Capping IT Off

Capping IT Off

Goodbye Flash, Silverlight, AIR and other plug-ins

Techniques that require a plug-in in the browser are dying. However Silverlight is not the one dying, since it was already dead just before it started (why even try to penetrate a market with a product that is not finished and with a competitor that is 4 blocks ahead). Therefore we can conclude that Flash will disappear, AIR will disappear and finally those ugly Java applets (who ever thought those would be useful on the web, waiting 5 minutes to have an applet loaded) are gone too.

For me it is clear that all plug-in based techniques will be replaced by more native techniques like JavaScript. JavaScript was forgotten due to some browser wars which ended up in the result with two rather incompatible implementations of JavaScript. However with the several incompatible implementations becoming less dominant (there now is only one implementation that is questionable at some points) and the rise of the libraries like JQuery, Prototype, SproutCore, YUI, MooTools, ExtJS it is clear that JavaScript is back on its feet again and it is running to overtake all these plug-in based techniques. JavaScript is platform independent, as most plug-in based techniques are not.

The limitation in JavaScript used to be the 'flashy' things: drop and drag, animations, interoperability and other nice and fancy stuff that was either limited by the technique or by the processing power of the client. Nowadays this isn't a limitation anymore, do you want nice animations, you could use script.aculo.us, JQuery UI, or Processing.js. Interoperability is arranged in almost all libraries. Do you want applications that feel like desktop application, think of your design and build it, just like 280 Slides, Google Docs, Zoho and SproutCore gallery. You are no longer limited, you can make these things happen with techniques that are native for all browsers: HTML, CSS and JavaScript!

It will be hot JavaScript winter, especially since everything is possible again. Flash, AIR and Silverlight aren't the only techniques that can make flasy desktop like nice solutions. JavaScript can do that too and JavaScript does not create a vendor lock-in requiring a specific closed source plug-in.

