Adobe has recently shipped a new beta of Adobe Flex 2.0, a set of client and server-side tools and technologies for building rich internet applications on J2EE application servers. This new beta introduces Flex Enterprise Services 2.0, which includes some interesting functionality: Flash apps now have a publish/subscribe messaging service available to them and Adobe has used this to create Flex Data Service, an elegant way to transfer data to and from the UI with much less development effort than has previously been the case with AJAX and other rich internet applications.
Development with Flex 2.0 has become easier with Flex Builder now being an Eclipse plug-in, better debugging tools, and a whole host of other improvements. A standalone command-line compiler and debugger will soon be available for use with any development tool. This tool will be part of the Flex 2.0 SDK that Adobe will be releasing shortly.
The Flex 2.0 beta is free and is available at Adobe Labs.
- relationship to laszlo? by Gerald Loeffler on February 03 2006 11:30 EST
- Flex 2.0 Beta 1 Shipped by bad mASH on February 03 2006 12:39 EST
- Flex 2.0 Beta 1 Shipped by Pavel Tavoda on February 03 2006 13:08 EST
- Flash can be useful by Aaron White on February 03 2006 14:42 EST
- Also... by Aaron White on February 03 2006 02:56 EST
- Less browser compatibility issues too by Rob W on February 03 2006 03:38 EST
- Flex in the Enterprise by Andrew McKenzie on February 03 2006 16:36 EST
- A sane post by Kapil Israni on February 03 2006 04:58 EST
Flex in the Enterprise by Marc Logemann on February 03 2006 05:07 EST
- Flex in the Enterprise by Aaron White on February 03 2006 05:15 EST
- Flex in the Enterprise by Orlando Roebuck on February 03 2006 06:40 EST
- Flex in the Enterprise by Andrew McKenzie on February 03 2006 07:44 EST
- Informed Opinions by Stacy Young on February 04 2006 06:13 EST
- idiots ? by romain romain on February 06 2006 03:44 EST
Flash by Konstantin Ignatyev on February 06 2006 04:44 EST
- Flash 8.5 by romain romain on February 07 2006 03:39 EST
some info by Mike Chambers on February 07 2006 04:13 EST
some info by Konstantin Ignatyev on February 07 2006 11:51 EST
some info by Michael Jouravlev on February 07 2006 03:02 EST
- some info by Konstantin Ignatyev on February 07 2006 04:05 EST
- some info by Michael Jouravlev on February 07 2006 03:02 EST
- some info by Konstantin Ignatyev on February 07 2006 11:51 EST
- Flash by Konstantin Ignatyev on February 06 2006 04:44 EST
- Flex 2.0 Beta 1 Shipped by Frank Bank on February 04 2006 03:12 EST
- Flex Pricing Comparison by Mike Chambers on February 04 2006 17:35 EST
- Flex 2.0 Beta 1 Shipped by Hans Omli on February 04 2006 18:32 EST
- Laszlo by Don Dwoske on February 08 2006 21:08 EST
- Drool Factor...Flex without Server? by Abhay Aggarwal on February 16 2006 00:09 EST
can someone please enlighten me/us what the relationship of Flex is to Open Laszlo?
can someone please enlighten me/us what the relationship of Flex is to Open Laszlo? thanks, geraldFlex is a Macromedia product that became Adobe's after Adobe was stupid enough to buy Macromedia.
Laszlo has nothing to do with Flex, except, maybe - inspiration.
They both suck and AJAX is going to kill them both :)
I simply hate websites which use flash. They all suck : they make you wait while they render useless animation .
Only complete idiots would use this for an entire J2EE application.
When you use it right way, time for loading should be OK. But AJAX will win because of direct support in browsers. Flash is today equivalent of advertisement. Except for some sites which use it for presentations I don't see reason to download it. It's event better not to have it installed because you will get less advertisement (blinking useles windows) inside page.
