the developer's interest, but it might well be in Apple's interest to have more decent free stuff on the iPhone.
who said anything about free? my point is that mobile web apps have no way to monetize, and that could be fixed by letting web apps in the app store, via a native bridge.
But how is this different from the monetizibility of web-apps on the existing web?
And don't forget - even if Apple were to acquiesce and make an app store for web apps, there would still be much wailing and gnashing of teeth. The claim would be that Apple is trying to take away URL-addressibility from the web. Heck, people already complain about this just for Objective-C Apps!
We all remember, but they changed their minds and now fully embrace the app store, and even use it as a major marketing point. Suggesting we go back to it now is a bit pre-mature, I think. JS simply doesn't run comparably to native apps, so if they want to talk battery life, they should discuss it there as well.
> Apple has adopted HTML5, CSS and JavaScript – all open standards. Apple’s mobile devices all ship with high performance, low power implementations of these open standards.
Not too many words there between 'JavaScript' and 'low power'.
They may be low power compared to other implementations of the same, but I'd be very surprised to hear that the same program written in html5 and JS didn't consume more power than an app equivalent. Ever use Meebo? Ever look at how resource intensive it is, vs Adium or a similar program?
There are plenty of apps that utilize web views inside of native iPhone UIs, like tab bar controllers and other widgets. They are "hybrid" apps and make the best of both and they get by the review process just fine. A lot of them are fairly big names, like Reddit.
The point is that if you build your application correctly no one will even notice it's not a Cocoa app.
Take the iTunes App Store for instance. Nearly the entire application is a Webview, the AppKit equivalent of stuffing a Safari window into your application. It's all HTML (or an HTML-like markup language) and CSS. It's a little clunky sometimes but you can't say it doesn't "feel" like a real application just because it uses web technologies.
Your link makes it seem like an evangelist is playing loosely with words; "available for download" could simply mean beta/RC status, not ready for release everywhere.
Battery Life
From your link: The results are amazing: I can watch a 4 hours and a half flash movie, without interruption, with a bright screen (no sleep mode) and sound !!!
From Steve Jobs post: The difference is striking: on an iPhone, for example, H.264 videos play for up to 10 hours, while videos decoded in software play for less than 5 hours before the battery is fully drained.
They seem to agree with each other? And only helps Apple's point.
It is true that some of these links are from evangelists. But we're not saying Steve Jobs is neutral are we?
Battery Life: It's a bit of hyperbole. We don't watch videos on the iPhone for 10 hours. Would it matter if Flash could play videos for 11 hours? It is a misdirection or a sidestep of an original claim that Flash would kill the iPhone inside of 2 hours.
Touch/Scroll: Another sidestep. A Flash issue? With that I answer: a:hover {}
I just don't get this argument from Apple. Steve Jobs pioneered high usability mobile UI because Apple took the reigns. Now we have Android UI taking a cue from Apple and "unlocking the mobile market"
Flash and HTML5 have a similar relationship. HTML5 opens it up a bit. That's great. But what about HTML6? Flash is Area51 for implementation study.
We don't need to implement a marquee tag and then take it away once we realize it sucks. How could we market test video conf. or canvas-based visuals if we had no way to benchmark them? Flash is good. The VS. argument is a misnomer.
> "We don't watch videos on the iPhone for 10 hours."
We do other things on the iPhone. I don't want to make the choice between watching a video and then restricting my own usage just so my phone has enough juice to make a call. I'd rather watch videos and use my phone, and the more power efficient everything is, the more the user can do this.
To put it in perspective, if Apple's numbers are to be believed, running Flash alone doubles the power consumption of the device. That's insane.
> "With that I answer: a:hover {}"
a:hover is no longer (it certainly used to be) a common tool used for site navigation. Which is to say, rollover dropdown menus are out like parachute pants - it's a non-issue on the web today. Just about every site works with "clicks", which is Apple's primary form of interaction with websites on touch devices.
The most compelling argument made though we've already been aware of for some time: cross-platform toolkits encourage lowest common denominator design. In Apple's world of pixel-perfect mockups and obsessive attention to detail this is abhorrent - and I'm inclined to agree. The iPhone's chief strength is its remarkably polished user interface - which for the most part remains true across its 3rd party app sphere, thanks in no small part to Apple's obsessive HID enforcement.
A Flash-compiled-to-iPhone app destroys this in the same way that a Linux Qt app hastily ported to Windows is the same.
> a:hover is no longer (it certainly used to be) a common tool used for site navigation. Which is to say, rollover dropdown menus are out like parachute pants - it's a non-issue on the web today. Just about every site works with "clicks", which is Apple's primary form of interaction with websites on touch devices.
This is precisely what I am talking about. We slice the pear in the middle and somehow end up choosing one side over another. True, a:hover is a deprecated form of menu control in HTML, but it is in Flash, too. Again, how you use the tool. Damning an entire developer base based on miscreant use is not a compelling argument. Sorry, but it isn't.
That's being a bit generous in favor of Adobe - the majority of the Flash development sphere is practically what one would consider "miscreant use".
Here's my impression of the different sorts of Flash use, based on my journeys on the web:
- Videos/Audio: Easily replaced by HTML5, and better too. More readily hardware accelerated, behaves as part of the core browser instead of a hijacking addon (e.g., Ctrl+T while focused on a Flash applet = no new tab), and much better battery life on mobiles.
- Crappy restaurant websites and other crimes against humanity: These should just die, and HTML5 will hopefully inspire better design ethic than this. I don't buy the argument that hovering is no longer common used as a critical navigation feature in Flash - I see it everywhere, especially for crappy restaurant websites. Flash is also impossible to crawl which reduces the usefulness of the web overall. Hell, you can't even LINK to content in a Flash site. IMHO the "full Flash site" (as opposed to Flash being a component of a page) needs to die a fiery, painful death.
- Games: The only place where Adobe has even a remotely legitimate claim that Flash is an empowering product rather than a crutch.
When I think of Flash these keywords come to mind: slow, buggy, crashy, badly designed, usability nightmare, gimmicky, unprofessional, and just plain bad.
> That's being a bit generous in favor of Adobe - the majority of the Flash development sphere is practically what one would consider "miscreant use".
The vast majority of websites in general are poor. It would be no surprise to see that the majority of Flash sites would also be poor. Both sides have exceptions, though.
> Videos/Audio: Easily replaced by HTML5, and better too. More readily hardware accelerated, behaves as part of the core browser instead of a hijacking addon (e.g., Ctrl+T while focused on a Flash applet = no new tab), and much better battery life on mobiles.
I don't disagree. I am grateful that HTML5 can handle this, but it took a "standards" committee several years to roll this into a questionably viable model. The immediate fanfare for the slow implementation raises my eyebrow a bit. We're all happy about having something we've needed for years. The company that created the fold so we could have sites like YouTube today? The devil.
> - Crappy restaurant websites and other crimes against humanity: These should just die, and HTML5 will hopefully inspire better design ethic than this.
It won't. Cheese is forever. If HTML5 can accomplish cheese, people will create cheese. We're all hopeful, but the human condition comes into play.
It's not the cheese I object to - but that the Cheese breaks the internet in fundamental ways.
Example: We just wrapped up Restaurant Week here in Seattle where a bunch of local establishments offered up prix-fixe specials for a couple of weeks. The website for this event was done up all in Flash, and made it impossible to crawl for a search engine, or even to link or bookmark a particular restaurant for later retrieval.
What did we exchange for this cheese and gimmicky design? Everything.
There's no regulating for good taste - you're right, most websites in general are poor. They, however, are still accessible - neon green text on black background with an annoying MIDI track in the background is still information that:
- you can link to
- you can bookmark
- you can copy and paste
- you can search for in-browser (doubly important if the website fails at layout, e.g. most Flash sites!)
> What did we exchange for this cheese and gimmicky design? Everything.
I agree that if you don't follow the proper protocol you will end up with an undesirable site. I can't argue that Flash == HTML in terms of accessibility. There are ways within Flash to harness some accessibility but it is up to the developer to implement these. An informational site in Flash is a bad choice. I would never suggest the use of Flash for broad use. A fun site (like http://www.myspace.com/fanvideo) is more appropriate.
> - you can link to
> - you can bookmark
Named-anchors and proper implementation (same as AJAX RIA)
> - you can copy and paste
"Selectable text" is an option in any Flash text field.
> - you can search for in-browser (doubly important if the website fails at layout, e.g. most Flash sites!)
> - you can search for on Google/Bing
Flash exports text within to the container. Google also can search SWF.
"Selectable text" is an option in any Flash text field.
Is copying/pasting it to the OS (or app) pasteboard an option? If so, will it use the same UI as native apps have? Or will it force a funky widget of its own in the middle of an app that otherwise looks like the rest of the OS?
The options are the same as the OS using the same UI. Identical to right-click menu in browser inputs/address bar. It is using the OS clipboard. CTRL/OPTION+C/V works the same as well.
Are you saying that Adobe reimplemented the UI? Or that they pass it off to the OS to handle? Because if its the former, the iPhone doesn't have a menu dropdown for cut/copy/paste. Or a ctrl/command button, for that matter.
I'm only asking because I've never had copy/paste from flash work well on the desktop.
Because <canvas> is used for just the "flashy" part of the site but everything else is still in traditional HTML with all of its advantages. Sure that can be done in Flash too, but, as we've seen, people often end up just doing the vast majority of the site in Flash. There's not the same temptation with <canvas>.
Why not? If <canvas> is being used as a drop-in replacement for Flash, why wouldn't it just be used in exactly the same ways? There's nothing special about canvas that would make it less prone to abuse.
Flash includes a ton of controls: buttons, scrollbars, text areas, etcetera. It has its own scripting language and memory environment that is separate from the HTML page and its DOM/JavaScript environment that contains it. There are ways to shuffle data and events between the two, but that means going out of your way to do so.
So if you're going to have a Flash-y data visualization component, say, it's easiest and most natural to have the controls for it (buttons adjusting parameters, display of those parameters, navigational controls) made in Flash too -- just expand the content rectangle of your Flash component out and drop in the controls.
In HTML5, the 'environment' of the <canvas> object is that of the surrounding HTML page. Adding buttons and other controls around the edge of your visualizer is most straightforwardly done by using "native" HTML <button>s and other such elements.
So it would be surprising to see <canvas>-using pages end up being done in a way where the <canvas> is essentially the whole page — it would require more programmer effort — whereas in Flash that's by far the easiest way to do it.
Although i agree that there are a lot of issues wrong with Flash, some of the things you mention aren't entirely accurate. For example, you can link directly to interior Flash content with SWFAddress (http://www.asual.com/swfaddress/). Also, there have been greater strides made in crawlers so that Flash content is more easily indexed by crawlers.
I will be the first to admit there are a lot of poorly designed Flash sites, but from an interactive design perspective, Flash still gives you the most freedom to create more innovative ways for user interaction. IMO this isn't something that should be dismissed so quickly.
> We do other things on the iPhone. I don't want to make the choice between watching a video and then restricting my own usage just so my phone has enough juice to make a call. I'd rather watch videos and use my phone, and the more power efficient everything is, the more the user can do this.
To put it in perspective, if Apple's numbers are to be believed, running Flash alone doubles the power consumption of the device. That's insane.
What's your battery life playing an intensive game or using a processor intensive app?
"But we're not saying Steve Jobs is neutral are we?"
I would argue that that's a reasonable point to make. Steve Jobs doesn't have a vested interest in seeing Flash die, he has a vested interest in seeing the iPhone platform succeed. If he saw any way Flash could help that, I'm sure he'd be all for it, but he's seen how Adobe's behaved in the past (late with PPC versions, late with OS X versions, late with Intel versions, late with 64-bit versions, crashy Flash, slow Flash, etc.), and he has no interest in subjecting the iPhone to the same issues that the Mac platform has had to endure at the hands of a slow-moving corporation who has little interest in being a good citizen.
I would argue that Jobs is 'neutral' in the sense that he's not supporting anyone else's agenda. He's made his choice based on the facts before him, and while you may not agree with how he got there, I would say that he got there from an initial position of neutrality.
The very moment flash was not included in the iPhone, Steve Jobs had a vested interest in seeing Flash fail. If Flash expands, user satisfaction with the iPhone goes down. If Flash fails completely, one of the strongest negative components of the iPhone disappears.
Why does he have a vested interest in making it fail? Could he not change his mind and allow Adobe onto the platform? It's not like he's never changed his mind in the past (the Intel vs PPC 'Megahurtz Myth' comes to mind).
Yes, this goes back to 1989 fight over fonts, and the 1993 fight about Display Postscript and the 1999 fight about Photoshop on Windows, and the 2007 withdrawal of 64-bit Carbon toolkit at the last minute.
Battery Life: It's a bit of hyperbole. We don't watch videos on the iPhone for 10 hours. Would it matter if Flash could play videos for 11 hours? It is a misdirection or a sidestep of an original claim that Flash would kill the iPhone inside of 2 hours.
If a phone can watch 4 1/2 hours of flash in software before the battery is drained this means that watching a two hour movie will eat half of my battery. By comparison I can watch a two hour movie using hardware decoding and still have 80% of my battery left.
About Job's remark of hardware vs software decoding of video:
In both cases, he's talking about H.264. An end user might not even realize there's a difference. But some encoders produce H.264 that can hardware decode, and some encoders do not. Not all H.264 is created equal, and this is one area that really matters.
You want an H.264 encoder that produces hardware decoder compatible video.
When it comes to battery life, you must consider the affect on the total user experience.
If the iPhone can play video for 10 hours, that means a 2 hour movie brings my battery life to 80%.
If the iPhone could play Flash video for 5 hours and I watch a 2 hour movie, my battery life is reduced to 60%.
Given other things I do with my phone battery life is important beyond "how much video can I watch?" If the repercussions of watching a two-hour video are so much less severe, I am more likely to watch the video in question knowing I won't wind up with a dead phone after a half day of traveling.
Apple modified WebKit to support most uses of a:hover by changing how touch interaction works when a :hover style exists on an element. Apple cannot do this with Flash, which means Jobs's argument here is right on the money.
No, but you might watch a film and a couple of music videos and then the battery is getting awfully low. My iPhone lasts a day on one charge (with heavy web use & lots of calls). Halving that with a 2 hour film could be a big problem.
Proprietary isn't bad. Open isn't great.
Proprietary vs. open is bad.
Proprietary + open is great.
Proprietary implementations are a (sometimes) necessary interstitial step towards great open implementations. Save the "VS." narrative for Street Fighter II.
> "Proprietary implementations are a (sometimes) necessary interstitial step towards great open implementations."
True, but Flash is a closed standard with a closed implementation. Adobe likes to claim it is open - but the fact remains that not only is it impossible to write an open implementation of their incomplete spec, but Adobe has actively sought to keep core components proprietary even in spec and prevent competing implementations.
That's the part that irks me about Adobe's role in this issue - they're playing themselves and Flash off as this open savior of the web when they've actively been evil in that regard. If you're closed, great, do your thing; if you're open, then actually open your spec - don't open part of it and then go after the other half with lawyers.
So Adobe gets tossed aside in favor of HTML5/canvas. Or an open version of the SWF format with no licensing issues emerges. I am fine with that. If Adobe wants to be stupid, they are more than welcome to.
Apple is calling the kettle black with the proprietary argument. We can drill down to specific layers but the effects are the same to the developer coding on that layer.
