Why Flex: What Flex can do, but HTML/AJAX can not?

Question

I want to understand the position of Flex in RIA competition. I used Flex because of AdvancedDataGrid and the Charts. It is impossible for HTML to do some thing like these.
The Transition Effects are nice, but not important in enterprise environment. The initial loading time of Flex is much longer than that of HTML although Flex is more reponsive than HTML/AJAX after Flex is loaded.
Any comments?

Answer 1

It's not always about what it can or can't do--a lot of this is about doing the same thing better.  And think about the origins of Ajax--it's basically a hack.  I know the same can be said about a lot of technologies (even the Internet itself), but why use something that is essentially a hack when you have technology that was engineered for the purpose you have in mind?  Building hacks and fixes upon more hacks and fixes is how we end up with a pile of junk at the end of the day in IT.  It's time to start calling things what they are, hacks, if they are that.
Flex has better performance, if you make use of AMF.
HTML and Ajax can't use AMF and has no counterpoint in this area. AMF is much faster than the crummy Ajax hack can muster (especially once you start throwing crap like Dojo into the mix).
http://www.jamesward.com/blog/2007/12/12/blazebench-why-you-want-amf-and-blazeds/
Check AMF times compared to JSON Ajax times, bandwidth.
Flex doesn't care about the user's browser.  Sure, the HTML/Ajax hack has frameworks to help you deal with that, but do you really want to worry about upgrading that trash (or framework bugs) when a new version of IE, or Opera, or Firefox, or Chrome, or Safari, or even another browser comes out and breaks something?  A newer version of Flash won't break a Flex app.  Also, other than Chrome, other browser JavaScript implementations are slow, slow, slow.  The more complex your Ajax app gets, the slower it will become--that's lame.
User experience. A well designed Flex UI is much nicer and easier to use than even the best "JavaScript/DHTML" hackfest.  Just look at these new Adobe forums--JavaScript out the wazoo and they suck on anything other than Google Chrome.   It's useful for marketing too--not just for commercial stuff, but marketing to upper management for enterprise applications.  Everyone that looks at a well designed Flex app will be blown away by it.  And then you get your pay raise.  (lol - assuming it works, too)
Superior development experience and development time.  Using Flex Builder you can mock up an entire "seemingly functional" prototype application within a ridiculous amount of time.  You can even run it without a back end or a server of any kind, just a standalone Flash player or browser.  You also have the GUI design mode, that shows you exactly what the user is going to see in the browser--no matter what browser they have, no matter what their settings are, etc.  No more HTML crap to deal with or (God forbid) tehnologies like JSP to generate your dinosaur code.  Documentation goes from the kludge of junk out there for HTML/Ajax/JS/etc to the excellent and very Javadoc-like ASdoc.
And no more choosing between the dozens of hacky and crummy Ajax "framework" patchworks out there, that are needed to workaround all the problems you'd run into without a framework.
That's just what I can recall at the moment.  I will NEVER go back to that God-awful Ajax/HTML hackfest people are calling RIA development.  Sure, Flex has it's issues and downsides too, but they aren't usually major enough to be of concern, even in an enterprise environment.
BTW - It's not really Flex vs. HTML/Ajax.
It's more like Flex vs. HTML/DHTML and AMF vs. Ajax/JSON.  But, Flex can talk to Ajax too, and it can even be embedded in HTML.  Although personally I feel that's a bastardization of the technologies, sometimes I guess people have a need to do these things.