Despite people's likes or dislikes I think Flash can be appropriately used to create applications which are practical.
Here is an example:
And personally, I haven't seen an AJAX site (and I've looked at a lot of supposedly "top" AJAX sites) which provides a user experience this nice, but that's my opinion, and it's probably only a matter of time before we see really good AJAX sites.
That being said, I haven't committed myself to one technology or another. I think it's too early to say one is better than another.
I think if this site were gmail.com, there'd be a whole different feeling in the industry.
One of the nice things about using Flash in this way rather than AJAX is that you by-pass the browser incompabtility issues - the app will work the same way in IE, Firefox, or Safari without any hacks.
It might not be buzz-word compliant, but there's something to be said for not having to deal with the idiosyncrasies of various browsers.
One of the nice things about using Flash in this way rather than AJAX is that you by-pass the browser incompabtility issues - the app will work the same way in IE, Firefox, or Safari without any hacks. It might not be buzz-word compliant, but there's something to be said for not having to deal with the idiosyncrasies of various browsers.
(Disclaimer: I'm inherently biased toward AJAX based on my occupation.)
Unfortunately this comes at a price, as Flash player releases for Linux tend to trail behind those for Windows and OSX. I'm already frequently having to switch over to my Windows machine to view certain web sites as Flash8 becomes more popular. While it's true that Linux support is often not a concern where a Flex application might be deployed, it can also be said that in such cases cross-browser capability is likewise not an issue.
In any case, it's my understanding that the popular AJAX frameworks take care of the majority of the cross-browser legwork for the developer.
This is true but if you go with openlaszlo vs. flex, it supports flash 6.
These compatibility questions are really relevant depending on the size of the dev and QA staff.
I also have concerns with flex 2.0 based on the flash player 8.5 requirement...I haven't seen if it will be compatible with older versions.
In terms of AJAX, I agree that the frameworks should eliminate compatibility issues, but I haven't seen a framework that provides a rich UI experience that is reasonably priced. I thought backbase was nice but the price is ridiculous....if you aren't a big company.
Please can we have an informed discussion on the merits of using this type of technology in the enterprise and hear from developers who have used the product especially in a commercial environment.
From my own experience this product is way ahead of the game. In terms of UI capabilities, richness and enterprise connectivity there is quite simply nothing like it in the market-place today.
For those of you making direct comparisons to AJAX then you are only just scratching the surface and missing more than a few tricks. Whilst AJAX obviously has the advantage of running without a plug-in, the implementations are still nowhere near as easy to design and develop and the end result is generally an approximation of what Rich Internet Applications are all about (usual suspects aside eg googlemaps).
I have personally seen significant business benefits brought about by using this kind of technology for a world leading corporation. We were able to develop a Flex Rich Internet Application in a fraction of the time it would have taken using AJAX + JSP / JSF / Struts technology. We made use of the inbuilt enterprise services (Remote Objects, Web Service, XML/HTTP Service) to provide seamless integration into a very large scale J2EE CRM application, all achieved without a single code amendment in the main CRM application.
The days of Flash as purely intro-screen eye-candy are truly over. Flex is a Presentation Services / Enterprise Integration Platform for those who like to get the job done whilst others drift in an ocean of web frameworks, un-maintainable script, cumbersome communication models, basic component sets that ultimately lead to an inferior user experience.
So go on folks, live life on the edge and download Flex2 Beta or Flex 1.5, take the top down and have a test-drive. You'll soon realise why this type of technology is highly significant as we move towards Web 2.0.
PS In response to the earlier post – “only an idiot would use this for an entire J2EE application” - only an idiot would make that comment without knowing the facts about a products capabilities.
I absolutely agree with you. How many times we have seen here, zealot's just bashing products/companies when they are in no way connected to it or have experienced it.
