I continue to be amazed at the effort Adobe is going to to make their own platform irrelevant. They claim to want to focus on "premium copy-protected video" but do they really think that only targeting that to desktop systems when mobile adoption is skyrocketing is a good move?
Adobe seems to be pretty much clueless when it comes to Flash if you ask me. They never bothered to make a solid version of the plugin for Linux and OS X until the whole Flash-on-iOS debate started and Flash really didn't need more bad press. Then they tried get Flash on the millions of different hardware configurations on mobile, even though none of the existing Flash content was made for touch, none of the handsets had enough power to run most Flash content well, and all despite all the trouble they had supporting just the 3 major desktop OS-es. All of this effort even though Adobe doesn't make any money on Flash player at all, only on the authoring tools, which everyone could see is a dwindling market now that you can do so much on the web without any plugins.
I think their current strategy on the desktop is not so much 'content-protected video' (Flash did that 15 years ago already), but high-end gaming. They have some seriously impressive 3D stuff in the pipeline, which coupled with Air would finally make some good use of Flash as a platform-independent way to create games. But my guess is that it's all too little, too late, and that the market for cross-platform advanced 3D graphics development is pretty small. The really big guys that would make good use of it already have their native engines ported to those OS'es that matter, and the small developers probably don't have the budget to create games that really benefit from the technology.
It's weird to see how Adobe seems to screw up so often, because they really make some very good products. The whole 64-bit Adobe CS for OS X debacle comes to mind, it took them about 5 years too long to transition from Carbon to Cocoa, even though it was clear from the very first OS X version that Carbon was going to be replaced and would never have a 64-bit version.
>The whole 64-bit Adobe CS for OS X debacle comes to mind, it took them about 5 years too long to transition from Carbon to Cocoa, even though it was clear from the very first OS X version that Carbon was going to be replaced and would never have a 64-bit version.
I think Adobe really believed they could strong-arm Apple on that. I think they were spoiled by Microsoft's dedication to backwards compatibility at all costs[1], and thought that surely it would hurt Apple more than Adobe to let PS languish. I think they were so sure of this, in fact, that they made few arrangements for the contingency that Apple might just call their bluff. So when that happened, they had an uphill battle to fight.
it took them about 5 years too long to transition from Carbon to Cocoa, even though it was clear from the very first OS X version that Carbon was going to be replaced and would never have a 64-bit version.
That wasn't entirely their fault. Apple actually did promise a 64-bit Carbon and included it in developer previews, but yanked it shortly before Leopard shipped.
> it took them about 5 years too long to transition from Carbon to Cocoa, even though it was clear from the very first OS X version that Carbon was going to be replaced and would never have a 64-bit version
Yet, they still managed to do it before Apple got around to making iTunes use Cocoa ("oh, the irony", right?). The reality of the situtuation is that sometimes, even for flagship products, even for first-party products, sometimes portability is a critical concern, and Cocoa's reliance on Objective-C makes it very difficult to nigh-unto-impossible to not simply end up with tons of code duplication.
I am not certain what you mean by "C code runs on Cocoa": "Cocoa" describes a set of Objective-C libraries that form a framework developers can use to build applications for Mac OS X. While you can technically drop to the Objective-C runtime functions (the Objective-C equivalent of JNI) the result is often horrific.
When using Cocoa, you will find yourself constantly needing your data in the form of some random Objective-C classes. Carbon, on the other hand, stores all of its core data structures in C objects defined by a library called CoreFoundation.
This library, CoreFoundation, is something that you actually will find a copy of along with your Windows version of iTunes in the form of CoreFoundation.dll. Along with another few easily ported C libraries, this allowed iTunes to maintain the same networking and storage backends on Windows and Mac OS X.
I'd say they are more indifferent than clueless. I've read that adobe makes about 5-10% of its revenue from flash. They can build html5 tools (see edge.adobe.com) and still retain their core business, and hence profits. I wouldn't be surprised if they have been removing resources from flash for years now.
Look at what it took for them to get to embrace HTML tools, all of which are very recent. Whiney full page ads, screeching blog posts, deleted 'fuck apple' tweet tantrums. I agree that they're getting into some interesting stuff, but the path they took to get there was less than dignified. If they're indifferent now, they started out maybe not as clueless, but petulant beyond belief.
> Look at what it took for them to get to embrace HTML tools, all of which are very recent.
