Is Flex/AIR a dying technology?

[RandomReado]

I develop Flex/AIR applications and need to know if this is a dying technology and whether there's any point continuing to develop on this platform?

[robglobal]

For the mobile world, sure. That is about 7% of the browser market, according to a seminar I heard at Ga Tech a few weeks ago. But as long as people use PC's, Flex/AIR will be supported. There are some enterprise applications out there that are using Flex, entrenched in it, such as SAP and Flex. So the process of phasing out Flex, if it were to be, would be a couple years to come at minimum.

I remember when Flex started getting big 4-5 years ago, and people were saying JavaScript would be dead in 1-2 years. Look where we are now.

Personally I love coding in Flex. But I am ramping up on HTML5 next year.

[RandomReado]

Just a few questions...

So is HTML5 aimed primarily at providing rich content for mobile browsers?

Will HTML5 replace desktop applications, such as those created in Flex & AIR?

Is it possible to run a HTML5 application without a browser, like you can do with AIR?

When Adobe say the Flex SDK is being donated to another organisation, does this mean the SDK will advance quicker than it did at Adobe?

Can we expect to see a Flex 5 in the near future?

[robglobal]

There are a couple threads recently out that answer most of these questions you pose:

http://forums.adobe.com/thread/923213?tstart=30

The only Flex 5 we will see will be an open source effort. Adobe is going to release the Flex 4.6 sdk and then move it to open source environment.

[drkstr_1]

I really don't think Flex is going away anytime soon. I have worked with many technologies throughout my career, and the Flex technology stack is the best thing out there right now for deploying rich front-end apps to as many screens as possible.

The main problem I forsee is pushback from the decision makers (clients) because of the bad rep Flash has gotten over the past few days (and longer). When that happens (and it will), my suggestion is that you turn the conversation around so it isn't about "Flash" or "HTML5". Make the conversation about what kind of features you can offer, and what devices you can offer them on. Flex will win hands down in that conversation, because the reality is that HTML5 simply isn't ready to replace it.

That's just one man's opinion at least.

[Guest]

ColdFusion has allegedly been a "dying technology" for what, like a decade now?  Way too many people are way too eager to declare things "dead", when they have NO idea what they're talking about.

Why are people so afraid of abandoning the mobile garbage market?  Most of those toy applications don't need an SDK like Flex.  It's about time we stopped wasting resources on Flex mobile, to focus on where it matters.  That market probably should never have been targeted in the first place for Flash/Flex.  If Adobe leadership should be ashamed about anything, it should be getting caught in the "mobile is all that matters" hype and hysteria going around lately.  Flex/FB 4.5 (mobile) was a waste.

[UbuntuPenguin]

I don't think Flex is dying perse.  But what most people including me are scared of is the perception of Flex's viability and long term support.  For instance, I have a little pet project I am working on that I am hoping to pitch to some companies, now if I go into a room full of MBA's I risk being turned down because they have heard this and that about Flex.  The driving issue is that it becomes harder to advocate Flex based projects because of the perceived lack of support by Adobe discontinuing the allocation of resources to it.

As far as mobile development goes, that most likely came about in the Flex world because of the perception.  No company wants to get left behind with their technology and having a mobile branch of Flex just made it that much more competitive, at least in the beginning.  With the proliferation of tablets and smartphones, I can't say that was such a bad idea.

From my username you can tell I know my way around FOSS, but I have a bad feeling about Flex going FOSS.  Let's start with Java comparisons. Java is a language that generally has its roots in the backend, firmly in the realm of neckbeards.  It is also the backbone of many corporations' backends and technologies, if not for the language itself, then for the VM (JRuby, Scala, Closure,etc).  This leads to the platform having MASSIVE corporate support from tech giants such as IBM, Redhat and others who can see to it themselves that Java survives.  ( On a sidenote, think of the operations coverage Java has, from DBs, Persistence, Networking, Imaging, Fronted("yuck") and even client side("double yuck") ) .