About the author

Rick Mans
12 Comments Leave a comment
AIR is not a browser plug-in. It's for building desktop applications using web technologies, including JavaScript.
So you can use a combination of your favorite JavaScript library and AIR to build desktop applications, including stuff like read and write file system access from JavaScript.
Javascript cannot even build multithreaded apps without hacking and you think it's gonna be desktop quality?? Flash, SilverLight, JavaFX and about just every plug-in will blow its pants off in a "flashy" contest. The only platform incompatibility issue is that they may not run on Linux which counts as what, less than 1% market share? As if anyone would care. LOL. The future belongs to those designed to be an application framework like SivlerLight(XAML), JavaFX and Flash(MXML), not the one designed to be document like HTML/Javascript. Repeat after me: Die Javascript, die.
rimans's picture
@Lars why would you install a closed source application to run web technologies on your desktop. A browser is made for these things and if you have a browser button fobia you could install Prism (which is a stripped version of firefox).
rimans's picture
@klimzk multi threading is not really an issue on the web and it is possible via a work around. Currently Linux is a rather good alternative, especially since Windows becomes an heavier OS every release. However the trend is smaller and more minimum PC's and laptops (eee-pc!), these laptops do run Linux. The iPhone is also a good example, it runs javascript, it does not run Flash.
It is not only about incompatibility, it is also about vendor lock ins. Why would you want to use software that can only be changed by one party. That can stop supporting your favorite browser at any moment. Is it your decision to use certain functionalities, or is it the decision of the vendor of your plug in?
@Rick Mans
iPhone in fact proves that end users prefer fat, rich & desktop like user experience to the thin & dull provided by a free handset coming with every cell phone plan. You can check their latest iPhone SDK which is loaded with support to build NATIVE cocoa apps. In case you miss it, let me stress it's NATIVE. If the trend is to get thin then why they bring the fat, heavy native platform dependent SDK into iPhones? The answer is obvious: javascript is not up to it. Even Linux, Ubuntu for example, is trying to catch up with Windows and OS/X in terms of rich user experience. With hardware becoming better and more affordable every passing month, the trend is definitely toward the richer end not the thinner one. You are suppose to make use of the better hardware instead of ignoring them.
As for lock-in, gee, who cares other than religious developers? iPhone is a big lock-in, but does that hurt its popularity? The end users don't care as long as it works.
And really what's the difference between locked-in by M$ and by Mozilla? They both have an agenda behind it, which is to stay dominant on the internet. Mozilla, back then Netscape, invented this lame Javascript and then pretty much didn't do anything to improve it for a decade until they smelt what Adobe and M$ were cooking with Flash and SilverLight. They want Google to insert AdSense in a web page and bankroll Mozilla with a share of that massive revenue, so they prefer a thin and text-based format under their control despite users clearly want richer and more interactive content. I don't think they are more trustworthy than M$.
If Javascript wants to stay competitive, it has to get as rich as SilverLight, Flash and JavaFX. I mean let's face it: the CPUs are heading toward multicore and yet Javascript still requires work-around just to run multi-thread? Come on, that ain't gonna cut it. End users want great UX, that's the key. Whoever meets that need is going to win the market. If Javascript doesn't address that, users simply switch to those who do.
As Lars points out AIR doesn't belong in this mix. It's a desktop runtime. Just like you need Java or .NET installed to rune Swing/Winforms desktop apps, you need the AIR runtime installed to rune AIR desktop apps like Twhirl.
I disagree that browser plug-ins will die. Sure JavaScript have come along way, and you can do some amazing stuff with the new modern JavaScript libraries. I'm a strong believer in progressive enhancement, giving the user the best possible experience depending on their platform. (http://en.wikipedia.org/wiki/Progressive_Enhancement)
I also disagree that Silverlight won't stand a chance. I've been working with the Silverlight 2 betas over the last half year, and is just amazed by the things you can do. Having a .NET runtime, inside the browser, running on Mac and Windows (and soon Linux) is just amazing. Being able to write client web applications in C#, VB.NET, Ruby or Python is really attractive, compared to writing ActionScript in Flash. I think that will be the strengths of Silverlight. Flash has a animation/design background, with a development platform "attached" on later, while Silverlight is a powerful development platform with a sleek UI layer on top of it. If you look at the enthusiasm about Silverlight in the .NET developer community I really think Microsoft has an awesome piece of technology on their hands. Sure Silverlight 1 hasn't been a big hit, and it isn't a really interesting release. It feels like something they had to get out there to take part in the RIA-race, but Silverlight 2 is going to be "the big release", with the adding of a .NET developer platform.
I think Flash is going to benefit from the competition from Silverlight, just as Java benefited from the competition from .NET and C#. Adobe have already announced several cool features to differentiate them self from Silverlight, or to catch up (like HD video support through h264) or adding hardware accelerated 3D (something you can't do in Silverlight).
Jonas Follesø
Senior Consultant, Capgemini Norway
rimans's picture
@klimzk please be aware that there is no vendor lock using e.g. Firefox. You are free to fork your own version (like is done by creating Iceweasel).
Whether Linux is catching up with windows and OS/X in terms of rich user experience or whether Linux already is way ahead of them, is just a matter of opinion and personal experiences. If I look at Linux, Windows and OS/X all three have the same functionalities and all three have the same bling bling possibilities. The other things are just a matter of perception.
I agree with you that eindusers prefer a great user experience, however that does not directly equal the use of an external plugin. Be aware of the fact that simple web applications build without these plugins already resemble desktop like behaviour. The plugin based techniques can do exactly same. These plugin do quite resemble what is published in the paper Hamlet's Blackberry:
Though paper appears to be a relatively “dumb” medium, it too performs tasks that require special abilities. And many of paper’s tricks, the useful purposes it serves, are similarly products of its long relationship with people. There are cognitive, cultural and social dimensions to the human-paper dynamic that come into play every time any kind of paper, from a tiny Post-it note to a groaning Sunday newspaper, is used to convey, retrieve or store information. Paper does these jobs in a way that pleases us, which is why, for centuries, we have liked having it around. It’s also why we will never give it up as a medium, not completely. For some of the roles paper currently fulfills in our media lives, there is no better alternative currently available. And the most promising candidates are technologies that are striving to be more, not less, like paper. Indeed, the pertinent question may be not whether the old medium will survive, but whether the new ones will ever escape paper’s enormous shadow.
