HTML5 Can’t Exist Without the Flash Platform
Let me start by saying I’m not anti-HTML5. I think it’s great. I think open standards are core to the DNA of the web and that’s part of the reason I’ve been encouraged that Adobe is moving more in that direction with AMF, RTMP, and some of our underlying protocols. But I also don’t think that means there is no room for companies like Adobe and Microsoft that want to see the web be better. I don’t think it’s an issue of corporate greed or vendor lock-in, but I understand the economics of how those end up being profitable. On the Adobe side I think it’s just that we have a lot of services and tools that we want to provide and we need a platform we can innovate quickly on. When the winds of technology shift we want to be able to make sure our customers can stay ahead. Video? RIAs? Real-time collaboration? Enterprise data management? All things our customers needed that we could build into the platform and make available to them. With the way open standards work it’s just not feasible to push those innovations through. So we’ve been very careful with the Flash Player. That’s how we provide our customers value. But it goes beyond features and into another problem with the web: a consistent experience. From browser to browser you’re going to get wildly different implementations of a “standard”. With Flash Player you can deploy your content and know it will run the same way for everyone that has Flash regardless of browser or operating system. There was an example today that made me chuckle and write this post. It’s not the best example; blurs and sparkly trails are not necessarily ideal uses of Flash, but in this context it was important. A group working on the NORAD Santa Tracker wanted to capture a specific experience; it added value to the overall presentation, and it was worth taking the time to create. Here’s the email:
Hey all-
My 20% at Google is the NORAD Santa Tracker, which gazillions of adults and kiddies use each year to track the progress of Santa across the world.
Last year, I used our Flash API to create the tracker map, primarily so that I could make a glittery comet trail of Santa’s past 7 stops. I tried it in JS with resized animated GIFs, and then decided that Flash would be much less painful – and was right. The code drew a line on a Sprite, blurred the line, then added some little sparkle movies and animated them.
You can see a screenshot with some trail here: http://www.wired.com/images_blogs/photos/uncategorized/2008/12/24/santa_nrad_edited.jpg
This year, I have ported the map to our JS v3 API, because we are doing aversion that utilizes the Earth plug-in (3d flying Santa!), and I wanted to make sure there was a version that didn’t require a plug-in. But, I really miss the glittery comet trail.
I wanted to get your advice on the best way to do this in JavaScript, given the following (flexible) requirements:
- Works in FF3, Chrome/Safari, IE 7+.
- Doesn’t require images (bandwidth is an issue).
- Doesn’t require a pluginI’ve contemplated canvas, but not sure if that supports foreground blurring,and don’t know if it would translate to VML with excanvas. It doesn’t seem very SVG-y. I could try animated GIFs again, but they don’t animate the same across browsers. I could also try window.setTimeout with some PNGs, but that will slow down the browser, require images, and won’t achieve the same blurred line effect.
Do any of you have a better idea? (Help prove to me and the kiddies that Open Web can make magic happen!
Thanks!
- PamElf (<- my elf code name)
Not an off the wall request by any means. It’s a pretty basic task and using open standards in this case is a great way to go. Especially considering that she’d like mobile support for this (later email) and HTML is still the best way to target every level of device. What surprised me were the responses:
My two cents:
SVG Web does support the blur filter so you could get this working cross browser if you force the flash renderer.
If you just forced the flash renderer for IE, that doesn’t solve the problem of Chrome/Safari who have not turned on their SVG Filter support.
SVG Web is a cool project to support SVG, but a requirement is that you have to use Flash. So even when you try to use open standards, you still have to fall back to Flash in some cases.
A hack would be to have a separate
<canvas>that draws the blur w/ an opacity and z it?
Seems like a decent solution, but requires some hacks. This is fine for a 20% project, but when you’re trying to be productive, are hacky solutions the best way to go?
How ’bouts IE 7+ with GCF? That’d let you try things like
<canvas>more directly.
A perfectly fine way to do it, but GCF is the Google Chrome Frame plug-in, which would mean another plugin and as Pamela says the message after that, it’s not ideal at all from an end-user case.
Now I understand that the major roadblock here is the IE 7 requirement. If it wasn’t for that you wouldn’t need the Flash fallback for SVG Web and you wouldn’t need to use something like the Chrome Frame plug-in to make it work. But that’s kind of the point of open standards. Before you can really take advantage of a technology you need to get a consensus from everyone. All of the browser vendors have different agendas and different priorities. Right now Microsoft is the hold-out, but what happens if something comes along and threatens another browser vendor’s goals? Then they’ll be the hold out. That’s part of the price you pay for standards; you have to get everyone on the same page before you can move forward.