OS has contributed to advancements in design, but proprietary initiatives employ a fleet of people to research, design and case study a technology. They do it all day and not as a side project after work.
Again, it isn't a case of one vs. another. I wish we didn't need to take a side and finally come to a conclusion that both are good in their own ways. Evolution is key.
There are plenty of people that are employed full-time to work on Open Source software. Don't perpetuate the myth that all Open Source software is created by people 'in their free time.'
The point is it will only ever directly effect products made by the same company. A device from HP or Microsoft will never be hurt by an issue with a closed Apple platform.
If there are issues with Flash, it will hurt every company using Flash, and there is nothing those companies can do about it other than bug Adobe to fix it or drop Flash support.
I don't disagree that Flash has the potential to create issues.
The argument just suffers from the fallacy of composition. The standoff motif is getting old. We're talking about software, not cold war against another country.
Apple doesn't want you to notice Steve behind the curtain. With Adobe we tend to pick and choose what we agree with and ignore the points that don't support our bias.
In what way is saying "Flash has problems, therefore putting Flash on your platform will expose it to those problems" a fallacy of composition? I don't see it at all. It's not wrongly attributing an attribute of one small part of Flash to the whole. Or are you saying the fallacy is somewhere else?
But you don't appear to have pointed out any specific flaws in the argument against it. So you're not making an argument, just an unsupported assertion.
All I am saying is that Steve Jobs isn't putting all the facts on the table as they are. He is feeding the anti-Flash gang mob with half-truths.
To say Flash is proprietary, possibly poorly executed format is a fair thing to say. To say it is completely without merit and has no place is a fallacy. HTML5 video is widely supported today. Flash might be a delivery vehicle for another unrealized idea tomorrow.
And the "death to the plugin arch" circle-jerk logic is almost intolerable. Only create within the confines of the parent layer's imaginative capacity? Sounds more closed than SWF.
I have no issue with him shutting out something that really sucks today at the cost of potentially missing out on something awesome to do with it tomorrow. The awesome scenario just doesn't seem likely to come to pass. If tomorrow there's a completely great thing on the web that can only be done with flash, then I think Jobs is pragmatic enough to reevaluate.
But the fact that the main use for Flash on the web these days, beside video and unusable websites, are animated flash ads that use up 40% of my much more powerful Macbook Pro's CPU time means I'm in no hurry to have flash on the phone, openness dogma be damned. When Jobs limits my general purpose computers, I'll be angry, but until then, I'm glad he's making the tough design decisions instead of just letting the device be a free for all.
> But the fact that the main use for Flash on the web these days, beside video and unusable websites, are animated flash ads that use up 40% of my much more powerful Macbook Pro's CPU time means I'm in no hurry to have flash on the phone
In my opinion, this is moot point. Annoying ads aren't going to be limited to Flash. HTML5 ads aren't going to be more fun.
My point was that that is one of the main uses of Flash, and that an extremely simple/common flash implementation is enough to burn a huge number of CPU cycles for no benefit.
No user benefit. But there's nothing stopping advertisers from making crappy HTML5 canvas ads, and those are guaranteed to burn even more of your CPU cycles. So I call shenanigans on that argument.
Apple controls the iPhone's HTML5 renderer and feels much more confident in its performance. "There's nothing stopping advertisers from making crappy canvas ads" is not the same thing as "canvas ads will suck at least as much power on average as Flash ads."
They were separate points. Canvas is inherently more computationally expensive than Flash at this stage of the technology. As canvas advances, so does flash -- unless Flash dies off completely, Canvas ads will always suck more CPU cycles than Flash ads.
"unless Flash dies off completely, Canvas ads will always suck more CPU cycles than Flash ads."
Who says they'll progress at the same rate? That's ridiculous. Flash is at least 10 years old and has been in extremely heavy use, so it's likely plateaued in terms of performance on most platforms. Canvas (and the rendering engines for it) are very new by comparison.
Fake Steve raises an interesting point about how close John Gruber and Apple are.
>"...we leaked the Flash essay to our Chief Apologist John Gruber and told him to start whipping up some buzz on Twitter about it, and he followed orders [http://twitter.com/gruber/status/12988151829]."
Apple do not have a Monopoly, unless you use a diseased meaning of the word Monopoly to mean devices produced by Apple. In that case, almost every company has a monopoly.
At a certain point, if you execute so much better than the competition and set yourself up differently (controlling and innovating in hardware too) and you maintain total control of your platform, you can dominate in the market to such an extent that you have a virtual monopoly.
Yes, but Apple do not have a monopoly in smartphones or computers. Maybe, just maybe you could say they have significant market dominance of mp3 player market, but who cares about that.
How is it so different? Microsoft at least did not control the hardware. Apple has always been worse than Microsoft in that respect - they even wanted to control the printer at one point! The only difference now is that for some reason its working for them now and iPhone is incredibly popular. But they're much more monopolistic than Microsoft ever was, and they always have been, and I would imagine they always will be.
I think Gruber remains independent from Apple.
It would be too conspiracy theory for Gruber/Daring Fireball to be a secret PR mouthpiece for Apple.
More likely Apple reaches out to him because he is by far the most influential Apple website outside of Apple (MacRumors is a distant 2nd in my opinion).
They don't have to pay Gruber anything; the increased ad revenue from being a member of the elite The Deck should more than compensate.
More likely Apple reaches out to him because he is by far the most influential Apple website outside of Apple
I also think he is one of the most levelheaded apple-centric websites out there. Gruber talks a lot about premium products (not just apple) and their success in various markets -- DF is like a premium tech/apple/good taste blogzine and I imagine that his style appeals to a lot of people at Apple, whether they reach out to him or not (and "legally" or not)
Gruber is fantastic when he's not in "scandal" mode.
The site quality degrades significantly, however, when he latches onto a "scandal", whether it be about stolen iphones or illegitimate wi-fi security claims.
With his recent ad nauseum stolen phone coverage, DF has felt more like an overwrought Entertainment Tonight scandal episode.
I've found the best Gruber reading algorithm is:
if (post < 2 paragraphs) { worth bookmarking } else { mark as read }
Apple's competitors have been clamoring to put Flash on their mobile devices because it's one more bullet point they can use in their marketing against the iPhone.
Yet it's 2010 and mobile Flash is still nowhere to be found. How is that not Adobe's fault?
Don't lose sight of the issue. 3.3.1 is the issue, not "Flash on the iPhone". By reframing the argument as "Flash on the iPhone", you are succumbing to misdirection.
So "Flash on the Web == Bad" because it's not open, but "Flash cross-compiled apps on iPhone == Bad" because it's not... open... ? The logic doesn't match up. As has been stated, it's a misdirection. Steve Jobs is setting fire to a house on the west side of town so that no one will notice that he's robbing the bank on the east side.
He's trying to come off as an 'open crusader for the web against Adobe' because 'Adobe is not open,' but on the other hand shoving Adobe (and others) off of the i{Phone,Pad} platform because it's not part of Apple's own 'not open' environment. I hate Adobe as much as the next person, but let's not dress Steve Jobs up as a knight in shining armour just because he also doesn't like Adobe.
"Flash cross-compiled apps on iPhone == Bad" because it's not... open...
Neither stated nor implied. The logic doesn't "match up" because it's a separate issue, addressed with different arguments:
"We have discussed the downsides of using Flash to play video and interactive content from websites, but Adobe also wants developers to adopt Flash to create apps that run on our mobile devices."
He then goes on about third-party middleware, e.g.:
"We cannot be at the mercy of a third party deciding if and when they will make our enhancements available to our developers."
Jobs does not make the openness argument in reference to Apple's proprietary platform at all. It seems to me that you have to ignore the last third of the article to think that he was.
He's trying to come off as an 'open crusader for the web against Adobe'
Is Jobs lying when he says Apple supports open web standards?
This has nothing to do with what's open and what isn't, it's about who controls what. Apple controls iPhoneOS, Adobe controls Flash. It's a prizefight. Don't get caught in a sucker punch just because you were paying attention to the fancy footwork.
I think it's almost entirely bullshit and misdirection. 3.3.1 doesn't apply to specifically to Flash.
"Open" is irrelevant. Apple is certainly no more open than Adobe.
The Full Web - whatever. Don't support it in the browser, don't support browser plugins, fine. Safari on iPhone still crashes all the time.
Reliability, security, performance - these are all things that apply to native apps built for the App Store, so why are apps built with Flash excluded in particular?
Battery life - more irrelevance. I'm pretty sure that there are games on the App Store right now that will suck away your battery's life faster than most cross-compiled Flash games. Game loops are intrinsically always busy - they don't necessarily have a natural idle point like GUI event loops do, especially if they are running at less than their desired framerate.
Touch - this is an application quality issue. I don't see any difficulty in simply not approving apps that have awkward non-touch interfaces.
The last reason, it being another layer, is the most insidious. UI difference, performance, etc. are all irrelevant - games have weird UIs, drain battery, and poor performance is a quality issue. The truth is Apple doesn't want to be disintermediated. Apple wants developers to be strictly dependent on its tools and APIs, and not use anything between the app and the platform. This is particularly offensive to me as a software developer, and it's far worse than anything Microsoft has ever tried to pull. It's the reason why 3.3.1 is phrased the way it is, and it has nothing to do with Flash and everything to do with locking developers in.
As I see it, Flash on the web is used for three things: (1) video, (2) casual games and (3) ill-advised corporate intro screens.
I'd love to see video not needing flash, but it isn't, yet. Having a sand-boxed third-party FLV player, so that it didn't need to infect the Safari process, would be nice to have. I don't really see that approving such a process would be necessarily harder than approving any other native application on the App Store.
I'd love for (3) to disappear. I can't bear those things.
And it would be great if HTML5 was universally usable and performant enough to work for (2), but it isn't, yet. I know of many casual Flash games, though, that would work really nicely as apps, and Apple is hurting me personally by increasing the barriers preventing those apps on the web from being available on my device. This policy is both anti-consumer and anti-developer. Only in the short-term is it pro-Apple, but I hope and expect Apple will fade in the longer term in competition with Android.
>The truth is Apple doesn't want to be disintermediated. Apple wants developers to be strictly dependent on its tools and APIs, and not use anything between the app and the platform. This is particularly offensive to me as a software developer, and it's far worse than anything Microsoft has ever tried to pull. It's the reason why 3.3.1 is phrased the way it is, and it has nothing to do with Flash and everything to do with locking developers in.
Microsoft did the same thing with DirectX and OpenGL. OpenGL was cross-platform. Microsoft stopped supporting it, and instead insisted that developers use DirectX. This essentially killed cross-platform gaming for Mac/Linux. DirectX games port between Microsoft platforms and Microsoft's Console, but require significant middleware to port to OpenGL and non-Win32 platforms.
Use of an API, public and documented, is a form of dependence and lock-in. Apple has done. Microsoft does it. Oracle does it.
There's nothing stopping you from using an OpenGL api that's built on top of DirectX, though. What Apple has done here is say that you can't put an API between you and the underlying API such that you are insulated to some degree from platform changes; but even worse, they've mandated which languages (!!!) you can use to interop with that API, which is almost criminal, technically speaking. (Of course it's not de jure criminal. Just for the reading comprehension impaired.)
Besides the fact that Flash is closed and proprietary, has major technical drawbacks, and doesn’t support touch based devices, there is an even more important reason we do not allow Flash on iPhones, iPods and iPads. We have discussed the downsides of using Flash to play video and interactive content from websites, but Adobe also wants developers to adopt Flash to create apps that run on our mobile devices.
We know from painful experience that letting a third party layer of software come between the platform and the developer ultimately results in sub-standard apps and hinders the enhancement and progress of the platform."
Agree with him or not this is his view. And I think its a valid view. The first macs with a mouse didn't have arrow keys. And the early generation touch devices don't allow intermediate steps.
This is not stopping anyone from compiling from lisp to html5. Apple is fine with this. Buy native apps on their platform, for the time being, wont have intermediate layers.
What? The legal notion of "legal personhood" and actual personhood are not comparable, not even legally, and especially not in the context of the discussion at hand.
"This legal fiction does not mean these entities are human beings, but rather means that the law allows them to act as persons for certain limited purposes..." Emphasis mine.
This stuff isn't even on topic, but when people insist on insulting my intelligence with snide pseudo-logical comments, I will insist on correcting them.
Regardless of what Apple think is best for the users, I would still prefer to have Flash on my iPhone. Not only to be able to use websites made in Flash, but also so that developers have a way to develop software without going through the App Store.
hmmm I too agree with this post somehow.... since long I am also not comfortable in learning flash and creating flash based website.. though the graphics is amazing but very slow.... the current decision by apple will definitely hinder the progress of Flash... Adobe need to come up with it's own version of mobile phone :) in the post Steve Jobs mentioned not to encourage the third party layer.... so they ll be slow to adapt the platform changes.... But today got mail from Appacceletor Titanium, they are quite happy bcoz they are developing based on JavaScript.... not sure..
What he describes as the "most important reason" is also the big mistake here. If using certain libraries or cross-platform development tools truly did result in worse apps, then the market would take care of that.
He also doesn't seem to understand that such tools do not all totally hide the underlying APIs. They fall along a continuum in that respect. And in trying to ban extreme cases like Flash, Apple has also basically banned compilers.
Every programmer I've talked to about this thought it was some combination of stupid and evil (there ought to be a word for that; stevil?) to insist that apps be written only in C, C++, and Objective-C.
Surely there must be people fairly high up in Apple who realize what a crazy move it is to try to dictate to hackers what programming language they should use. It worries me that those people aren't being heard.
I'm starting to suspect that Apple really requires proprietary APIs and decrepit languages for the same reasons Spolsky[1] once said Microsoft used proprietary and huge and ever-changing APIs. It's cover fire to make everyone else's projects soak up more hours, increasing the fraction of captive developers and shops who are only experienced in iPhone OS work, rather than being prepared for work on iPhone OS and other platforms.
I'm starting to suspect that Apple really requires proprietary APIs and decrepit languages for the same reasons Spolsky[1] once said Microsoft used proprietary and huge and ever-changing APIs.
Microsoft finally addressed this and stabilized their APIs but they were somewhere between 5 and 10 years too late. Circa .NET 2.0 + in my opinion as an ex-.NET programmer.
Apple could easily end up in the same situation if they continue down this road.
Every programmer I've talked to about this thought it was some combination of stupid and evil (there ought to be a word for that; stevil?) to insist that apps be written only in C, C++, and Objective-C.
Ya it's insane. I have no idea how they can be this out of touch with the development community.
Even the Microsoft developers that I know think it's evil and restrictive. Microsoft has failed in a lot of ways in the past but one thing they've gotten right is .NET. It's relatively open and they are beginning to actively support other languages. So instead of .NET and the CLR dieing off people can innovate on top of the runtime and continue to compete at some level with other dynamic languages. The same goes for java and the JVM.
My biggest fear now is Apple will decide to slowly but surely drop OS X support and begin the transition of moving the majority of users over to a new locked down stevil new Operating System.
I do think it is reasonably evil, but I'll be very surprised if, say, Apple enforces this against somebody writing software in a lisp, or all the games using Lua or Unity. But saying literally "You can't write software in Flash" makes for somewhat questionable legal language.
One of the regulatory tests for abusing market position is unevenly applying rules to lock-out certain competitors. If Apple does allow things that are technically forbidden, but ban flash, they may open themselves up to an anti-trust lawsuit.
