What ? Adobe Flash for iPhone OS is an open platform ?
labs.teppefall.com
labs.teppefall.com
Which pretty much sums up the entire reason for 3.3.1. Did they serious expect that, going in with this offering, they'd get no pushback from Apple? It's been abundantly clear from day one that the iPhone store is a closed platform, subject to the business ideals of Apple (i.e. make Apple more money). Any sane iPhone developer knew this going in, and really doesn't care (or they'd never have started).
I suspect the correct way to view the iPhone store is not "A horribly closed environment compared to e.g. Windows/the Web", but "A largely open market compared to the PS3/Wii etc". Closed platforms have existed for eons without the world ending, and they'll continue to exist in the future. The real novelty with the iPhone is it sits in the middle - neither open nor closed. People are freaking out trying to shoehorn it into one camp or the other, when it's just not possible.
Apple isn't so much concerned with people porting iPhone apps to other platforms as they are apps targeting a lowest common denominator feature set, or targeting a meta-platform feature set instead of the iPhone SDK feature set.
I just semi-ranted about this on reddit, but my best example at the moment relates to the "XBOX Live" features coming for games in iPhone OS 4. What motivation would Adobe have to Adopt these features in CS5? How long would it take Adobe to adopt these features (if at all)? What if Adobe instead wanted to create their own "XBOX Live" system? Apple would be at the whim of the business interest of the meta-platform, much like they are with Flash on OSX.
After all, Flash devs are going to stay Flash devs, they're not going to port everything to Obj-C just to get one or two features on one platform they're deploying on, they're going to stick with what their meta-platform provides. Apple knows this and they don't want to be stuck the hoping meta-platform want's to move at the same pace they do.
Thats what Apple is talking about when they say these cross-platform apps suck, and if you look at the big ones out there for the desktop (eg, Eclipse, Firefox), they're right, those apps don't take advantage of the OSX features (CoreData, CoreAnimation, GrandCentral), they take advantage of the meta-platform features.
That is exactly the situation that Apple wants to head off. Apple wants iPhone and iPad applications to fully integrate with the infrastructure they are providing such as fast task switching, social gaming and iAd.
Very good argument; a perspective that needs to be taken into account.
I think the "correct" way to view the iPhone store is to be both.
This is the "No True Scotsman" fallacy at work. The word "sane" is doing too much work in that sentence. Certainly a lot of iPhone developers got into the work because they honestly loved the platform, and wanted to code to what they used themselves. I think arguably it's this population that is responsible for most of the real innovation (as opposed to yet another racing game, say).
And its these folks are the ones who have been burned most by 3.3.1. To write them off as "insane" is missing a big part of the picture, I think...
The author forgets that GCC / the GNU toolchain alone is not sufficient to package, sign, and distribute a .ipa - instead, the developer is bound to Xcode, signing certificates, and the whims of the App Store approval process.
By declaring ActionScript, C#, and Java (all of which have open compilers and virtual machines) "vendor lock-in", the author glosses over the fact that the iPhone OS toolchain is not an open one, in fact, arguably more restricted than all of the above. Make no mistake - I'm not arguing that any of the above are "open" platforms, or that open platforms necessarily make for a better product ecosystem. But I would like to offer a correction to the article's imbalance.
Anyway, I wasn't getting my hopes up. The dream of "write once, deploy anywhere" software is generally one that disappoints.
What Mike Chambers tries to say, is that he believes open platforms (android, etc.) will succeed over the closed ones.
Also he might mean that Flash and Air are tools that are open to other platforms: "The primary goal of Flash has always been to enable cross browser, platform and device development.".
Personally, i haven't looked at Flash for several years, but developing for Air sounds intriguing, targeting Desktop OS' (Windows, Mac OS X, Linux) and most mobile operating systems at once. I will at least take a look at Adobe Air.
Edit: The article says that all major companies Apple, Google, MS, Adobe try to lock in customers. One thing i find positive about Adobe nowadays is, that it seems to be the only Company to bring different platforms together. Except for HTML/JS, i don't know how i would write apps/games for the variety of platforms. xmlvm.org comes to my mind, too. A shame that even this project won't be allowed on the iPhone (i guess).
Edit: wording;
And if we consider media consumption, the standard on the web at the moment is Flash. So for a company who's trying to be a one-stop-shop for media consumption, this type of acquisition probably would make a lot of sense.
On the flip-side, I think the culture argument is probably the most compelling. Adobe products don't feel like Apple products. And for a company that makes its money by selling powerful software to creative people, their tools don't seem all that well designed.
In other words, you are almost order of magnitude off.
http://finapps.forbes.com/finapps/jsp/finance/compinfo/Finan...
or
http://www.google.com/finance?fstype=ii&q=nasdaq:aapl
(click on Balance Sheet tab, cannot link directly).
The figures I quoted were for 2009/12. There are now for 2010/03, with cash & equivalent 10 bil and current assets 32 bil.
Total assets are 57 bil, current liabilities 12, total liabilities 17. Equity is 39.3 bil, but equity is not the same as "cash in the bank".
Apple is well-managed company on the financial side, but let's not get ahead of ourselves and apply reality distortion field to business side too.
I'm glad for this (to which I'm currently addicted): http://lab.arc90.com/experiments/readability/