In the end, that’s better for everyone. Once everyone gets moving in the same direction, that becomes the baseline, and the web is a better place. We’re seeing that start to happen with HTML5. And despite the “Flash is dead” rhetoric, HTML5 will be a good thing for the web. Things like playing basic video and doing animation/vector graphics are so simple and yet so integral to the web experience that they should have been done in an open standards way a long time ago. But if Flash didn’t exist those things wouldn’t have become as important as they did as quickly as they did. And even now, as the web tries to move forward, Flash is being used as the fallback so that the web can evolve even if not everyone in the group buys in.
To me, that says that Flash will always have value. We control the platform so we can keep innovating and pushing the web forward in our own way. Maybe that’s content protected video, maybe it’s P2P support, maybe it’s a binary transfer protocol, maybe it’s hardware acceleration, maybe it’s writing your own filters and blend modes, and maybe it’s real-time collaboration. Eventually, the open web will incorporate all of those things. But I still wish there was more cooperation between the companies who have technology core to the web and the open standards groups who create the future of the open web. If nothing else, the open standards organizations can see what features stick and which don’t so that you don’t waste valuable time trying to standardize something users don’t want.
In the end, everyone wins. Flash won’t die, Flash developers can still be on the cutting edge, and Adobe will continue to support designers and developers on both sides of the open web/Flash spectrum. We’re all part of the web ecosystem in our own ways. Just like any ecological system, if one side goes extinct, the other is going to suffer.
Posted in Flash, Rich Internet Applications
November 23rd, 2009 at 11:39 pm
Open web is future and Flash is the inspiration! Without the work of Adobe/Macromedia, today’s WHATWG/Open Web teams won’t even know what features they should integrate to HTML5. Youtube would have never become successful without Flash Video. That’s were these so called open web folks learnt how important videos are! Video is extreme example, talk about “rotation” of visual elements.
I have no reason to believe that open web/html5 can EVER stay ahead of Flash/Silverlight. I say this for a simple reason. I don’t see Adobe and Microsoft stop innovating there platforms. They are ahead of HTML5, they will stay ahead of HTML5 for next decade. Consider this:
According to WHATWG website HTML5 is still changing and will become final recommendation in year 2012. (towards the end of the world as we know it
Seriously a “final recommendation” in 2012? By then Microsoft is already out with IE9 and if HTML5 is still changing IE and other browsers won’t render things alike. It will still remain a mess. Chances are that MS will not incorporate an incomplete standard the way they ditched CSS3. So much for the cross browser HTML standard.
If you’re making cutting edge stuff today, you can do it now with combination of AJAX/Flash/Silverlight. What HTML5 brings in is nothing to be excited about. You will still be able to do the same thing with HTML5.
And all browsers of 2012 will still have to render todays page like they render it today.
I wrote a post on similar lines last month.
November 24th, 2009 at 12:21 am
Hi Ryan,
Great post! Way to take on a taboo topic head on
You make some interesting point. In the end, I’m betting it primarily comes down to how many talented-developer-hours (talent * # of Developers * number of hours) gets put into each platform.
Right now, I suspect that the flash platform has a heck of a lot more TDH’s than HTML 5 does, simply by virtue of how many years worth of TDH’s Adobe and Macromedia before it put into the platform, so it will certainly take a lot of time for HTML 5 to catch up.
On the other hand, HTML5′s openness has a lot allure too because of the possibility of how many different people can get something back from putting TDH’s into HTML5. IBM, Sony, John-charlie’s-donut-and-web-design shop could all through a few hours and making stuff better (theoretically). Google obviously has some 20% time (and probably some 80% time) going at the problem, and that’s a heck of a lot of TDH’s right there.
In the end, may the best platform win, but even if it turns out to be HTML5 10 years down the road, I suspect that Flash will simply because something different, and something better build on top of HTML5, throwing Adobe’s TDH’s on top of Google’s TDH’s for something all kinds of wonderful. (hey, engineers have dreams too!)
November 24th, 2009 at 9:06 am
Timothy, I have no doubt that the TDH gone into HTML5 is many, many time what has gone into Flash. HTML 4.01 was published as a W3C recommendation as of 1999 and XHTML 1.0 became a W3C as of 2000. So we’ve had around 10 years for HTML5 to evolve and it’s still not quite done, with a lot of TDH going into XHTML 2.0, which was eventually abandoned (or at least not to be used in browsers). Which is part of the problem, a lot of work in open standards is abandoned.