I'll buck that for you. I don't just dismiss it as stupid and evil.
There are 2 things that stand out to me which aren't being discussed. 1) Apple has maybe the very best record in supporting their customers' software through various multiplatform upgrades. If you've been on board with them, they've taken you from 68000 to PowerPC to Intel and they've had bloody seamless transitions. Perhaps that has something to do with their desire to specify how apps are built and which tools are used to build them. FWIW, they did their own iPad chip but they bought one off the shelf for iPhone, I'd imagine that they will have their own phone chip before too long. This just seems logical and sensical to me and it's possible they might alter the architecture in some way and maybe want to dynamically translate the existing apps.
2) If you look at iPhone and iPad, they clearly had some foresight and vision when they did the phone. They didn't accidentally build the iPad and it turned out to run the same software and be compatible. Perhaps there is something else that they have in mind in the coming years or two. Maybe you'll be able to use app store apps on AppleTV. Who knows, maybe there will be some other device too.
The market might summarily "take care" of iPhone itself.
There are many people (young people) who don't know anything about design or hacking, but they know enough to make an app in Flash--sort of. And when people buy those apps that look, feel and work like crap, they'll think iPhone itself is crap. Sure, they might pay for more expensive apps (those that are still in business), but once you get in the habit of paying little, it's hard to start paying a lot. I still buy cheap motherboards and curse myself later. And, once you get in the habit of getting crap, it's hard to expect anything more for your money. At this point, Apple has no edge over Android. Et cetera.
I can see a strong case for why it's evil, but to show that it's stupid would require thorough understanding of what a clueless, fickle public would do. Maybe the business morlocks at Apple are better at that than we are, and maybe we're better for it.
I can build a flash app entirely for free, legally, with zero restrictions, forever, on the tools I used to write it, or where I can deploy it to. Adobe has bent over backwards to make players (also free) backwards compatible. That's pretty open.
I must pay to build an iphone app, which I can only do on a computer owned by the same company, running an OS owned by the same company. This app only runs on hardware made by said company, and only run on an OS by said company. I am restricted in how I can build this, and must live by the whims of an opaque editorial process. In terms of openness, this makes Bill Gates look like RMS.
Apple makes Bill Gates look like Richard Stallman?
Let me ask you something. How many of Microsoft's competitors are building products that compete directly with Microsoft core offerings using the source code to Internet Explorer?
That's Opera Mini -- most other mobile platforms have the full Opera Mobile browser available (or soon). That browser is explicitly against the rules of the app store (as it runs interpreted code downloaded off the Internet).
[Edit: What's with the down votes -- this is factual information]
None, because the source code to Internet Explorer isn't based on an LGPL library and so can be kept proprietary.
Now, how many Microsoft OSes refuse to run in a virtualization environment on non-approved hardware? How many treat such execution as a license violation?
So your argument appears to be that Apple, having chosen to build the most popular (by vendor support) browser library on the market on a GPL codebase, is less open than Microsoft.
No, my argument is that your argument is misleading because it ignores the fact that Apple doesn't have a choice in keeping Webkit open. It's possible that they would want to and possible they would not want to, if they had a choice. When they originally chose KHTML to base Webkit on, in 2002, years before the first iPhone, iPad, iPod Touch, iPhone OS, etc., their attitute towards being open may have been different than it is now. No one is saying Apple were less open than Microsoft in 2002; people are saying it's less open than Microsoft now, and you counteract with ignorant, invalid arguments while ignoring the evidence you're presented with. You're behaving like a troll.
Apple has been a (or, THE) driving force behind Webkit for some time now. They weren't forced to start with LGPL code. They did because they were comfortable working under the GPL.
I object to the subtext here, of "Apple only does open source when they're dragged into it". No, in fact, Apple has been a vocal and active open source contributor for many years now, to a far far greater extent than Microsoft.
I started with WebKit because I think it's especially relevant to this conversation. WebKit is an important gift to all of Apple's mobile competitors. But move on from WebKit and there's still:
* LLVM/Clang
* OpenCL
* Grand Central Dispatch
* Darwin
* Quicktime Streaming
etc etc etc.
Name one bona fide open source project that Microsoft runs that any of its competitors in any market have been able to use. I'm sure there's at least one. Maybe if we go project-vs.-project, Microsoft will even come out ahead. I really doubt it, though. So for now, just one.
That's all true, but in the last few years Apple does seem to have undergone a rather alarming shift away from open systems.
I don't think anyone is disputing that they've done a lot of nice open stuff in the last few years, since jumping on the bandwagon in a big way with Darwin ... but that sure doesn't seem to be their M.O. anymore.
It's beginning to look like that whole attitude was borne out of desperation and a need to get MacOS in line with the times, and now that they're not against the wall it's back to the same old sealed-box model that was driving people nuts twenty years ago (and which was arguably a big part of why they ended up in trouble).
They just shipped GCD. FreeBSD has already ported it.
Meanwhile, in their most hotly contested market --- the one, by the way, where Apple makes more money than any other --- over just the last two years, Apple code has been adopted by virtually all of their viable competitors.
I don't think you can support the argument that they're shifting away from open source with any credible evidence, but I'd appreciate any stab you take doing so.
My concern is that all of their new products seem to be sealed boxes.
Yes, they still make Macs, and still sell MacOS, and MacOS is still fairly open. But it seems like the Mac is quickly becoming a legacy product, and Apple is no longer a computer company but instead a personal consumer-electronics company.
And this chart (http://www.businessinsider.com/chart-of-the-day-in-case-you-...) suggests that the Mac is no longer Apple's biggest product in terms of revenue. The locked-down iPhone is, with the iPod coming in for a close third. Add in the iPad and it's going to move even more in favor of their closed products.
I've always had a soft spot for Apple since they came out with OS X and cheated what most everyone assumed was corporate death, and I'm writing this on a MacBook right now. But I find their other products less than compelling, and since it seems that the direction they're going in is more towards sealed boxes and away from the open-system approach of the (post-2000) Mac, I'm not sure how much of their gear I'll be purchasing.
I do own an iPod just because I haven't found a player that's as good, in terms of being well-integrated with software, and the iPod at least doesn't restrict you to only loading music that you buy from iTunes. (I do wonder if, were Apple to bring out the iPod today, if that would be the case.) But I don't have an iPhone and I don't see myself purchasing an iPad. And their last few updates to OS X haven't felt like they're keeping up the pace of speed and feature improvements that they had been; it seems like it's the red-headed stepchild now.
There's a serious question in my mind as to whether the MacBook I'm typing this on will be the last one I ever use.
It's just disappointing because they brought out a lot of neat stuff.
They weren't forced to start with LGPL code. They did because they were comfortable working under the GPL.
They did it because the alternative - writing an HTML rendering engine from scratch - was not an economical option for them at that time.
On the whole though, I agree with your point that no other company has released and actively maintained an open source project that is directly benefitting it's competitors.
Though I would tend to believe that's more of an accident of history than any sort of belief in the open source movement.
"No, my argument is that your argument is misleading because it ignores the fact that Apple doesn't have a choice in keeping Webkit open."
Well, not to be pedantic or anything, but WebKit didn't have to be open at all. WebCore and JavaScriptCore (based off KHTML and KJS respectively) needed to stay open-source, but the WebKit framework that unites the two is a from-scratch Apple project that was open-sourced in 2005.
It could be argued that all the work happens in WebCore and JSCore, but the WebKit framework on top of them makes dealing with the two far, far easier than it would be to use them alone.
The codebase is arguably more open but the platform as a whole is not.
I think the closed platform thing is a lot more troubling and has the potential to be worse for the market as a whole than if they had decided to build a proprietary browser but allowed other browsers on it.
(This is not to defend Microsoft's portable strategy, though; they've got the worst of both worlds. They have a shitty proprietary browser and then don't let you install anything else on it, AFAIK.)
Can you really build flash apps for free? Honest question — I thought you needed to have Flash.
If you do need Flash, you have things backwards. For one, he's talking about web apps, not iPhone OS apps. Web apps are entirely open and Apple is huge on pushing what you can do with the web.
Also, building iPhone OS apps is absolutely free. The entire SDK and documentation is free of charge, you just need a Mac to develop on. You need a $100 membership to sell iPhone apps, which is very different from needing it to try out development.
Flash on the other hand, costs $700.
Flash is entirely closed. Apple is mostly closed and is totally up front about it, but they completely support open web standards.
You realise that Adobe lets you make any components you want EXCEPT FOR THE PLAYER? Why, because they want to retain control of it. I mean, they publish the spec, people have made players before, and they have been told by Adobe to stop.
They used place that restriction on those accessing the .swf spec, but that restriction has been lifted. I'd have to check, but I think it's been several years now, but it's certainly been more than a year.
> For example, although Mac OS X has been shipping for almost 10 years now, Adobe just adopted it fully (Cocoa) two weeks ago when they shipped CS5.
right. And the Finder - a piece of Software central to OSX was using Carbon until... when exactly? At most 6 months prior to when CS5 came out.
iTunes - a corner stone of the iPhone software world? Still on Carbon.
If I hate one thing, it's hypocrisy.
Aside of that, nice arguments and well-written. Also, from this perspective, MonoTouch and Unity might be in the clear too as they are not cross-platform toolkits reducing themselves to the least common denominator.
"Also, from this perspective, MonoTouch and Unity might be in the clear too as they are not cross-platform toolkits reducing themselves to the least common denominator."
Hell yeah! Let's pray. I'm not interested in using MonoTouch because I want to make a cross-platform App that also runs on iPhone/iPad. I'm interested in using MonoTouch because I prefer C# to Objective-C, if only because that's where the bulk of my experience is.
Apple purposefully wrote Finder in clean Carbon as an example to developers, to prove that it was possible to make good cross-platform software. iTunes was also written very carefully -- and very cleanly -- in Carbon to be cross-platform.
Adobe, however, hacked Creative Suite into Carbon with buggy, inefficient code and then took 10 years -- ten years -- to schedule the time to fix it.
It's true. Its fine on my MBP, but I can't stand it on Windows. It feels sluggish on a quad core processor. How that is even possible I don't know, but it does.
iTunes wasn't written in Carbon to be cross-platform with Windows--if anything, it would be easier to make a Cocoa app cross platform because you just need to port Cocoa, which is basically just OpenStep.
It was written in Carbon to be cross-platform with Mac OS 9. And because it was just SoundJam MP, which was written for Mac OS pre-X. (Though, to be fair, so was Photoshop.)
"Though the operating system for the iPhone, iPod and iPad is proprietary, we strongly believe that all standards pertaining to the web should be open."
Adobe can just as easily say:
"Though Flash products are proprietary, we strongly believe that standards pertaining to what software is allowed to run on your computer should be open."
It's doublespeak either way, and it's crazy to think that either of those two viewpoints is good for the industry.
edit:
Also: "If developers need to rewrite their Flash websites, why not use modern technologies like HTML5, CSS and JavaScript?"
Flash can handle touch events just fine. Rewriting all that flash code in H/C/J is a monumental task compared to just updating a few flash events.
Basically, this is a lot of corporate spin and not a lot of sound arguments.
Exactly. Apple vs non-Apple (or more generally, things a company controls vs things a company doesn't control) is the only relevant point here. Everything else about open standards and performance woes is just spin.
Point 1 is talking about websites / web applications. Point 5 is talking about native apps. Two different domains entirely.
With point 1, if developers are going to make a website, and they ditch flash, their work would work everywhere well (iPhone/iPad/iPod, Android, WebOS, Symbian, and Blackberry OS 6, as well as on the desktop). They wouldn't be limited to the platforms Adobe decides to support (or decides to support well).
With point 2, if you're going to make a native app for the iPhone, make one for the iPhone. Don't make one with Flash, which may not get around to supporting things like GameCenter, or multitasking, or any of the other hundreds of things Apple adds in new releases.
Choosing Flash for either one makes a substandard experience compared to the other options available.
I do understand your point about using the right tool for the right job, however Apple is arguing both sides. Either cross platform standards are a good thing, or platform specific features are a good thing.
While web applications and native applications are different domains, they are not different enough to mean different reasoning can be used. An end user doesn't really care if something is built using HTML5, flash, or compiled natively. They just care that it looks cool and does cool things.
Either cross platform standards are a good thing, or platform specific features are a good thing.
This is the classic fallacy of exhaustive hypotheses of the 2nd order, a.k.a. "a false dichotomy" or "the fallacy of the excluded middle".
There is a third possibilty: that both are good, and that which is better depends upon the nature of the app one wants to create. This third missing case also happens to be the reality of the situation — there are tradeoffs, and one selects which is better according to the upsides that one desires versus the downsides one is willing to abide.
(Surely you're not claiming that there are no tradeoffs between the native API and web technologies! I can think of several, not the least of which is the limited rate at which innovation can be brought to the web API.)
Cross-platform standards are a good thing for Apple. Getting people to use Apple's platform specific features is a good thing for Apple. Apple does what is good for Apple. There is no contradiction.
Remember, this is not an essay about why everyone else should support Apple's actions. It is an essay about why Apple thinks its actions are good for it.
"Apple does what is good for Apple. There is no contradiction."
I don't even think Jobs tried to hide this very well. It's pretty clear he is arguing from the perspective of what is best for Apple, and clearly and coherently explaining why.
Point 1 states that ubiquitous Flash co-opts the rate of advancement of the web.
E.g. You can't effectively differentiate, compete and move everything forward by supporting, say, h264 because Flash is already everywhere.
Point 5 states that ubiquitous Flash co-opts the rate of advancement of the mobile platforms it runs on.
E.g. you can't effectively differentiate your device by adding a new sensor, API or capability, because Flash won't bother to support it until the hardware is standard.
It's the same basic argument: Flash becomes the platform, it's not a particularly good one and it's not a particularly good situation for developers.
Would you rather Apple compete with Google, RIM, Microsoft and HP/Palm for developers, or have them compete with interchangeable Flash-playing-boxes, leaving developers to Adobe's whims?
"If developers need to rewrite their Flash websites, why not use modern technologies like HTML5, CSS and JavaScript?"
If you have to make significant changes to the UI portion of your application, why not thrown the entire thing out and start from scratch with an extremely new (less mature) technology stack that you're probably complete unfamiliar with.
I believe Netscape got similar advice in the nineties.
> [Jobs:] "we strongly believe that all standards pertaining to the web should be open."
Consider Python. Is it a standard? Yes, because there's an open source reference implementation.
Does it pertain to the web? Yes it does, because http clients and servers can be, and often are, written in this language, and its standard library and 3rd-party open source libraries support http and other web protocols such as HTML.
So if you're so keen on stand ards Steve, why can't I write web software in Python for the iPhone and iPad? Why are you such a hypocrite?
On a wider note: Steve, do you want to spend the rest of your life selling sugar-coated digital handcuffs, or do you to change the world for the better?
"why can't I write web software in Python for the iPhone and iPad?"
Ummm... you can. If you write 'web software' in Python, all an iPhone/iPad user needs in order to use it is the URL. Both devices have a best-in-class web browser that would be more than capable of running your web software well.
Is python really a 'standard'? Python's had trouble lately with non-mainline implementations because the grammar keeps changing. There's no standards document laying out how the language should behave, in the sense of 'Jython is compatible with the Python language 2.5'; rather, third-party implementations duplicate features and functionality from CPython based off the PEP they're defined in.