I think AJAX and Flash both represent viable platform's for rich internet applications. Both of them will continue to exist, however I feel one of them will probably have a lead over the other. AJAX has a lot of momentum, and is growing in standards, tools, industry support etc. Having said that, I have experienced a few flash sites and it provide's a credible UI alternative. I think one of the key points is design, sites dont wanna come off as too thick on flash UI. My experience is that it makes it slow and just does not give the "internet" experience.
Mind you flash has already played an important role in "rich user experience" for the internet. So far that role has been limited to static content. Its good to see its slowly getting out there for dynamic applications as well.
I think a good comparison is to look at yahoo's new email interface (beta - AJAX) and http://www.laszlomail.com (Flash based). My experience have been pretty ok with both the interfaces, though I will admit I havent used laszlomail as much compared to Yahoo.
But anyway, I downloaded Flex builder, so far it looks very neat. Looking at examples, I was able develop some basic apps with data being pulled from WebServices :)
I would like to see this tool be distributed free, that way you bring in more developers.
Andrew you are working for Macromedia right? At least your message sounds like you are working in the marketing department for Flex.
To me, the price of Flex is a joke when there is a competitor like OpenLaszlo. But i heard rumors that even the guys at Macromedia realized this and will radically change the pricing of Flex2. To me flash is a viable alternative to Ajax based RIA clients. Ajax might be the future but the frameworks are not mature and full features enough yet.But at the end it must be only the view. So i dont need all the bells and whistles from Flex, there OpenLaszlo is definitely enough in terms of client intelligence. One need a robust RPC/Webservice invocation framework and some events and some widgets. Thats it. No more no less.
So yes, there is a place for flash RIA. But right now i would allways chose OpenLaszlo. And yes, i made test drives with both products.
Look for the Flash SDK announcement on the labs site.
It looks like it will do a lot of what openlaszlo does and will be free.
To me, the price of Flex is a joke when there is a competitor like OpenLaszlo.
Laszlo's initial pricing of $10K/cpu left much to be desired also: http://www.infoworld.com/article/03/09/19/37TClaszlo_1.html
Then they dropped the price to $2K/cpu: http://www.asvguy.com/2004/10/flex_wins_the_f.html
Now, it's free.
The point is that the pricing will always come down, particularly when the competition heats up. It's currently happening with the Flex pricing model.
Additionally, I think most developers pay way too much attention to product/tool pricing. A tool is not necessarily better just because it is free and/or Open Source. The old adage, "You get what you pay for" applies to software also. IMHO, what matters most is which product will give me the least amount of headaches in development, and provide my client with the best UI-experience.
When evaluating Flex, and Laszlo from those perspectives, I think Flex wins hands down.
This argument is fine if your company can afford it. If you don't have a lot of money, it always comes down to price vs. performance. And if you are in charge of producing a product with very little resources, you have to look at the cost.
Sorry for sounding a bit <too> enthusiastic, In retrospect it did sound a bit like marketing-speil :) I'm not an employee, just slightly biased as I have used the product in question and my domain is integrating Java Desktop / RIA technology into large scale enterprise applications :)
I've spent many years developing enterprise desktop and web applications, six of those as a Java Swing developer and this is where Flex has been perfectly placed IMO. The programming model makes the switch very easy for Swing developers.
Yes, the orginal pricing policy was not cheap but still arguably good value for money when compared to project savings and out-of-box enterprise connectivity. After all, do corporations routinely question the cost of Oracle, Websphere, Vignette, Seibel ? - actually in some instances they might but in general corporations pay a premium for a quality product with support.
I agree, Laszlo is a very strong product indeed, uses the Flash VM as its rendering engine as does Flex, and is backed by some of the big players. Remoting connectivity is good too however it is some way off in competing with Flex in the enterprise arena especially when you look at Flex 2's data sync'ing and messaging technology.
Undoubtably AJAX has the momentum, the coverage, the books and crucially, the hype. It's a great step forward to providing a more engaging experience. This is a technology that will ultimately stick around but is still stop-gap in terms of where Web Applications will be heading. It is also capped in its capabilities.