If you look at the various HTML5 mailing lists and chats, or just blogs summing up the progress of HTML5, there’s a lot of arguements going on between developers, with lots of controversy over various decisions that’s made. Open standards are great but because of all the people and organizations involved there’s a lot of disagreement, so progress is always so incredibly slow. Of course once decisions are made, the next step is to actually get all major browsers to agree on using the specs correctly and then many DHT goes into creating different implementations of the same features.
I’m sure there’s people inside Adobe wanting to push Flash in different directions, but because of the way things run inside a corporation decisions are made a lot more quickly and we see the innovation happen with new players.
Which I think furthers Ryan’s point, as I think the evolution of HTML would be even more chaotic without Flash to point to as an example of what is working. HTMl5 is going to take years to finalize and is still not to par to what Flash Player 10.1 can do. By the time new features found in the current version of Flash are added to say HTML6, Flash will be a lot further along to what it is now.
November 24th, 2009 at 11:23 am
Hi Ryan,
as I already said on Twitter, I think that this piece has an unfortunate slant to it. Mike (Chambers) rightly pointed out that I should comment here, so I’ll try to expand a little on what I meant:
The word “deceptive” with which I described it earlier, might be a bit harsh, but I think that the argument you make isn’t sound and that the headline is very misleading.
Most importantly, “html5 wouldn’t exist without Flash” simply doesn’t follow from “historically, Flash inspired many of the features now found in html5-capable browsers”. Let alone the much stronger “can’t exist”.
One might well argue that browser evolution would have been faster if it weren’t for the Flash Player: Customers might have demanded advanced features earlier and with more pressure if they didn’t have the Flash Player as a solution to their demands.
Note that I’m not making that argument, because it would stand on equally shaky grounds as I think yours does.
An argument of the same structure, and one I don’t think you’d be happy with, is the following: “Some of the features in Flash Player 10 couldn’t exist without Silverlight because Adobe only added them because of the increased competition”. I guess the exact analogy to the first part of your argument would be “Flash Player 10 can’t exist without Silverlight”. I think that sounds silly, don’t you?
The second part of your argument as I understand it is that the Flash platform, in addition to inspiring html5, helps making development of complex web apps feasible by providing a fallback solution for IE users. That argument, too, can be turned on its head, and I think in this case, the upside-down version makes much more sense: If it weren’t for the Flash Player the fallback solution it provides for IE, the browser landscape might look somewhat different, with much more people switching to more capable browsers simply because they want to use web apps for which no IE version exists because it wouldn’t be feasible to develop one. As it is now, this is a relatively recent trend. In an alternate universe without Flash, it might have started way earlier.
Please don’t get me wrong: I know that you’re not mindlessly bashing html and are generally pretty fair when it comes to comparing it to Flash. (Which I really appreciate, considering your job!) I just think that this piece is much less balanced than it poses to be and makes rather strong assertions on very shaky grounds.
November 24th, 2009 at 11:50 am
Great, great comments. Thanks everyone.
@Sameer, I agree, I don’t think HTML5 will ever overtake Flash in features. And I don’t think they should try, it won’t work with their model. As @Timothy and @Matthew both note, we’ve been working on the HTML spec for a long time. We’re not going to stop innovating and Flash still provides value.
@Till, thanks for commenting. While I agree the title of the post is a bit….embellished, this is the internet, so that’s the nature of the beast.
When you take out the implication in the title, I think the argument still stands. I do absolutely think Silverlight had a huge impact in how Flash has evolved. H.264 support is a great example. It wouldn’t have gotten into Flash as quickly as it did without Silverlight.
I also think you put too big a burden on the customers. Without Flash *maybe* customers would have started to demand more features in their browsers or switched to ones that support it, but I have a hard time seeing that. Do customers have any concept of HTML5?
I feel like the success of Flash hasn’t been driven by customers, but by developers who wanted to deploy better content. Some of that is possible with standards, but it takes a lot of work (a subtle point in the post). Flash lowered the barrier to create that kind of content and showed how the web could be. That, in my opinion, helped the standards community rally and start to innovate with HTML5.
=Ryan
ryan@adobe.com
November 24th, 2009 at 12:45 pm
Small point of disagreement : the problem of inconsistent experience, caused by wildly different implementations of open standards is not because they are open standards.