Does it pertain to the web? It doesn't. Python isn't used 'on the web', it's used on the backend to generate web pages, as is C#, Erlang, and pretty much every other language ever. Are you suggesting that Apple should include a Python interpreter in WebKit? It sounds like that's what you're suggesting.
The simple answer would be 'because Python is not a language that is used in client-side scripting anywhere on the web'. No other browser supports it, and even if they did it wouldn't be widely used.
Your argument seems not only spurious but nonsensical. Python is used to send or receive documents over HTTP, so therefore I should be able to write a web app in Python to run on the iPhone? It doesn't make sense.
My home cable internet 'pertains to the web' too, and it uses documented standards like DOCSIS, so why doesn't my iPhone support DOCSIS? It's a ridiculous argument, as is yours.
In an important sense I think it is. If I write a program in Python I can be reasonably sure it will run on someone else's system (obviously I have to take into account different Python versions, different underlying OS, etc). If I write a program in C++ or Scheme, what assurances of portability do I have if someone else's system is running a different implementation, with different libraries available, and different underlying behaviour in edge cases.
> rather, third-party implementations duplicate features and functionality from CPython based off the PEP they're defined in.
Yes; CPython is the standard. It's better if the standard is a program rather than a document, because you can run a program and find out what it does, but a document just sits there and does nothing (and may be hard to understand or ambiguous; the spec for Algol 68 comes to mind here).
> Python isn't used 'on the web', it's used on the backend to generate web pages
It is used as an http client as well as a server.
> Are you suggesting that Apple should include a Python interpreter in WebKit?
No, I'm suggesting that Apple should open up their platforms and allow people who have bought iPhones and iPads to run software of their own choosing on them; I'm sure someone would port Python fairly quickly.
> The simple answer would be 'because Python is not a language that is used in client-side scripting anywhere on the web'.
That's factually inaccurate; I personally have written code in Python that acts as a web client. As have some of my friends.
> No other browser supports it, and even if they did it wouldn't be widely used.
You seem to be under the misapprehension that client-side web programming must involve running inside a graphical web browser. This is not the case. Ever heard of wget? Or scraping?
> Python is used to send or receive documents over HTTP, so therefore I should be able to write a web app in Python to run on the iPhone? It doesn't make sense.
Yes it does. For example, I might want to write a program to cache certain web pages so I can read them even if I'm in an area with poor wifi and 3G access. And I might want to write that program in Python.
The sad thing is Steve, along with many others, totally, entirely believes that selling sugar-coated digital handcuffs is changing the world for the better.
A reference implementation is a good test suite but a poor substitute for a standard. It's huge and you can't just sit down and read it, you have to simulate it in your head to really understand what it would allow or demand. (This is also an awful problem with the HTML5 spec.)
On the contrary, its a written standard that you "have to simulate in your head". If you have an implementation, you can simply run the code and see what it does!
If Jobs had said "We strongly believe that you should be able to develop on the iPad and iPhone using all open standards pertaining to the web," you would have a point. Your logic is all twisty.
"We only want to support open standards" does not imply "we want to support all open standards."
I guess by "standards pertaining to the web" Apple means HTML, Javascript (ECMAScript to be precise) and CSS.
What do you mean by "standards pertaining to what software is allowed to run on your computer"?
Packaging systems and their associated programming languages / APIs. Like Android .apk files, Apple .dmg files, Microsoft's .msi. All open and usable by anyone who wants to build one (though some fees may apply.)
Less of an explicit standard and more the previous consensus that the company that sells you your computer shouldn't retain control over what kind of legitimate software you can run on it after it's in your possession, based on some self-aggrandizing notion that they know what's best for you better than you do.
That's exactly the mentality that keeps 'users' stupid.
Apple is basically saying: "users are stupid, let's protect them from thinking." Imagine if we said that about our kids: "Our kids don't know how to handle the real world. Let's protect them from having to deal with it." (never mind that most parents do in fact go through this phase).
The reality is that users are just kids who haven't learned how to use computers. Two factors make this hard: most computer software is generally poorly designed (speaking in terms of the number of poorly designed products vs the number of well-designed products), and most software is not designed to teach users how to use it.
Apple is busy buying fish for starving people. The company that teaches users how to fish is the company that will win big.
That's not about keeping users stupid, it's about not making them worry about stuff they should not worry about.
The user may be the world's best neurosurgeon, does that mean that they have to learn about filesystems?
most computer software is generally poorly designed
You are right about this one. But the thing is that Apple does exactly that: offering well designed software.
It's not about buying fish, it's about hiding unneeded complexity. How do you drive the car: press the gas and it goes, press the brakes and it stops. Turn the wheel to the right and it turns to the right. You need zero knowledge about what's going under the hood.
Now take the iPad: tap an app and it launches, press the home button and it stops. Swipe to the right, swipe to the left…
"The user may be the world's best neurosurgeon, does that mean that they have to learn about filesystems?"
There's a bit of a divide as to what 'personal' computers are being used for. The original mainstream use was to create/edit files using programs. The modern use is to interact with other people via the internet. So no, the neurosurgeon doesn't have to learn about filesystems because filesystems are mostly becoming irrelevant.
I don't think software should be designed to educate people about the trappings of decades of computer cruft. I think software should be designed so that using the software teaches the user how to use the software. Apple's method (re: iPhone/iPad) seems to be to design software that doesn't invite learning, and at the additional cost of limiting functionality.
Apple's method (re: iPhone/iPad) seems to be to design software that doesn't invite learning
Could you clarify this, perhaps by comparison to something else? Everything in my experience strongly supports the notion that Apple's general approach leads to much greater levels of competency and independence than anything else out there right now.
Turn the wheel and it goes where you want it to. But if you want to go to Google Voice, you'll have to take a long detour few people know of. You see, the road to Google Voice hasn't been optimized for your comfort. Or Apple's profit.
More good news about 3.3.1 "must write in C, Objective-C or C++" clause: it's made even more obvious here that Apple only cares about commercial middleware UI-driven applications. Those of us with custom toolkits and compilers (and games developers) can probably relax. Though, I wish they would still change the wording. I dislike having to wait for my friend's cat to click the "I Agree" button, since I cannot actually agree to the terms as worded and not break them.
Well 3.3.1 is worded poorly then, you'd think their legal team could do better if they just wanted to stop Adobe. Its pretty explicit about banning other languages.
Yes, it's stupid. But with Apple, I feel I only need to fear their intention, not their wording. The intent here, it seems to me, is to stop Adobe from making crappy commercial middleware for Apple's platforms. The wording is overly broad and stupid, but I'm no longer as afraid of it.
If they were intent on getting rid of everything under the sun that isn't straight C, Obj-C or C++, there would already be an App-ocalypse underway.
(I was going to say "App Holocaust" first, but that seemed even less tasteful.)
I feel I only need to fear their intention, not their wording
I'm leery of their capriciousness. They obviously have no qualms about changing their terms of service on a whim, giving folks only weeks to comply, and not accepting any compromise. That is simply not the type of folk I like to do business with.
The intent here, it seems to me, is to stop Adobe from making crappy commercial middleware for Apple's platforms.
This is actually the biggest part of Job's argument that I don't buy. If their intent was actually to stop crap from being produced, they would just reject apps that are crap from the App Store as they do now.
It's a question of technical vs. subjective evaluation. A lot of apps on the app store are technically acceptable (i.e. they do things right), but do useless or stupid shit (i.e. they're junk). Apple's only lately started to police the actual purpose of apps, rather than their technical merit (or compliance with the HIG).
I suppose making this rule more explicit was the goal. The #1 complaint about the app store rejection system is that it is not transparent or explicit.
Still, 3.3.1 is poorly worded, so it's not a great solution.
"Those of us with custom toolkits and compilers (and games developers) can probably relax."
Which part of the submitted article supports this? (genuine question not snark). I see a lot of flash bashing but nothing indicating you don't have to " write in C, Objective-C or C++"
Not the parent but the article says: "The most important reason... We cannot be at the mercy of a third party deciding if and when they will make our enhancements available to our developers. becomes even worse if the third party is supplying a cross platform development tool. The third party may not adopt enhancements from one platform unless they are available on all of their supported platforms. Hence developers only have access to the lowest common denominator set of features." Suggesting that the primary reason for the clause is that it will prevent third party apps effectively blocking enhancements made the the development environment; if you are using your own tools this isn't an issue. Not sure I agree that this means apple will be happy with developers using tools outside Objective C, but it certainly could suggest it.
The rationale that Apple doesn't want developers shackled by third party development tools of which they have no control suggests that if you have control (e.g. you embed a Lua interpreter in an otherwise native app) or you still have full access to all the goodies (e.g. You use some sort of game engine for the guts of your 3D game) then the clause is not intended for you.
Still, it would be better for them to spell it out more clearly or have a mechanism for you to get a preliminary ruling before you embark on development.
We know from painful experience that letting a third party layer of software come between the platform and the developer ultimately results in sub-standard apps and hinders the enhancement and progress of the platform.
IMHO, this sentences leaves no doubt that Apple does not only target Adobe, but any kind of layer they consider harmful for their platform.
There's absolutely no need to specify languages for the purpose Apple is claiming they have in mind. They could simply say that you must not use an intermediate layer between your code and their API. There's no reason for them to not want you to write the application in Scheme (as an example).
Unfortunately in legal situations, it's often the letter of the law not the spirit of the law that carries weight. It's been pretty obvious from the beginning that Apple wanted to block Flash, but not, say, Unity. I think they'll find it hard to only enforce it against certain infringers though.
Imagine if a government introduced a law that said no-one was allowed out on the streets after dark, but then only actually enforced it against Hispanics.
Maybe, but it unfortunately is the way life is in many places, and it serves to make the point pretty good. Selective enforcement of this policy is a real risk.
There is an even more expressive quote for Adobe to peruse: "Perhaps Adobe should focus more on creating great HTML5 tools for the future, and less on criticizing Apple for leaving the past behind."
So that's one less reason to support Flash itself on an emerging platform (iPhone is pretty mature, but it is still very young compared to desktop OSs - see the bounty of new features and UI changes coming to iPhone OS4).
I worry that they are just retargeting Flash to HTML5/canvas instead of redeveloping/expanding the Flash production environment as a true standards-based authoring tool. It makes me think of exporting HTML documents from Word (or to be kinder, OpenOffice). Yeah technically it is [HTML5/Canvas/CSS3/whatever] but it's still the step-child, and will be treated as such. Given the HTML & CSS that Dreamweaver coughs up at the moment, my hopes are not high.
I'm also concerned about their choice to introduce FXG for web use instead of using SVG. The Wikipedia article on FXG links to a blog post by Mark Anders where he contrasts the role of the format vs SVG by saying "FXG was designed as an exchange format between tools that understand graphics, like Photoshop, Illustrator, or other 3rd party ones, and tools that understand Flex and MXML." Right. So why are they now introducing it as a format for the web (via JS/canvas)?
The sad thing: CS5 Flash could also export to native iPhone apps. A native exported app would run faster, consume less power, and potentially have better integration with the underlying platform, all with exactly the same interface as an HTML5 web app... In other words, a native app exported from Flash would be better than an exported HTML5 web app in all the same ways Steve describes HTML5 video as being better than Flash video.
Actually, the native-exported apps are mentioned by Jobs in point 6. Of note: these apps were slower than native apps, consumed far more power than native apps, used more memory than native apps, and had little to no integration with the underlying platform, other than running in the first place.
Jobs' point talks about this; the issue is that if Flash CS5 supported what the iPhone can do today, who's to say when it will support what iPhone OS 4 can do tomorrow? Will they update Flash on Apple's schedule, or their own? Or at all?
For one example, look at GameCenter, Apple's new XBox Live-type service. Would Flash support it? Would Adobe go to the trouble of adding full support for GameCenter to their Flash runtime, so that games written in Flash could make use of that feature?
That wouldn't make any sense, because then people would have to write a different Flash app for the iPhone than for Android; either developers use GameCenter on iPhone and nothing on Android, or GameCenter on iPhone and their own solution on Android.
My suspicion is that Adobe wouldn't implement it (because it's a lot of work to just support one platform) or wouldn't implement it well (because they just don't really care), and that if it were available developers wouldn't use it (because it would only work on one platform anyway). This leads to a worsening of the overall experience, as the flood of quickly and cheaply ported Flash apps into the App Store dilutes the value that's there.
Perhaps Adobe would add GameCenter support, but that might be a new feature for CS6, which might come out (for example) August 2011, after Apple's released iPhone OS 5, with more new features Adobe hasn't had time to implement.
Developers using Flash would have a substandard environment, and would produce substandard apps, assuming they even cared at all about doing things right. That's what Apple's trying to prevent.
I don't see Jobs addressing the performance of CS5-exported apps in point 5 ("Fifth, there’s Touch.") or elsewhere. The only place he says anything that could refer to Flash native apps is point 6, in which he argues against cross-platform development systems.
I think this is a red herring, though. Apple doesn't reject apps for not taking full advantage of the system. People can and will make Objective C apps that don't use GameCenter (or iAds, or the accelerometer/camera). If Apple was most concerned about GameCenter adoption, why not leave out the 3.3.1 change and instead mandate that all games must use GameCenter? That would guarantee Adobe would include it in Flash and then nobody -- not even the native developers -- could produce a "substandard" app.
But even if Adobe made a version of Flash that could target every aspect of the iPhone platform perfectly (maybe by open-sourcing the compiler and allowing Objective C extensions), Apple would find a way to reject it. It's not just a matter of making it easy to port things to the iPhone... it's also a matter of making it hard to port away from the iPhone. Apple currently has the majority platform, and they want to make the choice between platforms expensive, so that developers choose the iPhone first and -- as much as possible -- exclusively.
All his reasons are fairly obvious for the technical user and have been discussed in the blogosphere, but I think hearing it from the horse's mouth and at a very high level will benefit non technical users.
I think you overestimate the degree to which non-technical users await, with bated breath, a pronouncement about programming languages from high officials at Apple.
I love non-technical users. I was trying to explain to one of them yesterday that they can still use my website even after their school wipes their hard disk, and separately, even though the email address (login) is their home email address and not their school email address. She was quite convinced that I didn't understand the situation, because she was going to lose "all of her memory except the Hotmail at Google", like the tech guy told her.
I had precisely the same discussion yesterday with a new client. She didn't understand how our web application did not have a larger list of system requirements. She started repeating the situation in a louder, slower manner to help me understand.
Here's all the insight I can muster: non-technical people do not have an accurate mental model of how the computer works. They have a semi-accurate mental model of the steps they take to do routine things on the computer: I click this thing, then I type in this thing, then I click this thing, then the magic happens. Deviations from that routine will almost certainly cause task failure.
Example: you know how Macs have the menu bar at the top of the screen rather than at the top of the window? I have at least one customer whose mental model is that the screen ends where stuff stops getting painted to it, and then that grey gunk above it is scary computer stuff. It took a day to talk her through finding the Help menu -- "It says Help, at the top right of your screen" because she was looking under the grey gunk bar and literally did not notice it until I returned the screenshot she had taken with a big red arrow on it.
I thought it would help reduce the cynicism and help programmers make a better mental model of their users (because that's what we've got all wrong at the moment).
All the programs and data in your computer is in files and folders, even windows is just a bunch of files. You understand this and everything else is far easier to understand.
