Many, especially the zealous Apple supporters, would have you believe that Apple killed Flash. In reality, it's HTML5 that might be the ultimate demise of Flash, but it won't be this year and probably not next year either. A recent article at TechCrunch took an extremely myopic view of the whole thing and claimed that Apple "won" some imaginary war between itself and Adobe over the future of Flash. However, Adobe has already been ahead of the game as they're all about web development and not about tying everyone down to Flash. After all, they've got several tools already available to either export your Flash workflow to HTML5 or to create new web experiences directly in HTML5.
Knee-Jerk Sensationalism: Apple and Adobe Fight to the Death!
I asked Adobe about some of the claims in the TechCrunch article and about their path for the future and they simply pointed me to a couple documents which made me chuckle. They always tell me to contact them for comment, but then only had a "no comment" when I did. Anyway, a year ago, I already knew Adobe was on a migration path to HTML5.
I'm very anti-sensationalism in reporting and there are been some behind-the-scenes, knock down, drag out brawls between Mark and I over the wording of titles in the past because of it. I understand the need for catchy titles but I hate trying to pull readers in with something that sounds fake or over the top. Facts are facts and we just report the facts ma'am. Or rather, we do when we're reporting. When we're doing industry analysis, we use the facts as the basis for the analysis. However, some people fail to gather all the facts before spouting off some alleged analysis of what's happened or is happening and then they lure you in with things like, "How Apple Won The War Against Flash."
I'm not an Apple hater, I had a 1st generation iPhone, and currently have a brand new iPad. What I hate is the rhetoric that surrounds it all and is then propagated by seemingly tunnel-visioned journalists and publications solely to bolster their traffic numbers.
The HTML5 Landscape
First, some facts, thanks to LongTail Video and their State of HTML5 Video project.
Currently, 74% of the market can play a video in HTML5, the most notable exclusion is Internet Explorer 6-8. Even LongTail Video agrees that, "Eventually, mobile Flash support will disappear entirely." That link goes to an Adobe post talking about the increase in their own HTML5 contributions, from November of last year.
As far as the <video> tag is concerned in HTML5, there's some lack of compatibility still across browsers.
Preload and autoplay are simply ignored on mobile browsers. That's not a bad thing because it stops data consumption in the background or automatically which many mobile users still pay for outside of a standard monthly plan. So really, it's a consumer-facing move on the part of the browsers to not support that. However, there are far more technical issues that HTML5 can't yet address fully. For example, DRM, proper full screen support and control, lack of support for the text <track> tag for captions and subtitles, adaptive streaming including the new MPEG DASH, etc.
Clearly, at this point from a technical perspective, HTML5 is not ready for prime time, but it will be soon, I have no doubt. I also believe that Adobe will still be a major player as they have all the tools for the creation, deployment and support of HTML5 websites and video. Especially since Adobe Flash Media Server 4.5 can repackage Flash video and stream to iOS devices.
What Does Adobe Say?
In one of the two documents to which Adobe directed me, "Adobe roadmap for the Flash runtimes," I found this:
Adobe believes that the Flash runtimes are particularly and uniquely suited for two primary use cases: creating and deploying rich, expressive games with console-quality graphics and deploying premium video.
Given the maturity of the Flash platform and the fact that it can do things like DRM and adaptive streaming it will probably remain a major platform for premium video until HTML5 catches up.
One of the things that made me chuckle in the TechCrunch article was the author's complaining about the lack of 'full web access' which he then demonstrated by complaining about not all videos being available on the mobile web. However, what he fails to realize is that is most likely the result of the licensing agreements with places like Hulu and not in fact a Flash limitation. There are loads of premium video content that are available via the web on a PC but not even available via the Hulu desktop app, which would presumably be the same thing. Again, it's a limitation of the licensing more so than the delivery engine.
So Why Did Adobe Drop Flash Mobile Support?
Here's my best guess. First off, on tablets and smartphones, you generally have an app for wherever you are going to get your video content from, YouTube, Hulu, etc. They all have their own apps, so I imagine that Flash mobile browser implementation was pretty low to begin with. Why continue to pour resources into something that will probably never see widespread adoption when there are numerous stand alone apps that are the crux of the video viewing on those platforms? The same goes for the games, each one is its own app and I think that a platform like Unity or Unreal Engine will do far better than Flash anyway, because those engines were built specifically for games while Flash was built as an all around rich media platform. So again, why waste resources on continued support?
