Does HTML5 only replace the video aspects of Flash/Silverlight?

I see a lot of talk how HTML5 video tag will kill Flash. But while video is the most widely used part of Flash/SL, it's only a small part of their technical abilities. For instance you can write a game using full 3D graphics and socket connections in Flex, and serious business applications, etc.

Is the thinking that Javascript will kill those parts of Flash/Flex/SL? Because while that seems feasible now for even qui开发者_开发问答te rich web-apps, what about any kind of high-performance app like real-time graphics?

You can do some pretty cool things with HTML5, Canvas and the like. Check out the Google-hosted http://www.chromeexperiments.com/ (which, really should be named HTML5 experiments, as most run fine on HTML5 browsers). You can get an idea of what is possible, including 3D rendering, complex games and some pretty amazing animations.

Sketchpad by Michael Deal

basically, this question has been around already.

The answer in short: no, HTML5 will not replace anything. HTML5 will however offer a standard for features currently only available through plugins. Once HTML5 is released, which is scheduled for 2022, it will be a great thing. Still, even by then, if IE survives, I doubt it's support will be outstanding.

There has been a very similar question to which I provided a rather lengthy but detailed answer: should web developers learn flash

My personal opinion about anyone who thinks HTML5 will replace 3rd party plugins is, that they lack basic understanding of HTML5's role or sufficient knowledge about any plugin and have little if any grasp of how the web evolves.

Yes, the day HTML5 is sufficiently spread for large companies to rely on it, many of them will be able to replace 3rd party plugins within their web applications. However, as pointed out in my other post, the web is constantly evolving. HTML5 does not provide new features, that weren't available using plugins. And HTML5 does not provide all features currently available using plugins. New types of apps, services and content distribution mechanisms will arise. Also, as long as JavaScript stays fully dynamic, JS runtimes will never be able to provide the same speed as runtimes designed to run bytecode generated from statically typed languages.

Personally, for the client side, I basically only target the Flash Player, because to me it is the most convenient platform. I am not afraid that HTML5 might kill flash, for a simple reason:

Until HTML5 is really usable and largely supported through sufficient user adaptation, it will take several years. In the same time, all major plugins will continue to evolve, as well as their eco-system, including developement tools, cross-compilers and compatibility layers.
Today, you already need no knowledge of HTML, CSS and sometimes even JavaScript to create webapps, using GWT, qooxdoo or other tools.
HTML+CSS+JS represents nothing more than a platform that more and more languages are able to target. Using reasonable amounts of abstraction, one will sooner or later be able to develop apps in a totally platform agnostic manner, causing virtually no overhead for the lion's share of all apps: GUI (including localization, validation etc.) and application logics.

In the end, there's no reason to get excited. Currently, both Flash and Silverlight exceed HTML+JavaScript implementations in speed and features exposed through the available API. That is, why people use it. Altough unlikely, some day they may become obsolete. In this bright, bright future, far, far away, I will be happy to compile my sources to run in a JavaScript interpreter rather than on AVM2.

On a vaguely related note: check out Haxe.

There are many high-performance apps with real-time data streaming and graphics in production today that have been built with Flex. You can see some examples of these in the flex.org showcase. As always, deciding on a technology comes down to what is being built, who it's being built for, and who is building it. Flash, Flex, Silverlight, HTML5, etc will all coexist.

The discussions I've been part of on this subject, pretty much always seem to come to the conclusion that, while the HTML 5 standard will be a a great thing, once it's in place. It will by definition be a standard, standards take a long time to change, therefore innovation will be driven largely by plugins like Flash and SL.

And yes you can do a lot of what Flash and SL can do with Javascript, however Flash/Flex has the strong point of being used for desktop installed Air applications, and Silverlight has the similar thing with their "Out Of Browser" functionality. These are things where they clearly have the upper hand over Javascript.

At the end of the day it's not a matter of who will kill who, but rather which tool is the right one for the job and what skills does you / your development team have.

That's my 2c anyway...

Did Flex replace HTML forms? Have HTML forms killed Flex? Has C++ replaced C? Has Ruby or Lua or Python killed Haskell or Scheme or assembly?

Seriously, what is the obsession lately over whether HTML5 will kill plugins? Is it so hard to imagine people using Flash for content that needs to stand out and maximize expressiveness, and using HTML5 for content that needs to utilize standard metaphors and maximize accessibility?

In other words, no - the thinking among thinking people is not that the HTML5 stack will kill Flash/Flex/SL. If things work as they should, the plugins will keep innovating what you can do with proprietary techniques, and HTML will keep taking the most successful of those innovations and standardizing them. And if things don't work well - if a plugin fails to innovate, or a new version of HTML is done badly, then they'll fade into obscurity as developers stick with whatever solves their problems.

I think it depends on the authoring tools and the efficiency of the browsers. It is reported that the upcoming Flash CS5 export content directly as a HTML5 canvas. So the canvas can be a quite reasonable flash substitution.

For "high-performance app like real-time graphics", see WebGL. Mozilla and WebKit already have experimental support for WebGL, which is based on the <canvas> element.

For sockets, there are WebSockets, which are also experimental in some browsers.

For background processing, which can improve the user experience, there are web workers (experimental again).

A lot of experimental stuff, but they are improving them constantly, so at some point, even without an HTML5 standard defining them, we might see games specifically targeted for a certain browser (actually, there is some proof of concept but I can't find the link right now).

I think JS already has killed Flash/SL/JavaFX. JS is the most used language on the web, so just add a few features, enhance some tidbits and voila. Flash is alive because of videos and games, no serious website uses Flash nor any other RIA. Just wait, it's going to be a hell of a ride the next few years.