Met a guy at a party last month who said I could make lots of money on the internet! Apparently, if you're good at websites you can put ads on them and make a ton of cash. Who knew!
"We also know first hand that Flash is the number one reason Macs crash."
I can attest to the veracity of this statement. My MBP's processor spikes whenever I get on YouTube or any other Flash-based site. I've had a couple of major crashes on both Firefox and Chrome while using Flash. I've also heard other users express similar views with these and other browsers on Ubuntu.
The only kernel panics I've had on my Macs in the past year have happened while loading or watching Flash videos in Safari. Of course this indicates that both Adobe and Apple have critical bugs to fix, since there's no way Flash should be able to take out the OS no matter how buggy it is.
One of the big reasons I use Safari on Snow Leopard is that when Flash crashes, it doesn't take the whole browser with it.
That said, I've started using ClickToFlash, which prevents Flash from loading except when I tell it to, and my browsing experience has become far faster and more reliable than it ever was before.
I was expecting Microsoft to try to scapegoat flash for all of there crashing issues a while back. I would have made sense, and it would have been half true. The bottom line is flash is crappy software, it makes the OS look bad, but the OS gets all the blame.
The most brilliant part of Apple's marketing here is leveraging infighting and unintended consequences in the free software community. They don't care about open standards - that they would is, on the face of it, absurd. What they do know is that they will evoke much support for championing JS/HTML/CSS as an open plaform, and that it poses no actual threat to their App Store control. This is because of two reasons:
First that there is so much infighting between Mozilla, Microsoft, and Google on the PC browser side, that it will never be possible to deliver video in one way that is accessible to all users, and Flash will remain required on PCs. (Ironically, Mozilla is Adobe's best friend here, as lack of H.264 support for HTML5 in Firefox for the indefinite future - even if your hardware supports it! - ensures that we will never be able to encode once and run on any PC - unless we use Flash!) So HTML5 will never never be a real final solution, because of the PC environment, which makes it safe for Apple to support and champion.
The second reason why HTML is no threat to Apple is that it just is too hard to write real applications in JS and current (or forseeable) browser environments. I get flamed for this a lot, but I think most of the people who flame me have never tried to build and maintain large cross-browser compatible codebases in JS involving a changing team of developers over a long time. People who disagree here always point to nebulous future hopes or technolgies that don't work very well yet, and I think underestimate the difficultly and time this will take, let alone for adoption. And they also, in my opinion, underestimate the problems inherent in Javascript, the language itself, independent of the well known pathologies with browser DOMs.
And, the reason why this marketing strategy is so brilliant is precisely because posts like this one will be viewed as conspiracy theory! This is because people have so many emotional ties to their hopes and their own preferred ways of doing things.
> They don't care about open standards - that they would is, on the face of it, absurd.
They really do care about open standards. There is no reason why WebKit and several other projects have to be run in the open, after they stopped releasing several parts of Darwin, but they are.
At worst, this is because there is more than one person at Apple, and the people who want things to be proprietary don't care about this and have left it to the people who like open standards to deal with.
WebCore and JavaScriptCore are based off KHTML and KJS, but WebKit (which unites the two into a usable API) is Apple code and doesn't have to be open source (it was closed source until 2005).
It doesn't have to be run in the open; they have to release the source whenever they release a new Safari update, but they don't have to do project administration in public or let anyone else access their SVN. This is how several of the Darwin opensource projects work, notably Apple's gcc branch and the kernel.
I keep thinking Apple support for HTML5 is precisely because it has a long way to go, and does not address big media concerns. The gap is filled by the app store which they control, and wont open.
I wonder why Jobs felt he had to explain this situation, which (in my opinion) was fairly clear to anyone involved in technology. Perhaps the criticism was getting to him?
Also, I found this to be a brilliant snipe:
> If developers need to rewrite their Flash websites, why not use modern technologies like HTML5, CSS and JavaScript?
It's called control over the experience something that really shouldn't be a surprise for Jobs.
There are literally millions of experiences out there that can't be done with HTML5, CSS and JavaScript. if you want proof, go to theFWA.com and have a look around.
For every bad flash solution out there, there are some quite fantastic ones.
Flash gives you control of the experience in ways that by any metrics outcompete HTML5.
This might not mean something for most people here at HN, but for quite a lot of people (the same ones that Jobs is catering for with his iPad) it's given them many great experiences. Experiences that are not going to be matched even by the best HTML5 developers out there for quite some time to come.
You mean the Flash app linked to on the front page of theFWA that took 30 seconds to load and almost crashed Chrome on a late model Macbook Pro 17? Is that the kind of mobile experience you're after?
There are literally millions of experiences out there that can't be done with HTML5, CSS and JavaScript. if you want proof, go to theFWA.com and have a look around.
I don't disagree with your general premise about flash offering more robust design control, and it's possible I didn't look around enough on the site, but I didn't see anything on thefwa.com that you couldn't handle with a HTML5, CSS, and JS (heck, I didn't even see much that you couldn't do without HTML5). Was there anything specific about that site you were thinking requires flash?
EDIT: Ah... I think I get your point, I assume you weren't referring to thefwa.com itself, but the sites linked from there.
I would argue that dropping in gordon for a site like theFWA.com would easily solve the problem... seeing how everything they're donig with flash can easily be done with plain javascript
Adobe claims that we are a closed system, and that Flash is open, but in fact the opposite is true. Let me explain.
I think he states the major reason, Adobe seeks to characterise Apple as against the freedom of the developer. This campaign adds weight to developers that are jilted by the iStore. They can find a home in Adobe. I think that this should clear up why Flash would not be the friend of the iPhone developer.
That quote comes in the context of rollovers. Modifying a flash app so that it doesn't require rollovers is an order of magnitude less difficult than rewriting the entire program from scratch.
Analogy: "Instead of replacing the car's steering wheel with a joystick, why not just rebuild it as a boat?"
There are lots of good points in the article, but I found that line flippant and misleading.
I get what you mean, but it shouldn't be that surprising. He started out technical, and most people with that mindset don't just stop thinking that way.
Conclusion: being a great CEO in a tech company requires to have a solid tech background. Eric Schmidt wrote flex (yacc friend), Adobe founders were kick ass phds, Bill Gates comes to mind too.
I'm guessing a company that controls information as obsessively as Apple doesn't just "jot down" its official response to a major, ongoing controversy.
>We know from painful experience that letting a third party layer of software come between the platform and the developer ultimately results in sub-standard apps and hinders the enhancement and progress of the platform.
It sounds like Apple doesn't trust the market. If apps compiled from a Flash development framework target the lowest common denominator and don't take advantage of iP*d innovations, they will lose to apps that do.
A true open market is far more messy and painful, at least in the short term, then Apple seems to want for its app store apps. As the device and app experiences are tightly coupled in the user's mind, every high-quality app is an asset to Apple and every low-quality app is a liability.
Not saying that I think it's the best way to go, but it's worth understanding the intention.
I like using app store apps the same way I've come to prefer console gaming over PC gaming. There's at least supposed to be a minimum quality bar that things have to pass, and I can experiment with things without worrying that they'll muck up my whole system. I've never had an iPad game require me to download and install beta-quality video card driver software in order to be playable, and for that, I'm thankful.
* Flash websites will have to be rewritten because it relies on 'rollovers', which is not part of Apple's 'revolutionary' touch interface: overblown for two reasons - a) few stuff will have to be rewritten (many videos show desktop Flash content working on mobile phones without a problem); and b) this is not a problem with Flash, as any JS/HTML-based interface that relies on rollovers has the same issues.
* Flash is 'closed and proprietary': Flash is as 'closed and proprietary' as many of the technologies he's trying to portray as open, if not less. H264 is not open. Objective+C++ is not open. Flash (SWF) has a fully published spec and anyone can write tools to create, or play, SWF content, without royalties, or without having to respond to any entity.
* Adobe only wants cross-platform apps, not good apps; generic subjective bullshit.
* Comparing video decompression versus hardware decompression battery gains and claiming that's an advantage over Flash: bullshit, since Flash uses video decompression in most cases.
* Saying Adobe postponed the release dates of Flash 10.1 for smartphones: bullshit. He's repeating a story that has already been corrected on a number of places (that Flash's release was pushed to 2nd half of 2010) as fact. That he's ignoring the fact that the story was misinterpreted by one source, and simply repeating the wrong news, is staggering. FP is still bound for a release in the 1st half of 2010.
* Saying "Adobe has recently added support for h264 playback": gee, I guess 2.5 years is "recently"? h264 support was added with FP 9.0.115, released in December 2007.
There's a lot more, but it tends to get tiring to constantly dispel the most obvious FUD written for the general public and full of misinformation. Mactards will believe whatever they want to believe, so hey... I'll just let the platform speak for itself. In 5 years, let's see how well Apple narrow mindedness business decisions work out for them.
There are multiple implementations and Apple can tweak how it runs. It is open enough for their purposes.
> Objective+C++ is not open.
He never said it was. He said Apple has several proprietary products but believed the web should be open.
> Flash (SWF) has a fully published spec and anyone can write tools to create, or play, SWF content, without royalties, or without having to respond to any entity.
Yet Adobe has the only full implementation of Flash and if someone else were to try and implement it, it would always be behind the official release (just like Moonlight). And Adobe got lawyer-happy when open source guys reversed the streaming protocol in Flash so nobody is eager to try this anyways.
> Saying Adobe postponed the release dates of Flash 10.1 for smartphones: bullshit.
The fact is Adobe has been boasting about Flash on smartphones since Android first came out and there have been no working devices out except Maemo which totally sucked in every way Jobs described.
There was spin in this article just like his DRM article but I wouldn't qualify anything as FUD.
> It is open enough for their purposes.
Ok great, so Flash is also "open enough": fully published specs, anyone is allowed to use the format (creating, playing) without royalties. SWF is more open than h264.
> He never said it was
True. But if he wants to make this conversation an open-versus-closed conversation, I fully expect to compare the platforms in general instead of just a tiny bit of it.
> There was spin in this article just like his DRM article but I wouldn't qualify anything as FUD.
Saying Adobe postponed the release date is a lie. Saying they said it would be released before is also a lie, since Adobe never announced a date for public release (it was available for devs as beta). Maybe it's not your definition of FUD, but it's still a lie. I don't expect the CEO of a company this big to be repeating this bullshit.
> Ok great, so Flash is also "open enough": fully published specs, anyone is allowed to use the format (creating, playing) without royalties. SWF is more open than h264.
But does anyone do this? I see no usable reimplementations of Flash Player, and the spec cannot possibly be complete.
H.264 is a real standard with a conformance suite and bit-accuracy, and no scripting languages or multiple versions with different bugs to reimplement.
Flash also includes VP6 and RTMPE, which are proprietary and must be reverse-engineered.
This is clearly not true, though. The only implementation is closed and fully controlled by Adobe. Maybe it is more open in spirit but that doesn't mean crap in practical terms. In fact Silverlight is more open (in practical terms) than flash since at least there is a usable open source project. There exists right now a full featured and heavily developed open source h.264 decoder. You can't say the same for Flash. It is either Adobe Flash or spend a bazillion dollars trying to re-implement open flash that doesn't exist yet Flash. That's a non-starter.
* "Adobe only wants cross-platform apps, not good apps; generic subjective bullshit." Alright, so point to a single Adobe Air app, for example, that looks & behaves natively.
* "Comparing video decompression versus hardware decompression battery gains and claiming that's an advantage over Flash: bullshit, since Flash uses video decompression in most cases." Did you read the article? "the video on almost all Flash websites currently requires an older generation decoder that is not implemented in mobile chips and must be run in software" But hey, easy to prove wrong. Point to one mobile phone with Flash (not flash lite) support that has decent battery life. ... Oh, wait... Well, one mobile phone that has Flash support period? No?
Whether Objective-C is open is a weird discussion.
The Objective-C compiler is fully GPL'd, so that's definitely open.
Objective-C, though, depends on a runtime, and Apple's runtime is proprietary. While the GNU project has their Objective-C runtime, it performs markedly worse, and misses a lot of features that are available in Apple's version, including blocks, GCD-style multithreading, garbage collection, fast iterators, and more.
Really? I consider .net a closed platform, as it is controlled by a single vendor, there is no patent guarantee (as Miguel has mentioned) and the alternative implementations (like GNash does with Flash) always lag behind.
(I happen to like .net, by the way, I just think it's closed.)
.NET is a closed platform — but Mono, which does implement parts of .NET in addition to their own changes, is an open platform. All of Mono's design and implementation is transparent.
* We're talking about mobile applications like games being good enough, not desktop applications using native components. Please don't stray the conversation. Otherwise it's as dumb as me coming and saying iTunes should not be running on Windows.
* "Comparing video decompression" - The article (wishfully) ignores the fact that 'older' (flv) video formats are transcoded on the fly to a format that can be played by hardware. This technology has been present in the player for a while - ever since Flash Lite. Requires more CPU, but it is not the same as full software playback. Also, asking for mobile phones with battery life benchmarks is bogus, since it's still not officially released, so no proof can exist. It's like saying the next OS X is inherently shit just because you haven't seen it yet.
* Well, I'm not sure what other part of Objective-C that's not "open" you would be talking about, exactly...
* "We're talking about mobile applications like games being good enough, not applications using native components." Except we're not. That whole section was discussing specifically 3.3.1 and why they aren't going to allow cross-compiled apps to run on the iPhone.
* "Comparing video decompression" Oh, that's not gonna have an effect on battery life, nooooo. And so what, you're saying Apple should have included something that isn't even released yet in the iPhone?
I won't blame you for the confusion, as the naming is mixed up, but "Flex SDK" is the command-line Flash (SWF) compiler. It's free and open source. It's used to create SWFs or SWCs. It's the same compiler the Flash IDE has built-in.
I create full, rich Flash content with the Flex SDK, and has been doing it for a while. I don't compile with the Flash IDE. Not because I can't or because it's paid, but because using Flex SDK is better.
Maybe im nitpicking, but mxmlc compiles apps written using the flex sdk to flash bytecode. The flex sdk is open source yes, but the flash runtime certainly is not.
"Fortunately, there are over 50,000 games and entertainment titles on the App Store, and many of them are free. There are more games and entertainment titles available for iPhone, iPod and iPad than for any other platform in the world."
Clearly nonsense.
I'm willing to bet that there are many, many more games and entertainment apps written in Flash alone than there are on the app store.
Flash has been around a bit longer than the iPhone has, so no surprise there. Flash has been around since PointCast Network, if anyone actually still remembers that.
Wait a minute, how is comparing H.264 hardware support for decoding (I think you meant decoding, not decompression) vs. software bullshit? Shipping versions of Flash still use software for video decoding. The only versions that use hardware are unreleased beta versions of 10.1.
Your post sounds like an Adobe apologist making stuff up.
Hardware decoding of video content has been available since FP version 10, released in October 2008.
I'm not making stuff up. I actually have my history of gripes with Adobe, as if I've been working with their software for more than a decade. If anything, I'm just enraged that this article is making stuff up, and I don't like people taking ideology and fundamentalism into a discussion that should be tech-based only.
Hardware acceleration for h.264 decoding was introduced in the Windows betas of Flash 10.1 in October 2009. It hasn't shipped yet, and probably won't for another 6 months.