Additionally, since Apple is just under 30% of the overall market for smartphones, there's little justification to show that they had anything to do with the decision. The major news, as pointed out by the article, was that Adobe would not make Flash for Android 4.1 and, as you can see below, that currently is the major market share owner with 39%.
Why did Adobe abandon Flash for Android? Probably for the same reasons listed above. On top of that, with a new version will probably come new features that will fully integrate the full gamut of HTML5 <video> tag attributes and actions. Since Adobe already has Flash -> HTML5 paths available and the install base for Flash would continue to be low, it makes sense to slice that fat off and toss it in the fire.
Here's more from their road map:
Flash Player 11.1 is the last release of the Flash Player plug-in for mobile browsers. Adobe will not add support for new mobile device configurations (chipset, browser, OS version, etc.). Adobe will continue to provide critical bug fixes and security updates for existing device configurations through September 2012, and will also allow our source-code licensees to continue working on and releasing their own implementations.
Adobe continues to actively invest in enabling developers to create and deploy Flash based content as mobile (and desktop) applications via Adobe AIR.
For Android specifically it has this to say.
Beginning August 15, 2012, the Flash Player plug-in for mobile browsers on Android will only be available on the Google Play Store for devices certified to run Flash Player and which are preinstalled with Flash Player. Flash Player is not supported on Android 4.1 and users should uninstall Flash Player prior to upgrading to Android 4.1.
Why? Probably because there's no need to continue it. Meanwhile, for other platforms there are still plenty of reasons to continue for example:
Flash Player "Dolores"
Adobe is planning an additional Flash Player release in the second half of 2012. Code-named "Dolores", this release focuses on enabling features and functionality for the gaming market, as well as improvements for general Flash Player use cases.
Some of the features planned for this release include the following:
- ActionScript workers (enables concurrent ActionScript execution on separate threads)
- Support for advanced profiling
- LZMA compression support for ByteArray
- Support for hardware-accelerated video cards for Stage3D expanded to 2006
- Improved ActionScript performance when targeting Apple iOS
- Performance index API to inform about performance capabilities of current environment
- ActionScript 3 APIs to access the fast-memory opcodes (premium feature when used in conjunction with Stage3D APIs)
Plus, with Adobe AIR becoming their new dominant platform, for example on TV where they even say:
While we have historically licensed Flash Player for general web browsing on TVs , we do not recommend this approach given the difficultly of ensuring consistent and high-quality "full web" browsing on TV hardware.
So you can use it for video playback, but even Adobe acknowledges it's not all that great for full web browsing.
So Is Flash Really Dead? Did Apple Really "Win a War"?
No, and, uhhh, no. Flash is still, and will continue to be for a few years, a major platform for premium video delivery. Granted, it was never the best games platform and I think there are far better ones, but this isn't really about that. Third parties are still building flash-based experiences and it's still a leader in premium video content, because of things that HTML5 can't quite do yet. Will it be supplanted one day by HTML5? Perhaps. Is that day today? No. Did Apple win some war? Any tech savvy person should know the answer to that already. I trust you're all enlightened enough to deduce what they were really doing by now.
In my opinion, Apple simply pushed forward an immature agenda as HTML5 was not ready to be widely adopted, especially for premium video content, and it seems to me that it might have been more of a personal butt-hurt than anything. Adobe must have slighted them somehow in the past (perhaps by withholding an Adobe Flash cert in the first place?). Maybe that's why they refused to comment. HTML5 still hasn't been stabilized and certified itself, so forcing any industry to use it in order to play nice with your products really seems more like Apple wanting to be the control freaks they are instead of thinking of the best interests of the developers and consumers. With Flash, they had far less control I'm sure.
Besides, if Flash was dead, why would all the major online video providers have a fallback to Flash implementation? I mean, if it were truly dead, that would be totally unnecessary and just extra work with zero return which makes for bad business. Right?