There are lots of examples where standards work – OpenGL is a good example, USB, Ethernet, or PostScript & PDF to pick two controlled by Adobe.
The problems with web standards are more down to the difference between the de facto standard (IE) and the specifications (although some of the blame does lie with specifications that were ambiguous in the first place).
Nor should we forget that while people use both technologies for RIA now, their original intent was very different – HTML, and even early CSS, were not intended for graphically ‘perfect’ layout – if anything, they were intended to make the same information accessible across the wide range of machines people were using at the time.
Flash and PDF started from the opposite direction – consistency to the designers intention.
At some level, that is in the DNA of the two approaches.
I do agree with the point that evolution is more likely to happen outside the ‘standards’ route.
We only need to look at things like XMLHttpRequest and Canvas as examples of proprietary vendor driven extensions to HTML that have later become key ‘open standards’, for instance, and the whole WHATWG/HTML 5 thing is another example, of where a bunch of vendors had to bypass the official standards process.
(Oh, and if Flash hadn’t existed, ActiveX would have been more widespread than it is. And I really doubt Microsoft would have bothered with x-platform/CPU versions).
November 24th, 2009 at 1:53 pm
Hi Ryan,
fair enough: Lets leave out the hyperbole in your title and my Twitter reaction
I still think that you assert way more than what follows from your argument but I agree that my point is much weaker if I leave the title out of my considerations.
@JulesLt:
I don’t think anyone says that standards committees are a good venue for new inventions. That’s not how the W3C works, though (ok, it sometimes does, but it’s not supposed to). Features get proposed by private entities and included into the standards if enough vendors express interest in them. A very good example are CSS transitions (
http://www.w3.org/TR/css3-transitions/), which were first introduced into WebKit by Apple and then proposed as a CSS3 module.
The point about ActiveX is quite interesting, never thought of that.
Ok, enough rambling, now I’m off to releasing 1.0 of SwiftSuspenders, my first substantial open source contribution for (gasp) the Flash platform
November 25th, 2009 at 3:42 am
If Flash were an open source platform, I’d agree with you. If it were seriously targeted as an innovative platform where things can happen before standards, it would be open. For mine flash and Siverlight are slow, buggy blights on the world of the web and hopefully are made redundant by more open and inclusive solutions. The web is supposed to be open, hence the technologies it is built on should be the same.
November 25th, 2009 at 5:07 pm
In this context it would be interesting to analyze how the innovation of open standards based web platforms can contribute to your own business. The model Adobe chose with the Flash plug-in has been very successful. Google and Apple on the other hand are very business focused and profitable companies – but they still invest heavily in browser technologies supporting open standards. There are not many companies which can afford to spend large amounts of money on innovation without a real business model behind the technology they deliver. But we have seen a lot of innovation in web applications outside the Flash universe, and I still remember how stunned I was when I first used Google Maps.
Apple has shown with the iPhone and Webkit, with HTML+CSS based UIs inside the iPhone version of Safari that it’s possible to build compelling UIs for millions of users without using Flash or Silverlight. I’m not impressed at all by mobile Flash applications, Flash Lite and attempts to bring Flash&video to mobile devices in the past years.
I think Apple made the right decision to not install a Flash Player on the iPhone until now. Flash Player 10.1 might change the situation, based on the performance shown in first demos of mobile versions of the new Flash Player.
But then: Who is using a Flash based email client, and what limitations did Flash have – until the release of Flash 10 – rendering text within the browser? How long did it take to implement Unicode support, support for other alphabets and text directions?
We can’t change the situation we have today, with a smaller percentage of very modern browsers, and then many companies and users still equipped with old versions of IE. But the evolution of browser based technologies in the past years has clearly shown that a situation where one company dominates areas as much as Adobe did dominate the world of browser based animation, video, multimedia and convergence is not good for us. At the same time the creative tools developed by Adobe provide a lot of value to many creative people in areas like web design, UX and interaction design, game design, etc.
A huge step ahead for the open standards based web would be the release of visual tools for building modern UIs and user interactions – at the same time providing an integrated workflow for design, interaction design, UI engineering and application development. Creating highly interactive and animated UIs for the browser should be possible without deep knowledge of JavaScript, CSS and coding. And that’s an area where Adobe does a pretty good job (not a surprise when you realize that Adobe/Macromedia/Macromind have a total of 25 years of experience in developing timeline-based animation tools).