Now lets look at Flex,  I don't think it has the massive corporate support as Java since it can solve every single problem in the software field so that may hurt it.  More importantly, like someone else pointed out, what about the VM it runs on, or the idea that Flex is opensource, but EventDispatcher is not ? 

I don't think Flex is dead perse, but its market value in terms of people wanting to do projects in it has taken a hit since people are unsure of its fate.  I can tell you this though, I think HTML5 isn't quite there yet ( 6 months - 1 year ), it doesn't seem to lend itself to the same workflow type as Flex/AS3.  The whole JS world and language seems to defy any OO rules at first glance, maybe this is where Adobe is going to come in, I don't know.

[rexdtripod]

Kind of sucks.  I just got my Flex mobile app working on the IPad.  Pretty cool app.  And it's essentially the same code base as my PC app.  Few optimizations.  But it's hard to hit all these platforms with one code base and have things present in a homogenous way.  I think Flex mobile is worth the effort.  Not sure where I stand with my app now.

[UbuntuPenguin]

...Not sure where I stand with my app now.

That's easy, you are going to learn to rewrite you app in HTML 5/JS.  Not that your app isn't viable right now, but since you know how the app is supposed to look and behave in Flex, it would be great to begin the journey in replicating similar behavior in HTML5/JS.

[rexdtripod]

UbuntuPenguin wrote:

...Not sure where I stand with my app now.

That's easy, you are going to learn to rewrite you app in HTML 5/JS.  Not that your app isn't viable right now, but since you know how the app is supposed to look and behave in Flex, it would be great to begin the journey in replicating similar behavior in HTML5/JS.

Would like to do that but seems like a longshot at this juncture.  Simple websites can be reproduced.  Not complex OO apps. 

I do presentation software.  My current app does live video, synchronized slides and live chat.  Most importantly, via FMS, I literally reach out during  a live show and remote control operate your screen - resize video, slides, polling components, change screen views - all in full res (no screen sharing) and all wrapped in beautiful animated transitions.  And I can literally single users out and send special commands to only them if need be.  And the kicker?  I do all of this homogenously from one well designed OO code base and hit every browser/OS/device including IPAD via packager.  HTML5/JS is just not even in the ball game for what I do. 

Websites - easy.  Complex remote control FMS apps - not gonna happen.  I don't know how that real time remote control thing is gonna work.

And to boot, with the browser wars well under way each is lining up behind different video codecs.  I have no idea how I'm gonna get at metadata in live streams through web pages on different browsers *in a homogenous way*.

Each manufacturer, browser and device, wants everyone to line up behind them.  When it comes down to it, Flash is one hell of an equalizer.  I think certain monopoly minded manufacturers who made careers out of being anti other monopoly minded manufacturers just can't stand that.  They just hate competition.

I say to them, the Flash haters - my app is outrageously cool.  For all of your bluster about Flash being a hog, and sucking - let me see you build my app in HTML5 and reach everyone with one code base. 

[cymbals]

That sounds very cool. Even a control-any-device subset of your flex app is pretty impressive.

rexdtripod wrote:

Would like to do that but seems like a longshot at this juncture.  Simple websites can be reproduced.  Not complex OO apps. 

I do presentation software.  My current app does live video, synchronized slides and live chat.  Most importantly, via FMS, I literally reach out during  a live show and remote control operate your screen - resize video, slides, polling components, change screen views - all in full res (no screen sharing) and all wrapped in beautiful animated transitions.  And I can literally single users out and send special commands to only them if need be.  And the kicker?  I do all of this homogenously from one well designed OO code base and hit every browser/OS/device including IPAD via packager.  HTML5/JS is just not even in the ball game for what I do. 

Websites - easy.  Complex remote control FMS apps - not gonna happen.  I don't know how that real time remote control thing is gonna work.

And to boot, with the browser wars well under way each is lining up behind different video codecs.  I have no idea how I'm gonna get at metadata in live streams through web pages on different browsers *in a homogenous way*.

Each manufacturer, browser and device, wants everyone to line up behind them.  When it comes down to it, Flash is one hell of an equalizer.  I think certain monopoly minded manufacturers who made careers out of being anti other monopoly minded manufacturers just can't stand that.  They just hate competition.