Recent? I don't use Adobe tools one way or another, but recent is hardly the word I'd choose. It's been, what, over a decade now, or near enough to make no matter[1]. Yes, this starts with Macromedia, but it isn't like Adobe isn't working with a group to build useful tools. Just because they support their own format doesn't mean they aren't supporting other formats.
This doesn't mean they only focused on standards compliance. They are a tools company, and they build tools for people to use.
Adobe makes money from selling the Flash authoring tool, but development of the Flash plugin is an expensive cost center. Macromedia and Adobe have spent a decade trying to monetize Flash's huge user population without much success. They can't charge users, but they had some limited success selling licenses/servers for Flash DRM video streaming. HTML5 video is unlikely to support DRM to content publishers' satisfaction any time soon.
>hey have some seriously impressive 3D stuff in the pipeline, which coupled with Air would finally make some good use of Flash as a platform-independent way to create games.
That is true, but on the other hand it looks to me that try to scare developers away by charging them for using it (That includes that they have to open their books, who likes to do that?). I still think they are pretty clueless. Also I wonder why I should use Flash for platform independed game development, steam for Linux seems to be around the corner.
Purveyors of "premium copy-protected video" use Flash on the desktop because the process of distributing, installing and supporting PC software is a giant pain in the ass, fraught with problems and security risks, to the point that users no longer do it.
On mobile, users love apps and distributing and supporting them is far, far easier.
Further, flash runs so poorly on mobile devices, and has for so long, that anyone serious about their "premium copy-protected video" jumped-ship for a dedicate app years ago.
So it doesn't matter what Adobe does. No-one is going back to mobile Flash at this point. Providers' app investments have been made, the same source content (h.264 video) is used in both places, and they get far more out of their own dedicated app.
Further, Adobe's money is made via content creation apps. Flash was a means to that end, but they don't need it. They can still make money off production tools, particularly while people still need to maintain a content pipeline that churns out resources that work on mobile and desktop.
This is my conclusion. There was a period when they were making a lot of noise about how they were making Flash "touch friendly", releasing Air across all the platforms, etc. As time went on Flash became tolerable (and useful!) but never really "good" and I suspect the real killer was that Adobe started to see dire feedback in their sales pipeline because people were prioritising iPad and Adobe could not offer a universal solution. Then I think, as others have said, Google decided Flash support was hurting Android instead of helping it (once you have critical mass, cross platform tools become a competitive disadvantage instead of the reverse) and probably told Adobe they were on their own and that was enough for them to decide to kill it, and killing it swiftly and making money from people forced to convert was better than killing it slowly.
I do think they really tried, for a while, but it just didn't work.
Adobe has been, and probably will always be, one of the worst software development companies in existence. It is only within the last 5 years or so that they have become even remotely competent. I do believe all decent programs they have ever published (think Photoshop) have been bought from another company and rebranded under the Adobe 'seal of death'.
Prior to that Adobe's business was based on PostScript, and they developed Illustrator to take advantage of PostScript's capabilities. Then they licensed Photoshop.
It initially seemed like a good thing for Android to support Flash but as time as gone on it's become apparent that the end result is a lot of sites produced a premium iOS experience and sub-par Flash based one for Android. Whether this flawed experience was due to limitations in the platform or lack of effort on behalf of the service I wouldn't like to say but it happened either way.
Hopefully now with Flash unequivocally end-of-life'd (on mobile at the very least) we might see Android devices being served the same experience as iOS devices.
Not the same - both were desktop, and running on close hardware.
Contrast this with flash - Lots of flash apps require vast amount of memory, download not fit for mobile, and most importantly clicks done with mouse. For all other cases it's fine.
Those problems have almost nothing to do with Flash and everything to do with complex desktop apps not running on mobile. It is possible to write light Flash and heavy DHTML (hello Gmail).
There was (perhaps still is?) a long-running battle as the BBC added extra protections around the iOS stream and users built various scripts to exploit this and download episodes. For example, I think that the BBC site uses the installed certificates to verify that the player is really an iPhone.
Obviously all of this is somewhat pointless, but to work on Android they'd either have to come up with an equivalent method or turn off one or more layers of "protection". That's probably difficult politically, if not technically.
> Obviously all of this is somewhat pointless, but to work on Android they'd either have to come up with an equivalent method or turn off one or more layers of "protection". That's probably difficult politically, if not technically.
