Termination of financial support starling and feathers

[денисг22484613]

Guys, apparently the AIR is no longer an Adobe priority. Tell the developers when you are planning to end AIR support ?

[ASWC]

The "conversion" from AS3 to JS is limited and will stay very very limited. Conversion is not the end game of Animate and people shouldn't dream about an AIR/Flash to HTML seamless conversion in Animate, it won't happen for obvious technological reasons. Animate has become a very weird Software that creates HTML/Projectors/Flash/AIR/Slide Shows/Presentations/Videos .... etc ... At some point it's gonna get a make over and some options are gonna disappear like likely Flash/AIR/Projectors etc and eventually be only HTML software.

[PippoApps]

Of course it could be better, but I am pretty happy with Adobe's support to AIR. I do pay for a CC account since I use Animate and the whole suite, but don't forget you can leverage the whole AIR sdk without spending a penny.

On another note, AIR gets updated very often, and targeting 4 different OS they do a pretty good job.

I.e. I reported 2 weeks ago a nasty bug on iOS and today I have been contacted by the AIR team telling me that they are working on a fix. I couldn't expect more.

I only wish they continue to support it at the same pace. Of course I am so comfortable with AS3 and with AIR that I really hope they are not going to dump the technology. It is such a powerful solution, but it is really not known enough, that would be fantastic if they'd do some marketing around it. But since they don't and they will likely not do, I just accept things how they are.

I just want to give my thanks to the AIR team because they do a great job.

[PippoApps]

Anyway, last month I filed 2 AIR bugs. AIR staff contacted me for both. On the most serious one they are working on a fix on next release, for the second one they offered a solution and asked me to check. All within 2 weeks. Not bad at all!

It could be better of course, but I am really happy about their response.

Ad maiora. Le keep our hopes.

A warm hug to all fellow AIR lovers.

[DQvsRA]

Once again, it's nonsense that multi-billion company can't afford to support 10-15 engineers team, it's a drop in an ocean, they have millions of CC subscriptions. I believe that smart people who respect technology community will keep AIR alive and find the way how to evolve it.

[Leo Kanel]

The problem is not they can't afford a small team. I guess the current team something like 5-10 people somewhere cheap (like india). The problem is they don't want to. They failed to make AIR a paid platform, like so many respected game engines, so they are delaying the inevitable until they figure out how it is going to play.

I would love to see AIR open source, like they did with Flex. The Flex community managed to complete the AS3 to HTML5/JS compiler in 4-5 years.

[hferreira80]

"The Flex community managed to complete the AS3 to HTML5/JS compiler in 4-5 years."

Where do you saw that ?

Are you talking about Royale ?

If so, it's still a alpha/beta product.

In a near future, perhaps.

[Leo Kanel]

Yes I am talking about Royale. It is in beta quality since 0.9 and I know a few companies use it for released products. I am not saying it is perfect, I am just pointing out that this was received as pre-alpha and it was developed with the help of the awesome community.

This might be an option for AIR in the future. Flutter and React Native, are also open source, so anyone solves bugs.

[hferreira80]

"Flutter and React Native, are also open source, so anyone solves bugs"

Roayle (ex FlexJS) is taking a long time (even so would be nice to have it one day)

Meanwhile: Xamarin and the recent Flutter seems promising.

Soon I will let my hand on it and perhaps can forget AIR on mobile devices.

[Leo Kanel]

My team and I, already started with flutter. We are using it for mobile right now, but with the desktop option already available and web coming up, I think we will have a powerful tool soon.

Concerning Roayle, have a look at the awesome Tour de Jewel Component Explorer

[ASWC]

Xamarin is totally free so in your case "soon" really means whenever you want.

All AS3/JS converter ideas will always have extreme limitation and limited use cases. The 2 techs are way too different, even if a 100% conversion was possible (it's not) it would probably come at heavy price like performance and memory management problems, browser compatibility problems, etc ... There will not be a conversion miracle from AS3 to JS, you can hope for a fairly accurate conversion that gives you a BASE to start working on but a conversion to JS to production is not gonna happen.

[hferreira80]

The magic button it's not the solution (if will be here one day).

Also, I always been against HTML for Mobile Apps. HTML it's for Web Sites, period.