I say to them, the Flash haters - my app is outrageously cool.  For all of your bluster about Flash being a hog, and sucking - let me see you build my app in HTML5 and reach everyone with one code base. 

[DataProvider]

There's no difference at all. Your app is written in AIR. AIR is not touched by this change of direction. It's only the Flash animations that run on BROWSERS in the mobile space that are affected. In other words, the dancing penguins that advertise mortgage refinancing on the Yahoo pages, or what have you, will be gone from the browsers on mobile devices. Your AIR mobile apps don't run in browsers to start with.

[rexdtripod]

DataProvider wrote:

There's no difference at all. Your app is written in AIR. AIR is not touched by this change of direction. It's only the Flash animations that run on BROWSERS in the mobile space that are affected. In other words, the dancing penguins that advertise mortgage refinancing on the Yahoo pages, or what have you, will be gone from the browsers on mobile devices. Your AIR mobile apps don't run in browsers to start with.

Not quite.  There are two versions of my app.  One runs on PC's in the browser.  The other is an AIR/mobile app for the IPad.

The PC version runs "as is" in Droid tablets and phones. They will eventually cease to run there if Flash players are abandoned there.

As I mentioned in my post, client gave 400 IPads out to sales force.  Didn't want the hassle of asking all of them to install an app.  Just wanted a link to a browser app.  To you and I this may seem unreasonable, but to them, they run our PC app from just a link in a browser all the time.  They don't get the IPad distinction.  They're just business people.

[rolance2]

Flex will die if:

1. mobile (read: tablet) computing continues to replace desktops and laptops AND
2. tablets remain physically limited in terms of memory and speed AND
3. certain hardware manufacturers, who shall remain namless insist on controlling/gatekeeping their devices by limiting software access AND
4. HTML5+JS+CSS can catch up and get even halfway to where flex is in terms of functionality

Flex will not die if

1. tablet computing is here this christmas but gone before next christmas OR
2. tablets become physically capable of running relatively large software stacks like flash player without "missing a beat" OR
3. Android increases its marketshare over iOS and consumers go back to valuing the open-and-accesible over the closed-but-shiny OR
4. New open standards (HTML5) continue to develop at the rater they have over the past decade

I won't pretend to know what is going on, but from my vantage point, the form factor advantage of tablets is here to stay - I see them becoming ubiqutous. As for Item four, I can only scratch my head and conclude that someone at adobe knows a lot more about this than we do, because I don't see an open standard replacement for flex that has even half of its power coming anytime soon. I'll be happy to be proven wrong, but I think the cost of being in the development business just got WAY more expensive.

The money you might save by using open standard frameworks will be nothing compared to the money you will spend getting the code to work everywhere. I have my doubts that an open standard will ever be implemented consistently enough to rival what we have now with flex and flash player.

I don't think that the sky is falling, but I imagine the cost of staying dry is about to go through the roof.

[Guest]

I just completed my first mobile app in Flex. I'm comfortable with Flex and Java in concert and am in no hurry to move to HTML5. As long as I can write code in Flex quickly and that code compiles to every platform I work with (PC, Android/iOS), and my designer is building things in Flash and, and, and... I'll be using it.

[pauland]

Take a look here for a rather more expansive and positive explanation of Adobe's plans:

http://blogs.adobe.com/flex/2011/11/your-questions-about-flex.html

[RandomReado]

That's what I was missing - a post reassuring me in every way possible - thank you Adobe.

But what will happen about Flash Player and AIR development? Are these going to be open source too to allow them to keep up to speed with Flex SDK development?

[Guest]

According to the blog NO. Adobe will still develop the Flash Player. But they said that they will keep up with the speed.

Myself, I like it this way. I did not like them claiming Flex to be open source while they're the whole decision committee. Now, it's an open source for real.

Talking about mobile devices, I don't understand when people say that they just made a cool app. Adobe said they're not going to support anymore(i.e. no flash player upgrade). Your app should still be working unless you upgrade it in the future.