You can compare paper to the conventional browser techniques (e.g. HTML, CSS, Javascript) and the plugin based techniques to the alternative of it.
Full paper of Hamlet's Blackberry is available here: <a href="http://www.ksg.harvard.edu/presspol/research_publications/papers/discussion_papers/D39.pdf" rel="nofollow">http://www.ksg.harvard.edu/presspol/research_publications/papers/discussion_papers/D39.pdf</a>
rimans's picture
@Jonas I think using Silverlight is a matter of taste. Since I am not quite fond of the Microsoft tactics of yet introducing a new standard (as they are currently trying with OOXML, like they did with JScript, or like they did with the good old marquee tag in the old days), especially since flash is around that can fulfill all the market needs. However competition is healthy, though Flash and Silverlight are in my opinion just temporary fixes for the old days when Javascript didn't mature yet.
I agree with you on progressive enhancement, that is one of the best parts that can be done by Javascript. However my experience with plugin based techniques is dat progresse enhancement is hard and quite often not possible. What are you experiences with this?
A Silverlight related question: what do you think about the fact that Adobe, Google and Yahoo are currently to make Flash searchable (or better said: findable) via searchengines? Is this a blow for Silverlight, or has Silverlight distinctive features that makes it able to ignore this extra functionality?
Hi Rick,
You got some good questions and comments. I know Microsoft have done some ugly things when it comes to standards. But they are getting better. They’ve done a good job in the web service space, working together with other vendors defining the WS-* specifications. They have also gotten involved with the OpenAJAX alliance, as well as the ODF committee. The announcement of full support for ODF in Office 2007 SP1 shows that they are listening to customer demand, and understand that customers’ demands open standards. So things are getting better… I guess the “good things” with plug-ins is that it enables innovation without crippling standards, like you say perhaps as a temporary fix until JavaScript and other open standards catch up. Google Gears is another example of a browser plug-in providing a “fix” of offline storage while we wait for all the browsers/users to upgrade and get the offline support that is part of HTML5.
Online media (audio/video) is another example where we have to rely on plug-ins today. Even though HTML5 specify a video-element it does not say which codec to support. FireFox have made announcements that they will support OGG as the codec in FF3.1, but it will take time before Microsoft and Apple catch up. Will Apple be willing to support OGG, or will they try to enforce their QuickTime format? And I doubt Microsoft will be willing to give up WMV investments, and will probably support the video-tag through some Silverlight/WMV integration… All in all, innovation through standards takes time and moves slowly. At least Silverlight 1.0 with its 1MB install gives you an easy option to get cross-browser, cross-platform (Mac and Windows) HD video support in the browser...
With progressive enhancement I guess it depends on what you are using Flash/Silverlight for. If it’s a small “part” of the application/site, like a music/video player, it’s fairly simple. In those cases I just see Silverlight as another level of enhancement, above JavaScript. So I start with standard based XHTML markup, add CSS, add JavaScript, and if possible, add Silverlight. One example is a web site I’m building with my girl friend (who is a web designer) for a music artist. If Silverlight is enabled, the header on the discography page changes to an interactive, iPhone/iTunes like music player. With some nice 3D like effects, flipping covers, track lists, audio player, lyrics scrolling in sync with music etc. I guess I could have achieved some of the same things using a JavaScript library, but it was darn easy to do using visual design tools like Expression Blend (which speaks XAML, the XML mark-up used by Silverlight). If the user doesn’t have Silverlight installed they can still look at the covers, track lists and lyrics, and download the albums. But it’s not as “premium experience” as if they use Silverlight.
As for search ability Silverlight faces the same challenges as Flash applications (and to some extend AJAX applications). When it comes to search I think it’s a big difference between applications and pages, and Flash and Silverlight are often used to build applications and not present content (for that HTML is a better fit). But, you would still like to be able to find your Silverlight assets using search engines. It’s absolutely possible for the search engines to index Silverlight content. Silverlight 1 exposes the XAML files as plain text files, while Silverlight 2 uses compiled code and zipped XAP files. That being said, the search engines could still download the XAP files and look inside them. Another solution is to have “invisible”/HTML content alongside your Silverlight pages. This would fit nicely with progressive enhancement since you’re exposing the pure content of the app as standard HTML, but the user is actually interacting with it through Silverlight. The search engines could pick up this content when crawling the pages, but when people click the search result they get the Silverlight app.
Another, and perhaps the easiest way, would be to provide good site map files with URL end-points to your Silverlight application states. This is the “old” problem of bookmarkenabled, back/forward navigatable AJAX applications. Silverlight and Flash faces the same challenges, but using JavaScript and browser integration it is possible to create Silverlight applications that interact with the browser and updates the URL’s so that users can bookmark “states” in the application (even though they never navigated to a new HTML page). These “end points” in your application could be listed in a site map file that the search engines could index to understand the structure of the RIA.
There are already several examples out there on how to build searchable Silverlight applications that integrate nicely with the browsers navigation scheme.
Cheers,
Jonas
rimans's picture
Thanks for you extensive reply Jonas. I agree with you that HTML5 will indeed create standards for a lot of things that are currently handled via plugins.
P.D.Babu
E.Mail:babu_6531@yahoo.co.in
Mobile:9790339427
Carrier Objective
To Work In a creative and challenging environment that encourages learning and creativity, provides exposures to new highest and simulate professional growth
Educational Qualification
10th,Doing B.A in TN open univercity
Technical Qualification
MultiMedia
Technical Skills
designing
Photoshop,illustrator,coreldraw,indesign
Personal details
Father name : P.Dhamodararaju
Date Of Birth : 03.01.1990
Language known : Tamil,English&amp;Telugu
Nationality : Indian
Contacts
Permanent Address
P.D.Babu
No:85,Nehru Nager,
Tiruttani(TK)
Pin:631209.
Thiruvallur(dt)
Present Address
No.10,New street
Chittoor Road
Tiruttani-631209
Declaration
I here by declare that all the information mentioned above is true to the best of my knowledge.
Date Regards
Place P.D.Babu
<strong>Hypertext Markup Love 5</strong>

Currently my markup loves revives since HTML5 is being specified. HTML5 does contain so much good elements, functionalities and other technologies that were missing since the introduction of HTML4 (almost 10 years ago in December 1999). I will give yo...

Leave a comment

Your email address will not be published. Required fields are marked *.