Availability of Adobe AIR runtime in the long term?

[FerrieJB]

Hi all,

I'm currently maintaining an extensive portfolio of Flex (SDK 3.5) based products, that all run on browser based Flash Player plugins, and following the announcement of Google Chrome hiding Flash Player behind a click-to-start mechanism, I am reviewing what target runtime would be the best for our products.

I understand that Adobe still officially supports and develops AIR, but would there be any clarity on the long term (>5 years from now) roadmap of this product ?

Would converting my Flex code to run in AIR be a safe bet to ensure future compatibility on the platforms in runs on?

I've also been looking at alternative runtimes. There are some initiatives that try to compile MXML to HTML5 compatible CSS/JS - FlexJS, but those don't support stuff like AMF or E4X which make it useless in our situation. JavaFX could be another option but that would mean rebuilding our entire codebase from scratch.

We have about 10 years of development in Flex source code in our repositories and it seems like such a waste to have to rebuild it all. What are my options, and what guarantees are available, to ensure the availability of our products' runtimes in the future?

I presume there must be many devs facing this problem, and I was wondering if there will be a formal path from Adobe to follow (like f.i. an Adobe supported product that compiles MXML into HTML5 properly, or a guarantee that AIR will be available for at least certain period of time). Any suggestions or helpful links are welcome.

Kind regards,

Ferrie

[rewb0rn]

Hi Ferrie,

yes we are in the same boat. We are developing web and mobile games with Flash and Air and with the latest developments we are currently investigating if it is time to start over in another technology. However it is such a waste of resources to change the horse and I assume it would be much cheaper for everyone just to fund Adobe or someone else to seriously rethink Flash and Air and invest into its future. I seriously do not understand why Adobe does not try to monetize it better, I assume money would solve all of our problems.

Cheers

[Applauz78]

I would happily pay for this.. IMO it's wayyy better developing in AIR as I can target iOS and Android in 1 code base. .. Too expensive to do this as native development...  and terrible doing it in HTML apps.

[KarlCfs]

I agree. I bet we are a lot of developers that have many years professional work put into AS3 that would happily pay a monthly fee to make sure the technology stays relevant in the long run. Adobe please, take our money!

[ASWC]

All my Flash web based games (and code base) have been converted to typescript and run on canvas. I developed a specific tool to make the conversion (one month work). Now I can make a new Flash game, click a button and get my typescript project, compile and get my HTML5 game running immediately. My point is that this is doable depending on how your code base was build to start with.

[Colin Holgate]

Is what you describe a way that you can take a custom FLA TypeScript project and export for HTML5, or are you saying that you take your existing AS3 projects, magically convert them to TypeScript, and then export to HTML5? The majority of people who have done Flash work will have used ActionScript.

[buzzy@rhythmbee.com]

I too am using AS3 in Animate CC.

[ASWC]

No my tool converts the code from actionscript3 to Typescript and that's all it does. Because the assets and asset management is done via a ContextResource framework (also internally developed) that framework also converts perfectly to Typescript (same assets are used in both AS3 and Typescript). The conversion is direct and project compiles immediately in Typescript and runs flawlessly in browser, both output (flash/html5) are visually indistinguishable from each other.

[DaveCates]

Just want to say I agree with all of this.

But also...we need some major PR effort to show off the platform and it's significant potential vs others like Unity etc.

We need this PR because soon it will be an absolute nightmare to recruit developers who are skilled enough to develop with AS3 and Air. That would be an absolute show stopper...and avoidable with the right evangelism and PR.

I'm up for getting stuck in and helping out but would need resourcing...

[rewb0rn]

I fully agree, PR (as in credible commitment by Adobe) is the one thing Flash and Air is missing. Chris and the team are doing a great job, but they need to be backed by the company more. Other than that it leaves little to wish for.

I do not understand why Adobe is so shy about the platform. If you look at Unity, they make 125$ per month per developer now. I am not saying I am happy to throw the money out of the window but for sure we are willing to pay more if that fuels the future for the platform.

[leop66344762]

Hi,

I also agree that we need more PR and more credibility. As of now, I find it hard to find resources, hard to find credible information, majority of PR, tech announcements, evangelist posts date as back as 2010->2013...

If you show that to any prospective customer and say: look what its possible... the first thing everyone replies is: but this dates back to 2010! And then.... we have to overcome a big concern that the technology is dead and no one uses it anymore (Thanks to the bad PR against it).

I know this has been discussed 100x, and I also know that what we can do we AIR/Flash since its inception is still way ahead of several other offerings... But, final customers don't know this. And I hear over and over from my customers: We need to do it as a Native app. And in the end, it starts to make sense. I don't know about you, but I don't do games. I do business apps. And my customers expect UI that works exactly like native, looks native and perform like native. It is a big issue to recreate the wheel (and keep it spinning).