Adobe/Macromedia will have a new pricing policy for Flex 2. For companies wanting the bells and whistles, messaging facilities, data sync'ing and disconnected client capabilities then they will probably pay a similar license fee as before for the new Flex Enterprise Services product.
Where things get interesting is that Flex 2 Presentation services (MXML rendering engine, compiler, eclipse UI composer plugin etc) could well come in at an extremely low price, possibly gratis for most elements. They will be essentially aiming for mainstream adoption for certain types of RIA applications with this strategy. How successful they are remains to be seen but lets not underestimate the forces at play - Adobe/Macromedia now has a combined market cap of $19.37B!
Anyway, one things's for sure '06 is going to be a great year for UI engineers. AJAX or XML UI markup, just choose your flavour and lets start pushing the envelope. Or enjoy the best of both worlds - eg) Google Analytics
To see who's winning the battle right now :)
Laszlo vs Flex
AJAX vs Flex
Whilst AJAX obviously has the advantage of running without a plug-in, the implementations are still nowhere near as easy to design and develop and the end result is generally an approximation of what Rich Internet Applications are all about (usual suspects aside eg googlemaps).
Undoubtably AJAX has the momentum, the coverage, the books and crucially, the hype. It's a great step forward to providing a more engaging experience. This is a technology that will ultimately stick around but is still stop-gap in terms of where Web Applications will be heading.
Looks like you know where web applications will be heading and what RIA is about. Do you mind to share your vision?
If I were tasked to developing an application that was limited to html (which I believe is a legitimate requirement in some cases), sure, I would (and have) most definately leverage AJAX.
In cases where there's more flexability (no pun intended) with regards to runtime environemnt, it would be Flex hands down. In fact most of our internal systems are now flex-based with J2EE/SOA backends. All I can suggest is try both and make your own comparisons...you may be surprised.
Only complete idiots would use this for an entire J2EE application.
I agree that AJAX is good at upgrading current Web-applications to a higher level of user-interaction... But they are still web apps.
Flex 2.0 leverage enterprises applications to a stellar level (try it on a real, pro-level application, you'll see what I mean), and JEE is then a natural backend, oppposite to AJAX unstandardised-scripting concept.
Who are idiots then ?
My issues with Flash:
- Last time I checked Flash did not have concept of shared libraries;
- every flash works inside its own Flash VM or whatever its called, and about 10 flashes on page tend to slow down P4 3GHz machine significantly;
- most of Flash UI-s do not use liquid layout and do not scale, which makes them unusable on high resolution displays;
I would say that Java Applets (or JWS ) should be used to build applications, not Flash.
There are few minor changes are necessary to make Java Appets attractive again IMO:
-applets should be truly supported by Java Web Start;
-JNLP should allow multiple signatures on jar files and libraries from other sites (repositories);
Having the features in place will make Applets very fast to start by removing need to download common libraries on user machine over and over again (log4j, hessian, etc).
Flex and Flash evolves:
->It has the concept of Runtime Shared Libraries (RSL), even Flex 1.5 on Flash Player 7.0 has that concept
->The new Flash Player 8.5 has JIT compiler features (http://labs.macromedia.com/wiki/index.php/Flash_Player:new)
->Java is 10 years old, it all started with Applets, but unfortunately applets were never really adopted since then
->The REAL competitor for Flex 2.0 on the Enterprise Applications market will be Microsoft Avalon Framework, not AJAX... But so far Adobe is many years ahead in terms of adoption and openess.
Java is 10 years old, it all started with Applets, but unfortunately applets were never really adopted since then-Wheel is believed to be 5500 years old, so what?
In IT 10 years seems to be the time for developers at large to understand ideas and start using them productively.
And let me repeat that Applets lack very important features which prevented their wide adoption back then and still handicap Applets now (although high speed connections ease the download pain a bit).