Also, Flash does not use any of the multiple hardware decode APIs on linux yet due to it needing readback support which apparently they don't have yet.
'Welcome to the Flash Player preview release, code-named “Gala,” which introduces support for H.264 video hardware decoding on Mac OS X 10.6.3, the most recent release of Mac OS X Snow Leopard... The hardware acceleration functionality in the Gala preview release of Flash Player is expected to be included in an update following the release of Flash Player 10.1.'
I'm sure it will see the light of day. I'm using the release candidate on my media center PC right now because it is the first version of Flash that can play Hulu content full screen at 1080p resolution without dropping frames.
Running a beta of a Flash Plugin that kinda supports hardware decoding but still crashes occasionally on a Windows PC is much different than running a production quality release of a hardware accelerated Flash plugin on a mobile device.
Adobe still has closed projects like Flash Builder 4 and Flash CS5. But saying that, "Adobe’s Flash products are 100% proprietary" is false. FUD.
- "Second, there’s the “full web”."
h.264 is more modern than Flash's support for...? Oh, right. Flash has supported h.264 for over 2 years. FUD.
- "Third, there’s reliability, security and performance."
Pardon me while I go name a WAP "at&t wireless." Yes, Flash has security issues. Google mobile Safari 0day. FUD.
- "Fourth, there’s battery life."
As mentioned above, h.264 was adopted by Adobe years ago. FUD.
- "Fifth, there’s Touch."
I wrote a multi-touch Flash API for custom hardware shortly after Flash 9's release. Multi-touch works fine in Flash. Rewire some events and it's seamless. FUD.
- "Sixth, the most important reason."
Has Steve Jobs actually ever used Xcode? Beyond the native, crappy tools that Apple offers, some of us developers prefer to choose the tools we like to use when we write software. We like to use various languages based on what works best for us.
Apple is screwing over developers and they are using Microsoft level FUD to convince us its for our own good.
Regardless of how you feel about Apple or the Flash situation, you have to appreciate a CEO that can write and explain a company's motivations clearly. Would that we had more of that kind of frankness and less corporate-speak gobbledygook.
They seem to be arguing both sides. First we should have open standards. This is great because apps work cross platform. But then he goes on to say that cross platform libraries are bad because they stifle apple innovation. This is very reminiscent of MS's reasons for not having standard compliant software in Office, IE etc. You can't have it both ways. Proprietary innovation may be faster short term, but in the long term it only benefits Apple. History has shown that once a corp gets the lock in advantage, they stop innovating and slow the entire process down.
Sounds pretty reasonable, but common, if they needed to prohibit cross-compilers, did they really needed to do that just RIGHT BEFORE CS5 was going to ship? Any developer putting that much resources into something would be pissed. Adobe didn't make it secret, that they were devoloping an iPhone compiler, so Apple could've just told them right away it's a no-go.
And how do you know that Apple didn't tell Adobe well in advance? Adobe pushing to get onto the iPhone platform is every bit as much of a business decision as Apple trying to block them. Adobe may very well have been informed but decided that Jobs was just bluffing (not an unreasonable conclusion, there are very few CEOs with enough arrogance to do what Jobs has done on this particular issue).
did they really needed to do that just RIGHT BEFORE CS5 was
going to ship
I believe so. Otherwise they release iPhone OS 4, but customers get a flood of non-native, non-HIG aware apps coded by Flash developers and compiled for i* by the tool who knows nothing about new version of the OS.
That cannot be the great user experience.
I turn a steely eye to their marginalization of KHTML.
Otherwise, I was with it until the justification for the new clause. That justification just doesn't hold up. Developers are not going to sit around and wait for Adobe to implement a feature they need; they will do as they have done now, and implement it themselves as necessary. The justification is very lame imo.
As a developer who has many times decided not to implement a killer feature that works in a modern browser because I will have to come up with some graceful degradation or alternative implementation in IE6, I can say that it isn't hand waving.
Good, I prefer those features not exist. It's bad for the Web to depend on anything not thoroughly designed enough to gracefully degrade and be usable on all the browsers you don't happen to care about.
Rightly said. If we (can) budget out the development cycle a huge chunk goes into ulterior CSS hacks (JS is handled indirectly by jQuery/<x> fmwk) and an unnecessary manual test cycle dedicated to IE6. It's not without a reason Microsoft sent a bouquet to IE6's funeral!
I am a professional web developer. IE6 is one of the many platforms I have to support. Between the different versions of Firefox, Safari, IE, Chrome, and Opera its about 10 browsers that reasonably should be tested for a given site.
I guess what I was saying is give developers some credit. We've been working around a lot tougher issues than IE6 for a long time. Its a hurdle, not a brick wall.
I agree with you, and that's why we usually decide to support IE6.
But there's some things that just can't reasonably be done in IE6, and I think its OK to drop support in those cases instead of canning the feature entirely.
I'd also justify that by assuming that if they're still running IE6, they're probably not interested in your fancy app anyway. Obviously this isn't 100% true, but I'd bet its close.
There are very large organizations that have tied themselves to IE6-dependent code (or at least, code that their IT staff assumse is IE6-dependent) on their intranets, and they are not eager to commit the resources necessary to bring everything up to HTML5 compatibility.
And IE6 still represents about 20% of browsers on the Web (http://gs.statcounter.com/#browser_version-ww-monthly-200807...). So a business owner hiring someone to design a public-facing corporate Web site doesn’t want to hear the designer dismiss one out of every five potential customers.
I love this paragraph, check their hypocrisy, switch Adobe/Apple:
[Apple]'s products are 100% proprietary. They are only available from [Apple], and [Apple] has sole authority as to their future enhancement, pricing, etc. While [Apple]'s products are widely available, this does not mean they are open, since they are controlled entirely by [Apple] and available only from [Apple]. By almost any definition, [Apple] is a closed system.
Apple doesn't want to include Flash because Apple can't update Flash. Apple can't fix bugs in Flash, add new features to Flash, and so on. Apple has to rely on Adobe to fix these problems, and historically, on the Mac platform, Adobe has done a piss-poor job of it.
I'm not aware of any situation in which Apple has prevented Adobe from going about their day-to-day business with e.g. Photoshop or Illustrator, or has hindered their ability to fix bugs or add features to those products.
There is no hypocrisy here. At no point does Jobs say "proprietary software is always bad". Even the very next sentence is "Apple has many proprietary products too." This is not some dirty secret they're trying to cover up at all.
The point is that he is specifically taking about the web being open. Again: at no point does he say being proprietary or closed is inherently a bad thing. Only being so in the context of the web they consider bad. And Apple's actions in the context of the web support that.
There's a distinct irony that they're talking about openness, while at the same time explaining why they're preventing their own users from running software on their purchased devices.
If their arguments are correct, let Adobe port Flash and let users decide rather than telling them what they can and can't do with their own devices.
The article/memo says that Apple are waiting for Adobe to present a ported version of flash that works. It says that they promised early 09, late 09, early 2010, late this year...
I've met a lot of people who bought an Android phone because the iPhone 'doesn't support Flash', on the advice of friends or bloggers who complain endlessly about the issue, only to find that Android doesn't support Flash either.
When Flash is finally released and it turns out not to work on their pre-Droid devices, I wonder if they'll switch to the iPhone. I mean, either way, they won't have Flash to view MLB.com, but at least on the iPhone there are fantastic native apps to do things like that.
The web without Flash would be a better place, but the fact is that there isn't any viable open replacement. Yes, HTML5 can replace Flash for video. That's great and all, but it's probably the least interesting thing that Flash does.
Want to use Google Analytics? Sorry. Need Flash.
Want to see StreetView? Need Flash.
Hulu? Need Flash.
Aviary? Need Flash.
Games? Need Flash.
Apple's answer to everything but video is to buy it from the App Store. Really? Buying shit from the App Store is not the Internet I signed up for.
Aviary: neat app; there are image editors on the iPhone, but I'm not familiar enough with Aviary to find something comparable. Maybe there isn't anything.
Games: There are tons of games on the iPhone and iPad that don't need Flash. Command and Conquer is one of them, and Final Fantasy I and II are some of my favourites. There are also tons of independent games, like Implode, Tiki Totems, and MotionX Poker, as well as pseudo-indie games like Plants vs. Zombies and Peggle.
So, are you really so averse to paying someone for their time? Nothing comes for nothing, and if you're so broke that you can't shell out a few bucks for an app that makes your life easier, you probably can't afford an iPhone either.
That las paragraph strikes me as inconsiderate and a lot of guess-work. There's nothing wrong in wanting a website to work in a device who markets itself as something to browse "the full web".
As he said: "Buying shit from the App Store is not the Internet I signed up for.", but that's Apple's vision for internet. One they control, pre-approve and sell in their market.
You're basing a lot of this off of current implementations of these websites. Which one of these could not be done (at least a large portion of it) in HTML5? Each of these seems like there is HTML5 functionality to enable most of the functions and features of these websites.
I've jumped to the conclusions on the article and this puzzled me:
"The avalanche of media outlets offering their content for Apple’s mobile devices demonstrates that Flash is no longer necessary to watch video or consume any kind of web content."
Because media outlets are forced to make their videos available to the iPhone without being able to use Flash, this means that Flash is no longer needed?
Yes; for video-only uses, Flash is no longer needed. No doubt the avalanche is because a popular smartphone/tablet does not allow Flash, but Steve's point is that Flash is no longer needed because similar functionality can be performed in the more open HTML5.
Why the down votes? Of course a native iPhone app can stream DRM video, what does this have to do w/ Flash or HTML5? Flash is still needed for DRM video on the web. Hulu can create a native Windows, iPhone, Mac, etc. app but how does that cure streaming video in the browser?
Quote from Hulu:
"We're keeping a close watch on video support in HTML5. While as developers we are
always excited about cool new technology, we always analyze it from the
perspective of the needs of our customers: our end users, advertisers, and content
partners. To serve these customers well, we'd love to see further support in HTML5
for features including the following:
- Quality of service monitoring (e.g. detecting playback glitches)
I expect to have Flash available for my for droid soon, and for many of the reasons presented in this well articulated press release, i will not install it.
The thought of having to wait till a useless flash Ad render on my mobile browser, without any ability to block them is reason enough not to have it on my phone.
Unless Flash framework is completely revamped it will vanish to a place of old proprietary technologies, I am saddened by this, because flash offered a venue for application development to many creative people who are not programmers, and they made amazing stuff, but standards are important, and i suppose authoring tools for HTML5 will mature enough for non programmers in due time.
With all that said, The lock on the stream of revenue solely through the app store is to me, very hard to accept, developers should pressure apple to allow other venues, in the long run a single store is just not conducive to anything "open"
If Flash never existed and HTML5 was king, what would flashing, animated, noisy advertisements be created in? Content is not the fault of the platform.
What if, is not the question here. What is, on other hand, is more relevant, and at present the most obnoxious experience of web advertising is via flash.
"If Jobs and Apple are actually committed to creativity, freedom, and individuality, they should prove it by eliminating the restrictions that make creativity and freedom illegal."
We techies are guilty of always wanting everything. I see projects a lot of time, that won't move forward because the developers will not have the balls to stand up and draw the line between things the software will to, and the software will not do. It's just more comfortable to avoid confrontation. When I read this, I felt that Steve really knows what he wants, and in order to get it, he is willing to eliminate things that are less important (even if that means he has to hurt some of our feelings). We all can disagree with some of the points made, but in the end the iPhone is unique (and successful), because it has a clear definition of what it's not.
So this is all about Steve trying earnestly and selflessly to create the best platform possible, and not about trying to eliminate competition to his dominant platform by disallowing portable application environments?
I swear, I used to think the reality distortion field was a dumb metaphor...
You might be right about my vision being distorted, and I will think about that, but just so you understand me correctly: I did not say anything about selflessness. I have no doubt, that every move Apple makes is a business move; still, I don't think they only want to make money. They want to make money and deliver a great product. And the points Steve mentioned seemed reasonable to me...
I would say yes. From experiences with Adobe AIR and Java apps on Mac OS, Jobs knows that cross-platform toolkits are lowest denominator apps that don't work the way people expect them to. They're sometimes tolerable out of necessity, but they're never best-in-class, and they're often completely intolerable, unreliable, slow, and buggy.
As an example, I refuse to use Tweetdeck on the Mac, because even though it does exactly what I want from a Twitter client, it doesn't behave at all like a Mac app should. It uses too much memory, it's slow, it's crashy, and the keyboard shortcuts for moving around text fields, that I use in every other app, don't work. Instead I use Tweetie, which does almost none of what I want (except showing and making tweets), but behaves the way an app should, so it's less jarring to use.
The underlying situation here is that the runtime should be an open standard. Adobe's insistence that everyone should be dependent on them is absurd seeing as they have little market power to make such demands.
There is a huge need for a solid authoring tool for the canvas and Adobe is the right company to build such a tool.
Adobe has been whining for far too long about its inability to lock people into a proprietary runtime with poor performance.
Hardware and OS makers should have the ability to implement the runtime in ways that improve the experience. Apple is correct in insisting on such. This won't just be good for Apple, it'll be good for the entire industry.
I don't use Flash because Adobe forced me or insisted that I use it. Flash was the only solution to create highly interactive applications on the web for a long time and gained huge popularity because of this. The Flash plug-in was a failure on iPhone so they went out of their way to use the Flash environment to create native iPhone apps (i.e., no longer Flash).
He had me until #6 ("the most important reason"). Really, he could have just said "battery life" and I would have said "okay, that's a good point." But going on and on about cross-platform? Let the app store sort that out. I can guarantee you that the best flash is better than the worst native app, so the argument doesn't hold much water.
This kind of thing should be very actively encouraged.
Clearly there is spin in the post - but it is at least reasoned and tries to present a cohesive argument/point of view. More of that from every company please.
I think that even if you don't fully support or agree with the arguments (some of them have been debunked or much discussed after all) I think we can support this idea of company CEO's/management writing about their thought process in complicated or controversial decisions.
EDIT: Uh? I made the same point many others have.. is there any reason I've got the wrong idea?
Unfortunately, Adobe's track record of delivering solid Mac software dovetails with Apple's other stated issues. The fact that Adobe didn't step up with Creative Suite and Flash Player on the Mac lends credibility to Jobs' concern that Flash-for-native-apps would leave them captive to Adobe's schedule.
Reminds me, how long it took Apple to move Finder and iTunes to Cocoa? (Answers for those that don't keep score: 10.6 and not yet).
You have to remember, that Apple tried multiple rewrites of MacOS in 90' and OSX was yet another 'this time it will work out, I promise' project. If you were Adobe, would you bank your company on yet another promise? I know I wouldn't.
What promise do you mean? Cocoa has existed for 10 years now. It was never going to go away at any point.
Finder was kept in Carbon as a demonstration that large codebases ported from OS 9 would still be supported (it would take quite some time to rewrite them, after all).
iTunes's lack of a rewrite is probably due to having to run on Windows without having to immediately do the extra work needed to port Safari, but at this point it doesn't really have an excuse. At least it supports background scrolling, so it's still beating Word...
Apple had several failed attempts to establish new APIs/OSes in 90's (OpenDoc, Taligent). Why would anyone rational believe, that another project (OSX) is going to be success?
So during introduction of OSX, Apple had to include Carbon there, if they wanted to have ISV support for their new OS. Nobody was going to put significant effort to port entire codebase for new, unproved OS+frameworks+language put out by someone who has history of failures.