Facebook moved from HTML5 to native development in the past (just one of the biggest HTML5 defender).

And now Flutter directly from Google.

I think that's we don't need more proofs.

Hybrid/Almost native won the war (fortunately).

Adobe had the best technology (in the past): AIR.

Adobe had a great advantage but lacked vision and did not see that it was ahead of time.

My next focus on Mobile will be Flutter or Xamarin.

I believe that this 2 will be here like Windows and macOS and like Android and iOS (it's always 2 and always will be).

[ASWC]

"Always will be" is not the type of phrase that usually describes correctly technology in general. There sure are a lot of options when it comes to cross platform techs, they all have pros and cons. Some seem to be a safe choice for the immediate future, some might be more risky, and some like AIR might be extremely risky. Personally I went for Xamarin and typescript (HTML5 + pixijs), I was able to convert most of my AS3 to typescript and then convert again to C# (typescript to C# is a lot easier). In less than 2 years we have almost our 100+ apps fully converted and published. We still have a few remaining AIR apps that will be gone this year. Site note, typescript and C# have everything I dreamed for AS3 and knew will never ever come when Adobe killed ASNext.

[PippoApps]

Interesting. Do you use TypeScript+Pixi to build apps in Xamarin? Or TS+Pixi for web and then Xamarin for apps?

[ASWC]

TS for web and Xamarin for Apps but we are experimenting with ionic as well with the idea of having only one codebase (TS) for everything, results are encouraging so far.

[zwetan_uk]

I think you're underestimating by a lot what kind of dev community would be required to maintain the sources of Adobe AIR if it was open sourced.

Do a simple test

go there
GitHub - adobe/avmplus: Source code for the Actionscript virtual machine
and compile it yourself

then go there
GitHub - adobe-flash/crossbridge: Welcome to visit the homepage!

and compile it yourself

then go there
GitHub - adobe/WebkitAIR: Webkit used in Adobe AIR
and compile it yourself

now imagine that compiling all the above is easy compared to compiling Adobe AIR
which must have a code base that include all the above but even more complicated stuff

You really have no idea how good you have it with Adobe AIR

[Leo Kanel]

I don't at all. I know it is a hard task, but I prefer them opensourcing against to just EOL it. Although it is hard, there are many people who are up to the task (I am not one of them).

[-_Simon_-]

> Adobe AIR which must have a code base that include all the above but even more complicated stuff

I don't think anyone is disputing the developers' competence, but Adobe shifted focus away from developer tools some time ago now - and it's hard to argue the move has been bad from a revenue PoV. It's noticeable on two of those three repositories you've pointed to, the last issue is from an Adobe Open Source Officer asking if it's OK to archive them yet.

[PippoApps]

Hilarious, we are all debating based on guessing and suppositions

I think the only thing we can certainly agree is that it would be nice to have a semi-official statement from Adobe.

Even though, they might just let it live the way it is, and if it is brought to the attention of shortsighted management they might decide to cut it off :S

[Leo Kanel]

Maybe we should get back to the point of the topic though.

Is there going to be an official word for ETA for 64bit Android and any kind of Roadmap (even as simple as we are going to fix that bug from 2008)?

[DQvsRA]

For multiplatform apps AIR might not be the best option now a day. But its still good for games, heavy in graphics, especially vector graphics, hundreds of vector images, that is rasterizing in real time and it's very well fit into the designer/illustrator workflow with one button export and update, kind of "secured" in the app, and your framework tuned to use it. And of course, you can rewrite the whole app working full-time, not a big deal, clients pay for support, but when you are INDIE game developer spend all your free time working after hours and on weekends, sacrificing time with family to be able to release your dream project - it's hard, and dropping support for AIR, means dropping support for many people who put a lot of effort, believes and personal life in to the tech - it is not a gentlemen action. I see it as very arrogant and selfish, like dropping a "best" child from the family because somebody said that we don't employ him instead of cure and educate the way they need.

[Leo Kanel]

If you are an INDIE game developer, you should give game maker studio 2 a try. Good, fast, modern IDE, can handle heavy graphics and export for, web, UWP, mobile, consoles, desktops. A lot faster dev cycle.