Last time I checked Flash did not have concept of shared libraries;
It does not. Both shared libraries and SWCs (which are similar to DLLs)
>- most of Flash UI-s do not use liquid layout and do not scale, which makes them unusable on high resolution displays;
The layout is up to the developer. The Flex framework provides a fluid layout manager (among others) that developers can leverage.
hope that helps...
mesh at adobe dot com
>- most of Flash UI-s do not use liquid layout and do not scale, which makes them unusable on high resolution displays;The layout is up to the developer. The Flex framework provides a fluid layout manager (among others) that developers can leverage
The tool seems to discourage use of liquid layouts ( same complaint about DreamWeaver, it can do liquids but defaults are fixed). Could you point me to a site where I could see real Falsh application uses liquid layout?
On components: Could you point me to a guide that describes how to create own widget for Flex/Flash and reuse it across multiple Flash applications?
Could you point me to a site where I could see real Falsh application uses liquid layout?For example, www.laszlomail.com If you like OE, you might like this one too.
For example, www.laszlomail.com If you like OE, you might like this one too.Impressive indeed. Liquid layout works beautifully.
However still unusable IMO because I cannot increase font size, so it is all way too small :(
Also there are no shortcuts for keyboard operation.
For example, www.laszlomail.com If you like OE, you might like this one too.Impressive indeed. Liquid layout works beautifully. However still unusable IMO because I cannot increase font size, so it is all way too small :(Also there are no shortcuts for keyboard operation.
Now, after I have cooled off, these are the gripes with Flash-based iterface:
* older engines are slow, slow, slow. Have not tried 8.5 though.
* It runs in a browser window, but I yet to find a Flash app that has hooks to font size events.
* The L&F is different from my OS, but this is not a big deal for most people.
* Keyboard support (at least in Laszlo) can be improved.
* What about multithreading? Honestly, I don't know is it possible.
* Each time I leave the browser window and then navigate to it back, the app is reloaded. Can I store it on my machine?
Now some less-technical issues:
* I don't like windows (note the small case). I hate dragging them around and arranging them. Web apps provide "flat" interface and I like it. Flex/Laszlo reintroduce windowing interface again. What for?
* Window manager inside a browser window? Which is itself a window inside another window manager? This is too icky for me.
I would like Flash apps better, if they did not run inside browser window. How about standalone Flash runner? I believe there is one, but it is not for free, is it?
Otherwise, Flash is not that bad. You have one programming model for all your clients. Something to think about while MS is polishing Avalon stuff.
By the way while some Adobe/Macromedia reps are here: the yellow self-service booths at Burger King (or is it Carl's Junior?), are these machines Flash-based or do they run something else? Look very much Flash-y. If they are not, then such booths is a great market for Flash. ATMs too.
older engines are slow, slow, slow. Have not tried 8.5 though.
Whilst I haven't performed any personal tests, initial 8.5 benchmarks on the web come in at 12-20X faster for some operations. This is causing a lot of excitement right now as it up opens up many new possibilties for realtime apps.
>> It runs in a browser window, but I yet to find a Flash app that has hooks to font size events.
Flex fonts can be controlled externaly via CSS or programatically from within the app so there is no technical reason why you cannot change font sizes on the fly. However, I have not seen an api that hooks directly into the browsers font size event so the sizing may have to be implemented as a bespoke application feature.
>> The L&F is different from my OS, but this is not a big deal for most people.
Agreed, but I think this was a core design feature. Flex has its its own theme & skin management system with the default theme being Halo.
The following examples shows programatic skinning and OS specific themes.
>> Keyboard support (at least in Laszlo) can be improved.
Agreed. Not sure about Laszlo but keyboard support is generally good in Flex however it does seem to be missing shortcuts/mnenomics. All other accesability features are present and correct - tabbing, default keys etc.
>> What about multithreading? Honestly, I don't know is it possible.