To be honest, if there is no effort to bring it back to the top, it will fade away.

I wish someone from Adobe would comment on this, publicly, because I asked it privately and did not get any answers.

[davidrenelt]

I've moved on to Phonegap/Cordova .. not by choice, not at all .. but it is a painful but very obvious truth that there is no more momentum behind Adobe AIR, and a lot of momentum behind Phonegap/Cordova for crossplattform mobile development.

The reason why it is painful is because the flash rendering engine is a technological marvel that still has no equal. You could throw impossible things at it, and it just did it, while WebGL/Canvas turns into a performance nightmare every "wrong" step and forces you to sacrifice ideas simply because it would kill the performance.

So it's a very painful and tearful goodbye. I've developed a lot of fun things with Flash/Air the past 15 years, it has been a great ride, but when the love turns one sided, it is time to move on.

[Colin Holgate]

AIR gets updated every couple of weeks. Animate has already had updates, and I would expect more soon. I don't follow PhoneGap, does it get updates? I don't really need it though, I can already do what PhoneGap does using Animate. With that I can create the HTML5 content and also the web view app wrapper.

[leop66344762]

Yes, it is real that AIR gets updated every couple of weeks (sometimes more often).

And I know the team is pushing hard.

But this is what I said about PR. We developers know all of this.

But for the end customer, all they hear is "Flash is dead", "Google stops Flash player by default now", "Native apps are faster"... What I mean is, there is a lot of publishing going on with "the bad" of Flash, and little about its marvels.

I think everyone would really benefit if there were more publicity regarding the platform as a whole, and Adobe's long term plans regarding it.

[davidrenelt]

To be clear, Flash/AS3/Air is the platform that is discussed here. What Adobe has done with Animate is part of the problem.

I dont even think that it is a PR problem, i think the Management gave up on the platform a while back and is actively pushing towards HTML5/Javascript. All former advocates for the platform moved on to other things (ByteArray.org, JacksonDunstan.com etc), all AS3 Libraries / CodeExamples you find havnt been updated for years. Point me to some active AS3 site/community to prove me wrong .. i would love to be wrong ..

One guy that is still somwhat holding out: Yeah, But Is It Flash? But even he doesn't seem to put much trust in AIR as a plattform anymore. On the business end for me it doesn't matter much.. people are amazed and happy at what you can do in Canvas/WebGL ... not knowing that it's still a lot less then what Flash/Air had to offer. But i know. That's what makes me bitter sometimes.

As a seasoned developer (i'm doing this for more then 20 years now) i've seen a lot of things come and go. But somehow this one bugs me a lot more then past events. I think i really truly loved flash.

[Julien26]

I disagree with you, there's still some life in the AS3 / Air Community. For example you have Starling that released its version 3 just 2 months ago, and Feathers that released the version 2 also 2 months ago.

There's also the developer of as3lang.org who's realy putting some work for the Actionscript langugae. Etc. and Adobe is still releasing new versions of Adobe Air, faster than ever !

Yes, there are a lot of code that's really old on the net, but that code still works. There's still a loooot of libraries that aren't heavily (or at all) maintained. But you have that for all popular languages, people come and go, that's  normal. It's exactly the same for java for example. The problem here is that it seems that more people go than come.

But it's a shame, because Adobe Air is really really really a fantastic technology. Our app uses it, and I can do some things that I really couldn't do without a looooot more work on native code (and that are not possible with HTML5).

I don't think that Adobe has abandonned Adobe Air. It's still there, and I hope (and I need :-)) that it will stay there a lot more years. But I think the development team should be way bigger (if you take a look at the opened issues on the bugbase, you clearly see that they can't follow). They're doing what they can with what they have, and I'm thankful to them for that, but if they were more, that would be even better.

Now, I think that there are two main problems. First, you have ActionScript developers (like you) that switch to other technologies because they are not sure of the future of Adobe Air. That's why we lose some important advocates.

Then, a lot of people think that Adobe Air is Adobe Flash, and that it's dead. I can't count the number of times I had to explain to other developers that NO, it's not the same thing, and it's certainly not dead. People that could benefit from Adobe Air in their work, but they heard that it's soon-to-be-dead so they don't look further.

If Adobe went ahead, made a loooot of public relation to let the world know that it's still a thing, more powerful than ever, it could get a lot more users !

But for now they send misleading information, like their conferences where Adobe Air is not mentioned at all

Same thing for Flash Builder. Yes Animate CC is great for animators, but it sucks for (Air) developers compared to other tools like IntelliJ IDEA. Flash Builder was way better, but listing it in the "Previous versions" in Create Cloud don't really send the right message...