I agree about the difficult this poses. Given the kind of wide range of Android devices, simple content protection might not really be possible. Verifying a device as Android doesn't really tell you much about the kind of hardware that it is running on, so for business types this kind of uncertainty makes them unwilling to try alternatives.
Big Content will not really care about Android until it is easy enough and/or profitable enough to service. Given that Android doesn't have an easy way to implement DRM schemes right now, it is unlikely that many companies will be rushing forward to work on Android until some easier solutions are available.
The interesting thing is that the BBC were happy to get iPlayer working on the iPhone originally. That didn't have any meaningful DRM, at least nothing that couldn't be done to a similar degree on Android.
I suspect that this is because iPhones are "cool" and someone high up just demanded that it should happen.
Weren't iPhones better locked down at that time? I'm thinking it was a calculated risk. Much like start-ups will do risky things when they first launch. iPlayer was effectively an internal start-up too.
Targeting iPhone is like supporting a cable company's set top box, so highly prized because it's straightforward; whereas targeting Android is like supporting a myriad of smart TVs.
Targeting iPhone is like supporting a cable company's set top box, so highly prized because it's straightforward; whereas targeting Android is like supporting a myriad of smart TVs.
That's certainly true for supportability. It's easy to test and confirm that it works on the 2 or 3 iphone models, less so for all the Android devices. That's potentially bad from a PR point of view.
For drm-esque security it makes little difference though. The 'attack' in this case was from people running a script and pretending to be an iPhone, not from people running unauthorised software on iPhones themselves.
The BBC (website) has a long, sad history of picking crappy media formats. They used Real Audio and Video long after that was clearly a bad idea, then when they had to switch away they picked Flash.
This is a huge issue for me that I recently faced when I bought a google nexus. Flash cannot be found in the jellybean market, and the only way to get flash working is to install it through apk and use a developemental firefox build. Even at that, the quality is blurry.
I know flash is dying but unfortunately a number of sites (mostly video sites) which I frequent regularly have not made a full (if any) conversion meaning there is a whole sector of the internet that is not accessible from my tablet.
As I bought it as a media consuming device, I expected less of a harsh cut from Adobe and Google.
Unless I'm missing something, that won't achieve anything as it's using the built-in WebKit which probably had Flash support removed (otherwise he could just have used the built-in browser).
I suppose that's why he's using development builds of Firefox - they have their own engine. Opera would work as well.
Honestly, I don't think I would have even noticed my tablet didn't have Flash. I don't have Flash on my Linux laptop (I use Chromium there) and haven't missed it --its nothing ideological: if I ever need Flash I'll install it right away. On the Windows netbooks we do have Flash but only because there we use Chrome instead of Chromium and it comes with Flash. I don't think I actually use it for anything other than playing video games with my son, which is not very often (his favorite games now are puzzle apps he plays on an iPod Touch, not that he plays those that much either).
(Off topic: now that I mentioned it all, it felt like we have a lot of hardware, but then I realized altogether I payed less than the cost of a high end laptop for it.)
I don't think I would have even noticed my tablet didn't have Flash.
I noticed that mine did. It was impossible to ignore. My tablet would frequently become unresponsive when I went to sites that had Flash content. Or sites with Flash content would be unusable because they relied on mouseovers in a way that doesn't work with touchscreen devices.
After a week I got around to figuring out how to disable Flash on the device. At which point my desire to give up on the Android tablet and get an iPad after all diminished considerably.
My totally unscientific and anecdotal input is that, for what I end up watching (5-10 videos/week), it's in the 50-75% range (for HTML5 videos being available).
Of course, that doesn't mean much and I always have a Windows virtual machine running for just these scenarios.
No youtube video is worth that amount of time and energy.
EDIT: That's probably why I can't convince myself to put Linux on my desktop as the primary OS. I just don't want to take the time to compile github scripts or scour the Internet for drivers/workarounds/fixes to make all of the basic computing functions I take for granted on Win7 work.
It's actually quite the opposite. Scouring the internet is what you have to do to find Windows software. With UNIX-like systems the software is more organised and more often than not packaged for you in respositories.
There is no need for Flash. There never was. With a UNIX-like system you can download any video and play it at your leisure. No slow starts, hiccups, timeouts, or whatever annoyances people tolerate with Flash.
The video is a file. You get the file and play it. Simple.