And do not forget, that Adobe CS is much larger project than Finder, iTunes and Quicktime together.
I love how he didn't sugarcoat Apple's core motivations:
Everyone wins – we sell more devices because we have the best apps, developers reach a wider and wider audience and customer base, and users are continually delighted by the best and broadest selection of apps on any platform.
Yeah there's a business interest. But it's also good for developers, and it's also good for end-users. And they're all interconnected.
Mere mortals may have been tempted to eschew the business aspect altogether, but good ol' Steve isn't afraid to tell it like it is...
Until there is an authoring tool as strong as Flash for Html + Javascript + css (with regard to animation and design, and the ability to integrate these animations and designs with code) - then flash is still the best game in town for games, animation, and more...
I'd like to hear everyone's thoughts on the answer to animation. Sprites are heavy, but are the only other option I can see. Even still, there is no authoring tool better than flash for animation on casual games or the web.
I don't care if iAnything supports flash or not. They can ignore Flash, Silverlight, any other third party component other than the pure HTML. Steve might be right and has all the moral rights not to support them.
But banning anything originally written in Flash, MonoTouch or any other rapid or not rapid external "to Object-C compiler platform" should be out of reach of Mr. Jobs' jurisdiction. As long as it runs on the platform, it's none of his business.
Actually, he owns the platform (the App Store) and therefore it is very much his business. If you have beef with his policy for his platform, then you can jailbreak — otherwise, feel free to dislike his policies all you want, but you cannot disregard the fact that he's well within his right to enforce them.
Honestly, it's like telling a convenience store owner that he has no business refusing to sell M&M's in his establishment and that it furthermore makes him a big, old, bad guy(TM).
Anyone can supply him M&M's as long as they are in predefined colors and size, he refuses some of them by their taste. He is selling all the rest without any concern over the quality of M&M's. He was an ignorant middle-man for all other facts except taste.
Now, he starts banning some specific M&M's based on the production methods of them. No M&M consumer can tell the difference, all taste more or less the same way. Production does neither determine quality nor nutrition facts of M&M's.
If 3.3.1 was just about Flash, I'd agree with Steve. But I think that clause solves a lot more than just Flash. For instance, iPhone apps seem to be horribly unreliable in my experience. And I think part of this has to do with Objective-C on the iPhone's lack of garbage collection. I think that there's a valid point that a system like MonoTouch (which supports garbage collection) would improve the quality of iPhone apps.
Managing memory in Objective C is pretty simple once you learn the rules - namely, if you retain it, release it. People who write bad code won't stop writing bad code just because there's GC available.
All that adding a GC would do is move the burden of releasing memory to the GC thread; what would result is that in low-memory conditions (or when the GC runs naturally), you'd have a sudden performance hit as the GC churns through the objects in memory looking for free-able objects. On the desktop, this might not be as much of a performance hit, but on a mobile platform I can see this being an issue.
Likewise, the GC can only do so much. If a developer creates an array and starts stuffing objects into it, but never releases/unreferences the array, those objects will likewise never be deallocated. The onus is still on the developer to write proper code, but now they've voluntarily given up control of memory to a separate process that may not understand the code as well as they do.
Besides all of that, I haven't found iPhone apps to be unreliable at all. The apps I use are generally polished and work fantastically, and crashes are extremely rare (and most of those are due to OOM conditions directly attributed to WebViews).
Very interesting. Nothing that would make me want to buy into his vision, but then I've never been drawn to Apple's platform. It's nice to see his reasoning and product philosophy laid out so clearly. I think Apple could have saved themselves a lot of grief with mobile developers if this had come out earlier; but better late than never.
I hope that if Adobe responds, they'll do so in a similarly direct and civil manner.
What’s wrong about this argument? Would all Flash content that uses rollover just work on mobile touch devices or would some kind of modification be necessary? I’m really curious about that.
The other problem with this argument is that Jobs is saying that it's just as easy to rewrite the flash app in javascript as it to remove the reliance on rollovers. Whereas for the vast majority of the flash apps, removing rollovers is much easier than a complete rewrite.
I think you misread Jobs. He thinks of this as a transition of the web from desktop to mobile. He thinks HTML+JS is the future and the gains of supporting “legacy” stuff like Flash are diminutive because it wouldn’t “just run”.
If they're based on rollovers only, no. But again -- the argument is bogus because it makes it sound as if EVERY Flash content relies on rollovers alone (they don't - click or dragging is the norm) and that it's a Flash only problem (it's not, as I know plenty of websites that rely on rollovers for menus and as such just don't work on a mobile, touch-based browser).
I think you're reading too much into this. Many flash experiences do rely on rollovers for discovery. Obviously, not all of them do. Some of them do, but have sort-of-workarounds that are not so great. I mean, check out some of the most popular uses of flash, and note that they DO use pointer rollovers (say high to the newest youtube player).
He's making a very reasonable point: Lots of web UI decisions have to be re-evaluated within the scope of touch-based interfaces. Since we're rewriting, why use flash when open standards exist?
I feel like this is the #1 argument that flash-defenders dance around. What justification does Flash, as it stands currently, have to exist on the web in the face of HTML5? What does it bring to the table that HTML5 isn't busy providing a more compatible and open implementation for?
Well they are listing every problem that factors into the decision. I think it is about control more than just the bottom line, Apple don't have control of design decisions that are made in Flash, its much more in their interests to push people towards the app store and to a lesser extent open web technologies where they get to set direction.
All these facts also apply for desktop/latop PC's.
Remove Flash and notice you wont miss it. More and more websites are using HTML there day's. Removing Flash also improves browser speed and stability (just as removing Adobe Reader does).
I did a lot of Flex developing, but I think it's time to move on.
I have an extension that blocks flash from any page I visit. I can turn it on by clicking the in the little box where the flash content was supposed to appear.
I'll put it this way, my finger sometimes gets so tired clicking to turn the flash content back on that I sometimes just disable the extension.
He said developers (at least those at Apple's mercy) are required to use proprietary APIs, because only those have complete feature coverage. He didn't say they would be permitted to use modern programming languages, which is what most of us (other than those at Adobe's mercy) were offended by.
In what sense? This only addressed 3.3.1 indirectly. It made it clear that Flash was a target, but said absolutely nothing about other targets. Assuming everyone is in the clear from this would be a mistake, I think.
Right, but I don't think you would have the same reactive 'developer' lock-in if 3.3.1 was contextualized toward cross-platform right from the beginning.
For example, EA who uses Lua for scripting wouldn't be worried as they likely were due to this two weeks late press release.
Agreed, but the same can't be said for all concerned parties. The Unity3d guys, for example, allow cross-platform development, and many quality games are already using it. I don't think development addresses their issue at all (in fact, it might have made it worse).
For me there is one single data point that shines brightly through all this noise: we are well into 2010 and nobody has Flash running well on a mobile device. Not even the mobile companies that want it.
That's because that's the misdirection. 3.3.1 is the clause that's important, but it's not about running Flash on mobile devices. It's about Apple being disintermediated and losing lock-in with their developers.
Apple's policy of banning Flash doesn't really bother me, but 3.3.1 really, really, does. Having Flash on the iPhone as a means of cross-developing apps is the most interesting way for Flash to be on the device, for me: it would mean a lot of creative and fun casual games would be more easily available for me to play.
Video, I think, is a red herring; it's easier for web sites to switch to a different, HTML5-compatible embedding scheme than it is to rewrite Flash games in Objective C.
I don't see steve jobs allowing us to submit HTML5 apps to the appstore? They must be written in his closed proprietary format with a goal of locking you in to using their tools (xcode). Just like adobe wants to lock you into their proprietary format (flash).
His arguments sound good though. And it would be great for adobe to add HTML5 as a target but i don't think the spec is quite ready. Does webkit on the iphone expose all the devices to implement say chatroulette?
Flash is a great tool to develop cross platform apps until the html5 spec can expose enough of the underlying os.
I think the arguments are ones we've all heard before.
As a Flash developer myself, it only makes me second guess myself more. Not because it presents a new point of view, but that non-technical people will now harp in on the issue, and I'll have to once more explain why I'm developing on a 'dying' platform.
The thing that scares me though, is that if I want to move on in my programming career, should I be learning <i>another</i> proprietary platform to develop iPhone apps? Seems like Android is the alternative to that, but I just can't get rid of my iPhone.
All Apple's stubbornness does is leave more doors open for Android et all to get more penetration. Each thing you refuse to do becomes a selling point for your competitors.
It doesn't matter if you do not like flash Steve, it is used on a significant portion of the internet. Either you work with them and find a way to make it work, or someone else will.
Still vaporware... I'm just saying, it would be nice if Adobe can show a reference implementation of Flash on a mobile device, and prove that it actually works well at a decent framerate without sucking battery life. This would at least give Adobe some ammunition to use against Apple, and may make Apple rethink the "no Flash" rule.
If there is a choice to be made it should be up to the user. If they want to use flash, let them, if not allow them to disable it.
If they want to watch tiny movies on a tiny screen for only five hours instead of 10 hours at a stretch, let them.
Realize that when you make hardware it will be used for purposes other than the ones you intend, that's the whole idea of the power of the computer, it is a universal device.
To try to artificially limit what users can and can not do is a short term strategy, and a long term loss.
Adobe needs a new CEO, they have done nothing since 2008 but bad products. I wrote them a letter asking them to implement hardware rendering/acceleration in 2007 when Silverlight came out and Director 3D was dying. Now HTML5 beats flash as it is equal almost and Unity 3D beats Director 3D. They have failed to innovate since Adobe bought it from Macromedia. I blame the guidance and focus on being the market leader rather than staying fresh.
I am very glad to know that Apple assure us once again that they want a standards compliant flash-less web. This will definitely make the web a better place. As I've said before flash is only and option if you use Windows, Linux or Mac OS X if you use a *BSD based system or any alternative system you're pretty much fscked.
P.S. I am very tankful to Apple for taking this fight (whatever their reasons maybe).
Still doesn't compute. It's in the responsibility of the developer to create quality software, no matter what tool he chooses. So far Apple hasn't denied apps that are low quality in general, or have they? So how does this make sense.
I know for certain there are bad native apps in the app store.
All of this has been discussed before, though. It's not news, even though it's straight from Steve Jobs.
Clearly, none of you people commenting here are considering the flash gaming market. Yes, I can see how most video sites can ditch their Flash players once almost everyone has html5-capable browsers, and yes videos is the largest use-case for Flash right now, but games is the second largest, involves a lot more money, and cannot be easily replicated without Flash.
Also, the people making games in Flash are usually graphics designers, not programmers, and they will have a harder time jumping to new technologies, even if they reach parity. You would have to have an environment that is as rich as Adobe CS and as capable when it comes to graphic and animation design, but for html5/javascript, before you will see a significant shift there.
Flash is much more entrenched than you think, and no matter how much Apple or Google wishes Flash didn't exist, they won't be able to dislodge it until there are good alternatives for all the major use-cases for Flash, not just video or animated websites.
Of course they can, and of course they will, if a better technology and design environment appears, but that shift is going to be much slower than that of good web programmers.
These reasons are true enough, but I still believe what has been clear to me ever since the argument started:
Apple won't allow Flash on iP* because it would defeat the purpose of App Store.
With Flash support, you could make responsive, native-looking online apps specifically for Apple devices without having to go through the approval process.
This is all very good, but my faith in it is shaken somewhat by the blatant lie "There are more games and entertainment titles available for iPhone, iPod and iPad than for any other platform in the world."
Yeah, HTML5 is more open. But flash has advanced the web much faster than html and javascript have. We should not forget that HTML5 draft started in 2004. HTML4 was complete in 1995, and XHTML was complete in 2000.
Say what you will about flash, but I don't think going the pure html5 route is the solution. Flash and Silverlight and other plug-ins serve as great ways to test out new approaches to the web and add additional functionality to the browser that isn't there already.
I think HTML5 and javascript is the way to go, but sometimes you can't wait 5-10 years for innovation to be supported in ALL browsers with a consistent api.
I think, somewhat ironically, that the final "most important" point #6, is by far his weakest. A platform is made stronger by having more tools to create for it. If a tool does not keep up with new platform features then it will suffer in the marketplace. There's nothing about having tool x, that prevents anyone from using tool y or Apple's own toolset.
I do however think that Apple is in the right to require the tools to compile to native code using native UI widgets etc, and not be a blackbox runtime. The goal of ensuring consistent user experience is valid and worthy.
I am aware of the fact that today a good development stack is worth gold, and many tool stacks are using some sort of virtualisation to allow for easy portability...
I find it very refreshing to see Steve/Apple putting very strong arguments against wastefull portability layers...
One interpretation of this all I think is: mobile needs native apps, because they are less "resource" hungry.
Unfortunately, Mr. Jobs the average person does not care about why you don't offer it. They only care and know that what they are used to getting can not be had on your platform, but can & will be on other platforms.
This fact and doubled that you remain exclusive to AT&T may turn your juggernaut around and repeat history.
From a happy and estatic iPhone owner/developer; best thing on market hands down!
"For example, Apple began with a small open source project and created WebKit, a complete open-source HTML5 rendering engine that is the heart of the Safari web browser used in all our products."
He only failed to mention KHTML by name. He most certainly did not claim that they made WebKit on their own.
It is a very good example of putting spin on history. They don't lie, they just put the "right" emphasis. Depending on how much you know, you understand the text differently.
But failing to mention KHTML by name is an act, indeed.
Reminds me another masterpiece by Microsoft: "Java is a programming language proposed by the JavaSoft company", or something like that.
Whatever anyone's thoughts on this debate I think it's excellent that we're actually having this debate in public. This is the way the industry should work: open debate amongst industry leaders out in the open. In contrast to the behind the scenes skulduggery that seems to be the norm in business these days.
Wait. Isn't Final Cut Studio still written in Carbon? Talk about throwing stones in a glass house! What a strange statement to make given his obvious knowledge of the difficulty of porting such a large suite of applications.
What a brilliant explanation! It summarizes all the main points that different bloggers have come up with against flash. And if the death of flash means a boost for HTML5 then I am all for it.
Unfortunately for Apple, the average user doesn't care about any of this. The average user just want to go to a website and have it work. IJW technology: It Just Works.
What's really surprising here is such a long post with Job's name on it, in place of the one- or two-word emails we're used to. Is Steve going to start blogging now?
Let's assume for the sake of argument that all of Steve's points are a smokescreen for what ultimately boils down to "Apple is in business to make money for Apple and Apple's shareholders, not to appease peers, developers, or even customers beyond the extent that they continue to give us their money, and anyone who stands in the way of what Apple believes to be its clearest path to profits can get bent." Why is that not okay?
I feel as if there's this weird assumption that Apple must make other considerations, that as the dominant player (for now) they have some obligation to make room for everyone at the table, lest they become evil, unsuccessful, or both. Why? What evidence do you have that Apple's exclusive culture and persnickety habits will result in anything but more money for shareholders, more opportunity for developers to put software in more pockets, and more sophisticated but accessible products being made available to more users?