[PippoApps]

AIR is like a swiss knife. You can do amazing things in no time taht are simply unthinkable in other technologies. Of course it all depends on the project. Pros and cons, but workflow is simply the best. It is very very useful for graphic intensive presentations, complex kiosks with updated content, and marketing apps.

I.e.: at the moment I am working for a client to a high-end iPad app, with content completely updatable from a CMS, and connected via UDP to other applications running on desktop.

I have exported a Mac version ofr the client, who's design team can update static content and XML directly seeing the results on their desktop machines. This is just an example of where AIR is incredibly handy, days of workflow saved.

Another projct was an android custom build frame alarm clock, which turns into a 2 way video chat connected with a TV studio in real time. The studio could open connections remotely at presenters homes around the world, set remotely the best camera settings for network, and connect wideo streams to the broadcast equipment.

All this, guess whatm with an Adnroid AIR app and a Desktop AIR app in studio. Complete dev time: 2 months. Unthinkable with other technologies. And I did explore other options before, but it was just too time consuming and over budget. (check it on Behance)

I can tell you so many, where AIR allowed us to be in time and in budget with confidence.

Then, gaming... try to build a complex multiplayer casino for all devices connected to SFS server with players playing together (On-Line, cross-platform Poker and Roulette - Demo of multiple devices (Mac, Android, iOS) - YouTube ), easily customizable for new languages etc. in game maker... well, if you have money and time you just build 2 native apps... Or use unity, going crazy building the interface.

Cordova, Xamarin, Unity, Game Maker, are great and better than AIR for many things, but there are projects where only AIR can allow a realistic timeframe and budget. If it goes lost it will be a pity.

[DQvsRA]

I remember I made this interactive game installation in less than a month, and was using AIR for many other projects
Megafon Wakeup Racing Game - YouTube

This one project back to 2012 was also built with AIR in less that 3 weeks:
Hyundai MIAS 2012 - YouTube

Peace.

[PippoApps]

Beautiful products! Exactly, the time to market allowed by AIR is incredible.

Just today I had a meeting with a potential client, I showed them some stuff done in AIR with the timespan, and their jaw dropped. they couldn't understand WHY they didn't know at all about such technology. Maybe Adobe could answer that

[PedroFernandezEteria]

You are absolutely right.

What should be clear is the type of projects that AIR (Flash ecosystem) allows. I think there are some users who are confusing the goal of someone working with AIR.

If my goal was specifically to develop applications based on UI (mobile, windows, etc), then it is logical to think of alternatives because they exist.

If my goal was specifically to develop games, then it is logical to think of alternatives because they exist.

Xamarin, React, Ionic ... They are all solutions to develop UI applications.

Unity, Game Maker, etc, etc ... They are environments / platforms to develop games or interactive 3D.

None of these solutions are optimal for developing "interactive 2d multimedia ". Are there alternatives in this field? Very very few. Hardly any.

Perhaps it is that some of you have changed your business model: "before, we developed multimedia experiences, now we develop up UI applications or games ...". It's a different business...

Html5 canvas (most engines export to this format) as a "multimedia solution" is infinitely worse than Runtimes Flash. It's kind of like patch after patch. To optimize it so that it executes "well" everywhere you have to work a lot (convert ALL to bitmap including texts, BEFORE the execution). In devices you have to "pray" to make it look fluid (sometimes webgl is worse than normal canvas ...)

Nobody talks about this, but flash runtimes are especially good in gpu mode to render vectors, texts + bitmaps in a very fast way. It is only necessary to optimize and convert a few things from vector to bitmap in runtime. I do not know of any other more effective solution. ALL the rest are based on converting ALL to bitmap to obtain certain performance. In Flash / AIR, gpu mode none of this is necessary. It works fast, out of the box !.

An example of an opinion that nobody comments: "Flash Player has a very optimized software rendering engine which is much faster than canvas for certain things. If you are using a lot of vector graphics ...".
https://community.openfl.org/t/direct-comparison-of-medium-sized-game-flash-as3-vs-openfl-haxe-issue...

Html5 DOM is faster than canvas, but it hardly has equivalents to do similar things. Unless you place everything "manually" (yeah, it can be done....) ... As in the old way. No visual editor to assemble designs, without consistent language (api) ... Dom is for layout documents. The rest is a patch.