I would guess there are some video formats that Windows Media Player, not to mention other Windows video players, will still choke on. This does not happen with Mplayer (which is also available for Windows, but is truly a UNIX-style program).
And of course audio can be extracted from any video. Youtube is like a giant Napster. But the mp3's are hidden in flv/mp4/3gp/webm/etc, and Youtube has better lawyers.
My youtube downloader is 30 lines of sed and can use any tcp client (wget, curl, whatever -- there are so many). It takes me about 15 minutes to write and could probably be smaller. But there are plenty of more complex solutions, e.g., there's a nice one done in Lua called quvi.
It is trivial to set up a server that takes the youtube watch?v= url and returns the download url for the video file.
Obviously this probably makes some people uncomfortable. But this is how the web works. Anything that is uploaded can be downloaded. IMO, it's more respectable to try to be honest than lying to people that video can be "protected" from download by using some convoluted Flash scheme.
Flash is fading into obsolescence. Youtube is getting stronger every day. And the lesson from that is clear, at least to me.
And as for all thos people who love the concept of "streaming", youtube still uses progressive _download_.
If I want "streaming" I download to a file on a ramdisk and let Mplayer read from that file as the download progresses. With a fast connection, using ffmpeg to do the download will allow you to do transcoding on the fly if you need it. Mplayer gives flawless playback, every time.
The link JoshTriplett gave is a simple script, nothing that takes any time to make it usable after you download it (assuming Python is installed).
But, different strokes for different folks: and choice is usually never a bad thing. Of course, I learned long ago not to throw the baby out with the bathwater.
Personally, I have a few Windows virtual machines and an old, cheap Windows XP laptop around for those rare occasions as well as a Windows 7 dual boot install for .Net development.
However (again, anecdotally), I've not had a problem with using Linux day-to-day. Nothing has been difficult and I appreciate the power the CLI affords me.
I run a Linux VM in Windows on my desktop actually, but it's my gaming machine too. Last time I gave it a shot as a full install, I couldn't for the life of me get a brand new USB wifi adapter to work in Ubuntu. All the searching on the topic seemed to indicate that 64-bit users are screwed (which was me), and 32-bit users needed to use a driver emulator and dig through generic broadcom drivers to test.
I'm a Linux user, have no flash installed and use Youtube. It's not as good as i could be, but it works. Most content is displayed with HTML5, for the other stuff i use mplayer witch is a bit hacky, but works for me.
Flash still works on my HP TouchPad, but it's not that good. Video almost always lags the audio which makes them awful to watch and Video playback is the only think that matters with flash anymore and it fails at that so, I really see no purpose for it. I wish all the sites would just adopt native video.
Flash is always going to have a place on the desktop, but as a developer having to make a choice between Flash & HTML5 or just HTML5 is going to be pretty simple for a big chunk of purposes.
"The things Adobe is working on for desktops--advanced 3D gaming and premium copy-protected video--already exist in mobile through native apps and HTML5. As mobile and desktop platforms merge, Flash Player will get overtaken."
There's still things you can do in Flash that can't be achieved in HTML5, so until there's the ability to layer ad's in videos effectively for YouTube or strap DRM into the streams Flash will be a goto choice.
One of my friends who made a living as Flash consultant for about a decade - he has transitioned - last year, he had one Flash project - the rest were HTML5
Actually, while doing some research (for a MMOT class) I was told by several of the local web shops that their customers are finally "getting" it, and now worry about usability, Google positioning, etc.
Most of their customers are people that had their first page made by a friend, nephew, acquaintance, cheap freelancer - and now realize that nobody uses it and that people don't find them on Google.
Android users get screwed over with the iPlayer app. Not everyone wants to install Flash (and now no one can even if they want to), yet for some reason they feel need to serve the video in a different way to the perfectly fine iOS stream.
Adobe took the pragmatic approach.
The HTML standard is maturing to the point where Flash will become redundant technology sooner rather than later for regular media consumption.
Pulling the cord on Flash Mobile is the right thing to do. Seriously, which relevant company still builds anything for mobile on Flash?
well nice going google, first effectively kill or setback html5 video by torpedoing everyone with their anti h264 campaign and then get f*ed up the bum by them.
(here's a guy who 2 years ago said that flash is effectively dead long live html5 video)
pps. i won't really miss flash. but i do miss the time where every site had h264 vids up