The Flex ui programming model is for most parts totally asynchronous. Components / properties can be bound to events or datasources, xml streams etc so traditional multithreaded apps such as financial stock tickers, live data graphing (using the flex charting components) is entirely possible.
http://www.igindex.co.uk/ shows live market feeds via the Flash VM (flash not flex)
>> Each time I leave the browser window and then navigate to it back, the app is reloaded. Can I store it on my machine?
A flex app, dependant on your browsers cache settings, should be cached so it will be stored on your machine unless it has changed at the server end in which case it will be downloaded again.
> I don't like windows (note the small case). I hate dragging them around and arranging them. Web apps provide "flat" interface and I like it. Flex/Laszlo reintroduce windowing interface again. What for?
Windowing or MDI (Multi Document Interface) is a valid design principle but it should be used only in certain circumstances. In most cases I agree that a "flat" interface with minimal popups is more desirable from a usability perspective. Many RIA examples that show draggable, resizable, collapsing windows offer no usability benefit other than the fact that it is demonstrating that that feature is technically feasable.
> Window manager inside a browser window? Which is itself a window inside another window manager? This is too icky for me.
Agreed - generally poor design leads to this.
>> I would like Flash apps better, if they did not run inside browser window. How about standalone Flash runner? I believe there is one, but it is not for free, is it?
This is a valid point. Sometimes you need to have fine control over the experience without browser clutter.
You could roll your own desktop Flex/Flash app runner using JDIC (Swing embeded IE/Firefox browser). I've done this for one of my projects and the code is very minimal indeed. That way you will have finer control and remove all the unwanted browser functionality. What you are left with is essentially a RIA viewer.
>> Otherwise, Flash is not that bad. You have one programming model for all your clients. Something to think about while MS is polishing Avalon stuff.
I agree. Even as a devoted Java UI (Swing/Web) developer I'm a convert now and genuinely believe that Flex, whilst not a perfect fit for all applications, is best of breed technology for thin client Enterprise UI. Avalon will undoubtedly be a strong contender. I've recently checked out the Sparkle product and it looks very promising but still seems to lack the standards approach (XML markup aside) that Adobe have adopted. Flex also sits very nicely inside your J2EE stack and hence is easily integrated into the web container.
>> If they are not, then such booths is a great market for Flash. ATMs too.
Hope this helps.
Please contact me if you would be willing to discuss Flex in the Enterprise. We have a financial system that we are having developed and I would like to discuss technology / topology / etc with someone experienced in Flex Development for Enterprise Applications.
FSAdvisors (Future System Advisors)
Baton Rouge LA
Does anybody have any comparison of OpenLaszlo solutions with Yahoo! widgets (http://widgets.yahoo.com) -Bhaskara
Flex is for RIAs, with a rich widget set and functionality that AJAX monkeys can only dream of. It's not for demos and annoying ads.
I also have concerns with flex 2.0 based on the flash player 8.5 requirement...I haven't seen if it will be compatible with older versions.
The 8.5 VM will actually consist of two runtimes (from my understanding) so backward compatibility will not be an issue.
OpenLaszlo is pretty nice and would be even better once the tool support matures.
I advise downloading Flex Builder (which is a Eclipse plugin-now...nice) and playing with it. I was impressed.
The great thing about Flex and other Flash-based toolkit/frameworks is that you don't have to deal with all the browser incompatibility nonsense that plagues AJAX. In any case, I think Flash and AJAX are complimentary approaches.
Just to clarify some of the questions on pricing for Flex:
Flex SDK - Command line compiler, Flex Framework, docs, debug player. FREE with no limitations. You can create complete Flex Framework based applications just using this.
Flex Builder 2.0 - Flex / ActionScript Development IDE built on top of Eclipse. Pricing not announced yet, but will be under $1000. This is not required for Flex development, but like any good IDE, will make development easier.
Flex Enterprise Services - provides real time data services (push and pull), and tons of other features. Pricing scales from free (limited connection) to much higher (havent finalized exact details).
