HTML 5 SHOULD Kill Flash and Silverlight
June 16, 2009 31 Comments
A great article has appeared about how HTML 5 really should finally kill off the proprietary Flash and Silverlight browser add-ons.
HTML 5, a groundbreaking upgrade to the prominent Web presentation specification, could become a game-changer in Web application development, one that might even make obsolete such plug-in-based rich Internet application (RIA) technologies as Adobe Flash, Microsoft Silverlight, and Sun JavaFX
All this focus on Flex/Flash and Silverlight is really beside the point in my opinion. Sure maybe today, we have to rely on these proprietary browser plugins to deliver content to users, but the real innovative developers and companies are going to standard on HTML 5 and in turn revolutionize how users interact with data. We all want faster web applications and the only way to deliver this is to use HTML 5. Of course some companies can’t get their act together to support it (I’m looking at you Microsoft), but given how positive people have been toward the Google Chrome browser and how it works with their web applications, I think we are really very close to a revolution here. The question we need to ask ourselves is to you want to be in the front, or the rear of change?
MechaHTML5 pushes proprietary browser add-ons to the side.
HT: DF
About James Fee
We’ve been focusing a lot of the javascript API for web mapping, and this is pretty much why. When the choice came down to Flex or Javascript (we had already dropped the possibilities of WebADF and Silverlight), one of our programmers said, quite simply, Flash should go away with HTML 5. While we’d almost certainly redesign a few times before that happens, our redesigns could focus more on improved functionality and taking advantage of more features of html 5 as they become available in different browsers; instead of worrying about when we would have to get off of ActionScript and how much of the code base we would be able to port.
Was it the right move? Time will tell, but we are not hitting any roadblocks yet going the JSAPI route (especially when we start looking for opportunities to add in Ext, jQuery, openlayers, and other libraries).
James:
As a proud philosophy major, I’ve been trained to note when the prescriptive “should” has slipped into the conversation. “Open”, “platform agnostic”, not “vendor controlled” are positions as unassailable as Mom and apple pie.
However, if Microsoft doesn’t play with full support in IE, then it’s hard to see where HTML 5 is going to make inroads on the desktop. The experience of the “open” SVG standard is instructive.
No, if HTML5 is going to make it, it’s because it’s the standard on mobile browsers. With the mobile marketshare being dominated by Apple, and Android making its presence felt, there’s no MS incumbency to deal with and developer momentum may carry over to the desktop.
Finally, as geospatial-types, we’re allowed a dose of skepticism as to whether open standards bodies can effectively keep pace with the ever-increasing rate of technological change.
Brian
James, I think Brian nailed it… wait and look at the eventual deliverables, and what can actually be done with them.
Reality Distortion Fields made with the cube of time….
jd/adobe
John, don’t get me wrong. I understand why I have to use Flash today. I think the architecture of plugins is wrong and just because something is available doesn’t mean we should use it.
The endpoint of all this has zero to do with Adobe and Microsoft and everything with delivering applications to clients. Standards are standards and we should build upon them rather than foist noise between my users and their data.
@Brian: Should was meant with one eye on the looking glass and not my personal opinion. I believe that Flex/Flash and Silverlight are not how we’ll be developing applications. Sure mobile is the perfect example (Apple and Google), but the desktop can benefit so much more with Canvas. Sure Google’s hacks to get IE to support HTML 5 just put ourselves back into the proprietary plugin problem, but we are not that far from WebKit/Firefox owning 50% of the browser market.
Tide is turning.
I think this is the most important quote in that article
“The HTML 5 timeline states that it will be at least a decade before the evolving HTML 5/CSS 3 efforts are finalized, and it remains to be seen what parts will be implemented consistently across all browsers.”
A decade eh?…
Utopia is always a decade away. Fusion power is always two decades away.
and this is why when people blab about browsers supporting standards I always roll my eyes.
The article ignores a significant issue: the video codecs used by HTML5 (Ogg varieties) are considered, at least by some, not really adequate for HD.
Win7 comes with H.264, and the above codecs are likely to rapidly catch up with the current proprietary options, but at least in the short term it’s a problem.
I’d like to think Wave will be one of the killer apps that drive browsers, including IE, to HTML5 much faster than they might move otherwise. But Microsoft gains nothing from increased Wave adoption, or any other new HTML5 application. Such things have the potential to make the modern operating system just another easily abstracted away layer.
“However, if Microsoft doesn’t play with full support in IE, then it’s hard to see where HTML 5 is going to make inroads on the desktop.”
It’s against Microsoft’s best interests to support these new features when they are pushing Silverlight. Their dominant position in the browser space has held back innovation for years. The only way to get support for these features in Internet Explorer is to force them via applications that require HTML 5. Considering how much faster and cheaply RIA apps can be developed using modern browsers and the features of HTML 5, it should be possible to convince clients to go this route in some cases.