It blows I know. But most of the mobile device apps are real simple anyway. Some drag-and-drops, some big buttons, some feeds... There are not much of these data presentation and heavy interactive interfaces. At least as of now.

But Adobe should really keep it on desktop. Add more features that HTML5 will never match. Like Pixel Bender, Stage 3D. I would really love too if you accelerate the sound card, that allow us to play with Fruity Loops like but Flash.

[Karl_Sigiscar_1971]

Now, the best thing that can happen is for the community to create a roadmap for Flex 5 that builds on top of Flash Player 11 features. If some top Flex developers contribute and the project is well led, this could possibly work.

On the roadmap, I would personally add:

- All Spark (nothing to do with Transformers 😉 which mean migration of all MX components

- GPU accelerating rendering of all components, leveraging Stage 3D (the old trick of 2D in 3D as in the MacOS X's Quartz API)

- Video component that leverage Stage Video

- Time slicing

Eclipse and Firefox, after all, are open source too and they are top software. To survive, Flex needs to become a top open source technology.

The best Flex engineers out there should volunteer.

[Fletchgqc]

Right now perception is playing a crucial role. I reckon Adobe should put a good designer onto building a beautiful page/site for Flex under the Eclipse foundation, assuming they want to encourage success for the project under its new open source foundation. Sometimes a great website design makes a huge difference in how people evaluate the project, this could be a huge help at a small cost.

[rexdtripod]

mahatana wrote:

...

Talking about mobile devices, I don't understand when people say that they just made a cool app. Adobe said they're not going to support anymore(i.e. no flash player upgrade). Your app should still be working unless you upgrade it in the future.

It blows I know. But most of the mobile device apps are real simple anyway. Some drag-and-drops, some big buttons, some feeds... There are not much of these data presentation and heavy interactive interfaces. At least as of now.

Nope.  Like I described.  My interface is not simple.  It's highly interactive.  That's the point of the app.  And yes, we do plan to upgrade the app in the future.  Don't know of a software provider that doesn't.

The point is that Flash is doing things for me that nothing else can at this point.  There's a power to it - a power that doesn't exist in HTML5.  And building and maintaining individual strains of the app for each platform is simply not manageable.  Simply put, this is a regression.

Maybe something comparable will come along down the road some day, but those who say "just convert your app to HTML5" don't see or use Flash's power. 

Still hoping for the best.  Just have to keep trudging along and hoping for a positive development.

[Karl_Sigiscar_1971]

Just updated:

http://blogs.adobe.com/flex/2011/11/your-questions-about-flex.html#comment-6556

[Guest]

My app is also not simple. And, I agree that it's a regression if the Flex mobile ability to write once, compile to any platform goes away. However, from what I read, it seems this is now going to be opensource. If it's opensource, won't it continue to be developed and the support just shifts from Adobe to the community? I wrote in Java before it was opensource and I still write in it today now that it is opensource. Also, it would seem that Adobe does have plans to provide cross compile tools (the PR says to compile to HTML and JS). Is it sangine to believe that whatever the future holds, those of us who have already developed complicated cross platform mobile apps (in my case that do not run in the browser) will have a solution that does not require an entire re-write of existing code?

Given that Adobe says the following, it would seem that all is fine in the mobile app development world or am I wrong?

"Adobe will continue to support applications built with Flex, as well as all future versions of the SDK running in PC browsers with Adobe Flash Player and as mobile apps with Adobe AIR indefinitely on Apple iOS, Google Android and RIM BlackBerry Tablet OS."

[Yozef0]

For Crying out Loud!  Flex is Alive and Kicking! Stop Worrying - Jeez.  You'll need to learn HTML5 just to see how the Flex framework is superior - In Every Way!  AIR? You're kidding me right? It's on every Mobile & Desktop. Stop Nagging and Create Cool Apps!

[rexdtripod]

So the customer that just pulled a 25K event about two hours ago because I couldn't deliver it reliably to the IPAD browser is just nagging right? 

I described the app earlier in the thread.  Sounds like we have a volunteer.  Let's see you do it in HTML5.  I'll expect a post on it soon