I wrote far too much on the first pass, so I just excised a bunch, and here's the salient point: Apple didn't luck into this. They're not a clumsy giant who happened to end up with a market advantage despite themselves. They're in the position that they are because of the decisions they've made; they've deliberately created this reality for themselves. Apple doesn't wield authority over developers because tens of thousands of you decided to do Apple a solid and start writing iPhone OS Apps out of the goodness of your hearts. They managed, against really unfavorable odds and heavy competition from longstanding incumbents, to create what is arguably the best and indisputably the most successful device of its kind. Because they made decisions which resulted in that reality, should you choose to play by Apple's rules you have access to a large installed base of users—many of whom have indicated with their credit cards that they're willing to pay premium prices for premium products. I mean, how many normal humans do you know who would consider buying ANY software for a cell phone before the iPhone? I have many friends and acquaintances who have either told me outright or given the impression that they don't understand why one should need to purchase any software for their DESKTOPS beyond what comes bundled, yet each time I see them they have new paid apps installed on their iPhones. But, if you want access to them, you must play by Apple's rules. In the event that Apple adjusts 3.3.1 to state that developers must submit video evidence that they wore tricorn hats for the entire development process, if you want access to the users, you must play by Apple's rules.
You're free to determine that Apple's decisions of late are unsatisfactory to you, and that the potential business isn't worth the development time/unsavory feelings/brain cycles/whatever to play in Apple's game. But to imply that they have some obligation—moral or otherwise—to consider any interests but their own isn't just inaccurate, it's lunacy. Apple doing exactly whatever the hell they wanted is what created the marketplace to begin with.
An aside: I'm not super pleased that I have to post under this inane pseudonym, but I work for Apple Retail in the Family Room (One to One, Genius Bar, etc) and I'm prohibited from discussing Apple on the web. Yet another compromise I've made with Apple, but (for me) making decent money and having the opportunity to work with people I don't hate while I finish school is worth attempting to resist the temptation to be sucked into silly arguments on the internet...most of the time. In an attempt to preclude fanboy assumptions/accusations I will note one compromise I haven't made: I use an Android phone because I've had horrible experiences with AT&T in the past, and I refuse to pay them one red cent, though I still hold out hope that on one of these hardware revisions the iPhone will be made available on other carrier(s).
> But to imply that they have some obligation—moral or otherwise—to consider any interests but their own isn't just inaccurate, it's lunacy.
You just say that it's perfectly normal for Apple to behave like a psychopath. Now, replace "they" (Apple) by "any for-profit company in the world". And watch the world burn.
It's a fascinating combination of good points and utter rubbish, but mostly rubbish. The twisted hypocritical logic here is almost entirely devoid of the introspection I would expect from Jobs. More important, if that Jobs felt the need to release this kind of statement to defend this absolutely absurd position. Yet none of what he says makes the rational claims that Gruber has already made, i.e. Apple wants to limit the platform to exclusives, by preventing cross platform development environments, it makes it harder to not make exclusives for the dominant platform. It's a calculated move designed to keep people developing for the i* devices and ignore the Android devices. Instead we get the kinds of BS nonsense that we've seen people speculate on, but would indicate a kind of madness on the part of Apple.
For example:
Adobe’s Flash products are 100% proprietary. They are only available from Adobe, and Adobe has sole authority as to their future enhancement, pricing, etc. While Adobe’s Flash products are widely available, this does not mean they are open, since they are controlled entirely by Adobe and available only from Adobe. By almost any definition, Flash is a closed system.
This is fantastic news, direct from Steve Jobs himself, Apple is going to go completely open on their products! This means that I can go and buy a copy of iWorks from Oracle now, and get a Mac Clone from Dell and an HTC built iPhone! I wonder how long it'll be until OSX is available from Canonical? so excited, so excited
Other that this, the point is almost entirely incorrect. There are plenty of authoring and playback tools for flash not made by Adobe. It isn't exactly an "open standard", yes, but Adobe is not the sole source of flash stuff these days. Even a cursory search on google for flash creation tools brings back a bunch. Mr. Jobs, please follow this link http://lmgtfy.com/?q=flash+authoring+tools
Other points are better made, but some are still bad:
1) HTML5 is better
2) Webkit is neato
3) Apple never said they support the full web anyways
4) Most places have HTML5 video support nowadays, except for Hulu
5) Who needs flash for games anyway, i* devices have tons.
6) Flash = bad security
7) Flash = poor performance
8) Flash = eats batteries
9) Flash doesn't work great with touch
10) Third party content development tools suck! The people who use them suck! The crap that comes out of them sucks too! Anybody who think differently was born wrong!
11) Adobe should be making HTML5 authoring tools anyways.
What Apple should do is put Flash through the exact same approval process it puts other apps through. Make Adobe work for it, but don't just simply cut it off like a tantrum throwing child that can't share. If these are the problems with Flash, kick it back with comments and force Adobe to fix brokeness to get approval through the store. Adobe being open or not being open is a garbage point as I noted above. Apple doesn't require any of the apps in the app store to be open standards compliant. The mechanism for dealing with all of these problems is already there and it's simple -- the app store. There's no need to "work with Adobe", just reject it from the store the same way lots of apps are rejected after the screening process. Lots of apps eat batteries. Is that a new rejection criteria, "apps may not consume battery life."? If I want to run down my device in 5 hours instead of 10, that seems to be my problem.
1) HTML5 is still very very immature technology with almost no good toolchain support for authoring HTML apps. It's still a pile of languages glued together in a browser that sometimes makes something useful or interesting, but also really really processor intensive and slow. We all ooh and aah, everytime we see a canvas demo that eats up 100% CPU time on a quad core system displaying something that we all saw in 1996! By just this simple test, points 7-8 are almost completely invalid. Does flash perform worse than native apps and use more battery? Sure! So does playing an intensive game!
2) Webkit is neato, but webkit is a different thing than Flash. It makes no sense to compare the two. It's like comparing webkit and iworks for the iPad. By this logic, Apple should also not sell iWorks for those devices but just point everybody to Google Docs.
3) That's okay, that's why there are 3rd parties who can build stuff to support it for you.
4) Good point. This is definitely the wave of the future, Hulu needs to catch up. The point about hardware decoding is also fair I think.
5) i* devices do have tons of games true (Steve, I hope you now see the connection between entertainment software and platform sales, it's eluded you on the Mac for years). But those games are not flash games. I want to play flash games on my i* device. Period. I am the consumer, and I get to demand what I want to purchase. The only fair point I think is that most flash games don't work well via a touch interface. I have to agree with this. But there are also plenty of flash apps that are not games that would be amazing on an iPad. I want to use those. http://news.ycombinator.com/item?id=1295052 And no bit of HTML5 will replace these kinds of things for at least 12-24 months. Flash is not just about video and games.
The i* platforms also do not have the most pieces of entertainment software of any platform. Apple needs to simply stop exaggerating their claims to the point of absurdity.
6,7,8) see point about app store rejection above.
9) Agreed, though some flash apps might work well. Or apps could be written to work well with it exactly the same as making an HTML5 canvas app work well with the i* devices. This BS about mouseover events, and device features is also true of the web, yet that doesn't seem to be a reason to yank Safari mobile off of the devices.
10) This is the most noticeable pile of utter garbage in this entire thing. Somebody here did a survey of the top i* device apps and many of them were made with third party tools or contained elements that are now in violation of the new terms. Most of those apps were of superior quality. Just because the tool chain is made by somebody else does not mean the output will suck. Conversely, just because the toolchain is Apple approved, does not mean that all apps that use that toolcahin are going to be free of suckage. There are lots of shitty apps in the app store, lots, and I'd bet the vast majority of them were written with the approved toolchain.
11) I think we can all agree that HTML5 is immature as a technology. Adobe makes great authoring tools. Therefore, Adobe should make great HTML5 authoring tools. I think we can all agree on this. Flash is eventually going to go away. But in the meantime it's still a useful piece of technology. Just because quartz timing devices will all eventually go away doesn't mean that we should just stop using quartz timing devices. But even if Adobe started today, flash would be a predominant force on the web for 3-5 years.
Finally, this is all a giant pile of misdirection. Point 10 is the key one not flash.
We want the web to be open and free while we keep everything we do on our end closed and make tons of money. Oh, and don't let anyone else try to profit from their proprietary stuff either (unless we get a cut).
OK. Apparently, this was silly enough to warrant a down mod. So, let me ask this simple question:
Who here stopped using their desktop, to rely exclusively on an iPhone and an iPad? Who plans to? Why?
If Tablet PCs ever displace conventional ones, they won't have Treacherous Computing build in. Even my mom understands that depending entirely on a single company for much of her work and personal life is not a good deal.
It's good Jobs attempts to lay out his reasoning with facts but his facts are not all that correct.
Flash has been ported to mobile device. Windows Automotive, a version of WindowsCE, has flash. The Ford Sync system's front UI is completely flash and action script. I know because I work on the system.
I mostly agree with Jobs reasoning though. He makes very good points but it would be nice if he could get his facts right.
There are so many things wrong with Jobs explanation I don't know where to begin.
But two claims really pisses me off.
Battery Life
If I activate push notification it sure isn't nice to my battery. So I switch it off.
The same thing could be done with flash. Or even better, make flash an opt in. What's the problem.
Yes I know apple can claim that it will affect peoples impression of their products, but so do not having flash.
Touch
This really is a strawman of enormous proportions.
First. There are plenty of html pages out there that have roll-over. Plenty of menus that expand as you roll over them.
Second. Rollover states is not a problem for most flash websites that users use. In fact Apple could simply allow for rollovers to be activated on push down and click on release.
I love apples products, but they are simply in the wrong here. I love my iPad but not being able to see flash sites is really getting quite annoying.
I don't understand why you reacted so strongly to the roll over argument. Jobs didn't state roll overs as the raison d'etre for not including Flash, he simply used it as yet another example why the tools and, more importantly, thinking of the previous generation of web developers will not advance the platform forward. (As an example, I have been frustrated at some sites reliance of roll overs when using my iPhone).
The battery life argument is also stronger than you make it seem. When users turn on push notifications, the phone warns you of a reduced battery life (also, when enabling 3G) and is usually not toggled by the majority of the population. Flash is much more insidious; it will live in the browser and to the average user, will just be a part of normal browsing. Ditto for watching movies over flash.
The touch argument just isn't valid. What exactly is it that can't be done with flash just because it's a touch environment. Flash don't care about what the input is. It will listen for what you ask it to listen to.
Regarding battery life
And why couldn't the phone warn you off reduced battery life if you turn on flash?
You simply make it an opt in, the first time someone is running flash you overlay a warning saying that flash reduces battery life and you can turn that warning off.
Its simple interaction design, apple are masters at that.
None of the problems stated can't be solved or are not already there with regular html (rollovers on plenty of normal html sites)
These are all problems that could be solved, this isn't the argument. The argument here is solving them is not in everyone's best interest and is artificially keeping the platform in the past.
edit: why are you so angry? Apple are providing a beautiful product at a great price to the public. If it doesn't have a feature you want, buy another product that supports it. I never understood being angry at a company because of not putting effort into what you think they should be putting effort into.
"Flash was designed for PCs using mice, not for touch screens using fingers. For example, many Flash websites rely on “rollovers”, which pop up menus or other elements when the mouse arrow hovers over a specific spot. Apple’s revolutionary multi-touch interface doesn’t use a mouse, and there is no concept of a rollover."
Flash don't care. You are not forced to use rollovers in flash you can but you are not forced to do it.
Drag and drop is a historical mouse feature. You have that on the iPad too. So apples touch is not more revolutionary than it still uses these.
Every other mouse metaphor is in use besides rollover.
No — they don't have to use rollover. But the fact of the matter is, particularly with games but also to some extent websites (navigation), that many (perhaps the majority) do. Apple is speaking about the state of things as they are, not how they could possibly be if people so chose for them to be that way.
But that is true for so many HTML/js sites too and nothing absolutely nothing prohibits people from playing flash games with a touch interface. Most flash games are point and click
Making flash opt-in wouldn't solve the problem. The VAST majority of non-technical users will not know about or use that setting and will blame Apple for the poor battery life, not Adobe. It's a lot about saving face here. Apple wants to be in control of their own success, which becomes hard to do when you're allowing proprietary technologies to run on their hardware. Jobs points out his frustration with Adobe taking so long to adopt Cocoa - this is just him trying to avoid a similar situation on the iPhone.
Speak for yourself about the lack of flash on the iPad, I consider it a feature on mine. 90% of the videos I run across on the web on my iPad are youtube or vimeo videos and work perfectly.
You know, what draws most of the juice from the battery?
1) Radio,
2) Display.
These two facts made VAST majority of non-technical users aware, that the more they use their gadget, the shorter the battery life is.
Fortunately for them, Flash is pretty much interactive, it does not run in background and does not draw energy when the users thinks it is off. It is pretty easy for the users to comprehend that they were using flash and that's why the battery is empty.
I am speaking for myself, who else do you think I am speaking for?
Regarding making flash an opt in being a problem. It's a pure interaction problem. By no metrics is it harder than to turn push notification on and off or setting your time.
Simple solution: use the accelerometer to aim a mouse cursor for sites with rollovers. I don't suggest Apple implement that, but some app developer could, or maybe Adobe.
re: battery life, if you want a phone that is infinitely tweakable get a Nexus One. Apple devices are unapologetic about restricting you in ways that Apple thinks provide for a better user experience.
I don't buy this at all, although some of the reasoning is sound there's some jaw-dropping rubbish in this statement. esp. around 'openess'. This is one big red herring - Adobe pissed Jobs off, this is his revenge, end of story.
Wow, when did HN become all Apple fanboys, all the time? - I thought this was supposed to be a rational argument but it seems as long as Jobs is destroying Flash he's ok with everyone, even while he's locking down the rest of his platform.
> Wow, when did HN become all Apple fanboys, all the time?
Not all the time, but I think it's quite telling that a gushing 'I love this, with a slight reservation' gets 150 upvotes and you are sitting here at the maximum downvotes for simply stating your opinion.
Maybe most people on this thread are just blind. Blind to he "Treacherous Computing" which is applied to the iPhone and the iPad. Blind to the fact that these devices are fully fledged computers. Blind to the fact they wouldn't like to be locked out of their desktop PC in the same way.
Apple pulled out quite an American dream here: they made a sleek product, and were wildly successful. On top of that, they spurred a new gold rush (the App store lottery).
So, Apple looks great on the short term. On the long term, however, they make steps towards a society few would like. (Did you saw the mention to the "PC era" at the end? This is probably the most dangerous thing in the whole post. Yet it went unnoticed.)
The problem is, few think long term (it is dismissed as "idealism"). Most prefer short term (calling it "practicality"). HN is just no exception. (Opinions like "privacy is worthless" or "the BSD licence is more free than the GPL" are both short term-based, for instance.)
Apple already provides some amazing documentation and tutorials for making offline-capable mobile web apps, but they're still treated by Apple as the bastard cousins of "real" iPhone apps. They should buy or provide an officially-supported analogue to a tool like PhoneGap[1] or Titanium[2], allowing HTML5 apps to appear in the App Store, and perhaps provide an Apple-supported framework such as their mysterious PastryKit[3] or contribute to SproutCore[4], giving HTML5 developers better tools to conform to Apple's HIG.
HTML5 apps built on open standards should be on an equal footing with apps built in Cocoa Touch. That would be walking the walk.
[1] http://phonegap.com/
[2] http://www.appcelerator.com/
[3] http://daringfireball.net/2009/12/pastrykit
[4] http://blog.sproutcore.com/post/531215199/introducing-sprout...
EDIT: "officially-supported analogue"
EDIT: footnotes