I know that IntelliJ IDEA supports ActionScript, but to my knowledge you can't use it with a projet that uses a .fla file, and it's my case

TLDR : Adobe Air is not dead, it really shouldn't die, and Adobe needs to acknowledge that publicly so developers can trust it again.

[leop66344762]

Julien,

I agree with several of your points. Most importantly, that Flash Builder has been left to dust, when its really a very solid IDE. Try doing in FlashDevelop what FB can do... honestly, I was never able to do so (complex multi-project referencing and compiling...I might not be the smart guy around...but.. in FB its very easy)

Interesting enough, today I watched a presentation from a google engineer regarding Web Assembly... go figure, the guy bashes plugins (read flash), but advocates for a Binary container that runs compiled code and claims the best feature is a Multi-threaded environment with shared memory space. And is runs inside the browser, in the same "layer" as the DOM/JS.

Wait a minute, since Flash 11.4, we've had workers, and they can share a byte array, and the code is compiled, and it is distributed in binary format, and it runs in a sand-boxed environment, and it can talk to JS and the list goes on. Yet, this is received with applause as a new glorious discovery.

Now.. tell me, we are going in circles. It just took them something like 7 years to get to a stage where they are showing a non impressive ray traced sphere moving up and down at  50fps ...claiming it is about 2x faster that pure JS. What is not¿?

To me, there is a very simple agenda:

Flash was/is too powerful. One can do exactly the same games that you sell on the stores, and distribute online. But if one distributes online, there goes the 25%-30% chunks for the Store owners... So, NO..Flash cannot run on mobile, else it would destroy the app stores.

I know I often deviate from AIR to Flash, and the sole reason is because to me, one fuels the other. The VM is in grand part shared, and the whole idea was that you could run the same code on web, devices, pcs, tvs... It is of great benefit for everyone to just ensure both live a very long life. We need to remember, while Mobile is all about Apps (AIR), Business apps are more about Web... and serious Apps can be done inside Flash and can perform way better than JS counterparts. Installing Apps is still a friction point on the desktop (like it or not, quite a lot of things cannot be done on mobile/tablets).

Since the very first Apollo open beta I have used AIR. I've enjoyed every step of the journey since then, and I truly hope it keeps up and moving forward.

[davidrenelt]

I totally agree that it shouldn't die. And yet, I think it already has.A platform is dependent on the enthusiasm of it advocates. JavaScript/Webtech has a lot of those now. It is a mess, to be honest, but there is excitement about it, so it spreads. All the big players are pushing into that direction, including Adobe.

But ... again .. i'd love to be wrong. I have a lot of apps i have to maintain in the AIR Ecosystem. So i'm very thankful for every year that it's still around and updated for newer OS releases. I think we made our case .. it's up to the OP to decide.

[rewb0rn]

Maybe we just need to hold out up to the point where all enthusiasm about HTML5 has faded away and people have lost enough money in the process. I am not sure that will happen, though.

[ASWC]

Starling and feather have never been professional frameworks, the 2 authors (payed by Adobe) have decided to target beginners instead of professionals for their audience forcing people like me to develop their own (the performance of Starling is just awful and without mentioning its design).

The Flash community has never been known for its willingness to support open source, I used to be for years a very active open source contributor (and actionscript.org super moderator) with framework, apps, tools, etc ... But even with thousands of people using my stuff constantly at the most I got less than a dozen click the "donate" button and this along a period of 6 years. I had a SoundManager framework, the best you could ever find out there, downloaded 80k times and got 3 donations for it totaling $7 ....

Flash people just don't want to pay a cent for anything and this is the main reason so many talented developers stop providing/helping with open source projects.

The truth is, it's hard to find out there a community that is LESS whiling to support open source. The Flash community is cheap as hell and they got the open source projects and the future they deserve.

[buzzy@rhythmbee.com]

Still working on the whether I can believe in Adobe AIR - and I want to.

Do you get any negativity from customers who have to give the application permission to access everything on their computer? That is a scary dialog box.

[rewb0rn]

We are deploying with captive runtime, i.e. .exe file. No scary dialog box there

[buzzy@rhythmbee.com]

Is that the AIRI file?

[ASWC]

AIRI is an unsigned (no certificate) AIR file, as such it cannot be deployed as executable on a computer.

NewB0rn is referring to native app compilation which requires captive runtime and export with native app extension (eg. .exe), this also allows for deeper access to OS features (eg. NativeProcess)

[rewb0rn]

Hi,

ASWAG you are correct, in our case there is still the standard permission dialog that warns the user about an application requiring access to the system. But it is not a custom AIR permission box, that no one has ever seen before, but the default Windows dialog, that is well known with the users. In our case we have even acquired a certificate so that it states the origin of the file in the dialog (this is also possible with AIR, but the problem remains, that the user has never seen the AIR dialog before with any other application).