Microsoft is well on their way to getting below 50% of the browser market share, they have to support such features. The latest Google 3D warehouse embedding feature supports IE and uses no Flash. http://twitter.com/mweisman/statuses/2198493751
Where’s Mothra?
What about The Twins?
What would you say about http://www.openlaszlo.org, it can generate flash and dhtml.
And we are working on HTML 5 support making a smooth transition from Flash to open standards possible!
From a developer perspective, I am loathe to develop in JavaScript. I need an IDE with GUI, compile & debug, variable type-checking, etc. I dumped JS/HTML because developing in that framework could not be more backwards. I love the idea of HTML 5 and agree we need to ditch Plug-ins. But until JS gets a FlexBuilder-type development environment I’ll never consider it for large-scale apps. And a single standard for consistent rendering across browsers is nice too.
Just my (limited) perspective.
Josh:
I don’t do much JavaScript work myself, but have you looked at Eclipse-based IDEs with JS support e.g. MyEclipseIDE?
http://www.myeclipseide.com/htmlpages-func-display-pid-57.html
MyEclipseIDE is aimed primarily at Java/JEE development, but the latest versions claim to include Ajax support.
Hahaha, three letters: G-W-T baby!
Brett, when you start looking for opportunities to add in ExtJS, jQuery, OpenLayers, or other libraries to your JS API approach, I strongly encourage you to look into GeoExt (http://geoext.org/). It’s a nascent effort between Camp2Camp and OpenGeo and a community of JavaScript developers to integrate OpenLayers into ExtJS and provide solid reusable components for (truly) web-based mapping applications.
Things that need to happen for HTML 5 to take off.
1. Real development tools
2. Standard implementation of the spec across all major browsers.
3. Standard video codec
4. Since HTML 5 is really just an engine you need some type of framework to make it easy to work with it. To create sprites and spatial objects.
I don’t see this happening for a very long time. Plus this is assuming that the plug-in world is just going to sit by and twiddle there thumbs. Not creating new features in Silverlight/Flash.
Canvas is also never going to have the performance or scalability that a plugin will have. You’ll never run quake using canvas.
Quake in silverlight: http://www.innoveware.com/quakelight.html
Its an extreme example, but do you really want to be limited by the spec?
Classic GIS dev complaints. Real development tools? LOL, they want bloat to go along with their bloated plugins. Frankly using EditPad to develop web code is perfect, optimized and clean. Quite a jump from mapping vectors in the browser to running Quake. If you want to talk about running Quake, then I don’t want you anywhere near my code optimized for mapping.
No wonder people look down on GIS developers.
I guess you need to put your bifocals on.
Did you miss the part where I said “extreme example”.
Maybe if I wrote it using a emac editor you would of seen it.
Emacs editor? Luxury! In my day we had to code in binary with 1′s and 0′s, and if we didn’t have 1′s we had to use lower-case L….
Lefty, I’m sure you do just fine developing “web code” in EditPad.
I’m more interested in developing integrated web apps as efficiently as possible. To that end, I prefer to use a robust IDE.
I’m not a GIS developer, but I do find them adept at getting jobs done with the tools they have available. If they find those tools lacking, I think they’re right to expect more.
2 cents.
Pingback: James Fee GIS Blog » Blog Archive » Adobe’s Views on HTML 5
Not one word on how HTML 5 will improve Javascript’s painful shortcomings…?
I’ve been doodling with Silverlight for 3 months and right now I’m loathing having to go back to my Javascript app and upgrade it to work with my new database. The wholesale changes I’ve made to the DB took 2 days in Silverlight to account for – I’m estimating a week’s work in Javascript..
What impact do you think this may have on HTML 5?
Browser Vendors Force W3C To Scrap HTML 5 Codecs (Slashdot)
That was a short revolution.
Silverlight and Flash here we come!
Nah, HTML 5 is good. http://www.w3.org/News/2009#item119
Aw, now I’m just getting confused…
Meanwhile, look at all the pretty pictures in Firefox 3.5…
http://www.mozilla.com/en-US/firefox/video/firefox-3.5.html
Isn’t that nice? Fast, pretty, no memory issues. The future is a great place.
“The future is a great place.”
Yeah, send me a postcard when you get there!
javascript is one of the biggest hurdles
we’re seeing more and more CPU intensive RIAs doing things like processing image buffers, tasks where flash struggles and javascript would be just unworkably slow if it had the APIs required to do something like that
also, a large number of programmers feel that type safety and static binding are important, and javascript forces a less reliable model on those programmers
there are also many, many features in flash and silverlight that html isn’t even proposing to address
html is essentially falling further and further behind
flash and silverlight will be hosting hardware accelerated 3d games on par with quake 3, and full applications that duplicate desktop applications like photoshop, before html can do what flash was doing 10 years ago (audio/video streaming, basic vector graphics with a sensible frame rate)
the fact is that adobe and microsoft are spending far more human hours and effort on their specifications than is going into html, and it shows