You can download public betas of all of these at:
Mike, just wondering, has there ever been any thought at java integration in the runtime? May sound silly, but personally, this is the one thing I am looking for, as there are tons of things I'd like to be able to do, but can't strictly within the flash environment. It looks like, without actually reading the docs to enterprise services which I just downloaded ;-), that the pub/scribe facilities will be proxied via jms connections on the appserver, although obviously I could be jumping to conclusions. It would be nice, even as an intermediate solution for some of us that are heavily vested in Java Swing UI's, to be able to more easily call into java code running along the flash runtime, so we could, for example, reuse logic code already written, or directly access remote facilities such as jms directly.
Mike, just wondering, has there ever been any thought at java integration in the runtime?
Yes. This is something that is oft requested, although currently there is no way to do it natively.
Hope that helps....
mesh at adobe dot com
Mike, thanks for the info. It looks like Macromedia/Adobe is doing all the right moves with lowering the barriers of entry as AJAX is hyped to no end.
Anyone who invests some time with the beta rather than spouting off uninformed opinions of a technology they haven't used will agree the publish/subscribe model mentioned in this message is superior--both for development and for user experience--to the RPC-style model required in AJAX as well as previous versions of Flex. After building an app with the Flex Enterprise Services beta, I would not be able to go back to AJAX without a sense of loss. Give it a try, it's cool.
I've been using OpenLaszlo pretty extensively for the past six months or so, and I've been able to do things with it that I honestly don't think I could have accomplished with AJAX in a reasonable amount of time & cost.
When I began my current project, Flex was immature compared to Laszlo, but it's made great strides. I still think the object model and XML data binding is better in Laszlo, but the XML manipulation (using e4x) is much better than the Laszlo DOM-ish like classes. The UI components are similar and coding styles and time are similar. I'd perhaps give the edge to the Flex developer tools, and of course, the price (free) to Laszlo.
Combined with a REST back-end (mine written in Java) the architecture is easily understood and straightforward.
There is absolutely a learning curve with both Laszlo and Flex - you can't leverage any of your existing web skills for the UI... but I think it's really worth it.
For the "easy" parts of the website, using standard web technologies are better served, but for really complex UI's - you really can't compete with a Flash based application. AJAX really can't hold a candle to the ease of development and results.
With either Flex or Laszlo, you can compile the application in standalone mode so that it's just a regular Flash .swf file. This can be deployed inside your war like anything else. The Flash runtime is capable of making asynchronous calls to the browser which returns XML data. This data can be bound to your UI components. We actually manipulate this XML data and send it back to the server. (full CRUD capabilities) There are no additional runtime requirements other than a Flash client on the browser. It's integrated smoothly into an otherwise Struts/JSP application.
This technology should really only be used for difficult web UI's... that's the sweet spot... if it's not really a RIA.. don't use it.
I've been reading about the upcoming Flex 2.0 and to me, what seems to be the major drool factor is the fact that you can now publish and deploy swf files without requiring the Flex Server:))
Admittedly it could be done earlier, but it's the little things that go a long way in making the life of the ordinary developer (aka guys like me!) easier... For reference on how it could be done earlier, visit these 2 wonderful and descriptive links:
Finally someone at Macromedia (now Adobe) woke up to the simple fact that people don't want an pricey server to deploy swf files. The command line compiler will now, apparantly, not include the 'secret code' that required the flex flash file to be deployed on the server. Now, simply bundle up your swf into a WAR/EAR and resend (resell) as many times as you want to. A lot of companies' pricing models could see an impact, especially small & medium offerings. Apps that looked cheaper in HTML will now hopefully be looked upon with the Flex angle, which, combined with the already incredible interface that Flash/Flex provides, could see some interesting, and hitherto ignored, solutions come out in the market, thereby increasing adoption.
Alternative approaches like OpenFlash/Laszlo are there,and will continue to be, but IMHO, corporations still like to see an enterprise solution on which they could base their decisions.
My 2 cents...