AIR33 Where is it?

[Leo Kanel]

Is it me or AIR33 Beta is delayed? We are a few days before April and there is no beta version yet. It looks like the 1st quarter release is missed.

[Leo Kanel · Accepted solution]

The announcement has been made. Harman - Adobe Partnership - HARMAN

[PippoApps]

This is fantastic news.

Also: "AIR 33 will support 64-bit Android devices and will be available on a commercial basis"

I hope this means we will finally be allowed to PAY for Adobe AIR. And finally Adobe will PROFIT from it. So will finally ADVERTISE the beauty of AIR.

Am I a dreamer?

[federicor30206780]

Great news finally !!!!

A revamp for Air ? I'll be more than happy to pay and have the support we deserve.

[PippoApps]

This HARMAN seems like a serious company... hope, hope, hope...

[federicor30206780]

PippoApps  escribió

This HARMAN seems like a serious company... hope, hope, hope...

Yes, it's a big one so let's hope a real change and a move forward.

[William Spence]

Was just reading about this, but don't think I entirely understand what is going to happen.  Is Adobe still going to do AIR development and Harman is just going to provide support, or is Adobe handing off AIR entirely to Harman and they take over from here??  Will AIR now become a paid service, if so, how much?  I think I like what I am reading because it feels like it is going to provide extended life to AIR, but I have to admit, I don't understand at all what is actually happening.

[PippoApps]

Also how AIR integration will work in Animate CC in the future.

Well it is going to be interesting times...

[ASWC]

This is good news instead or at least this gives ONE chance for AIR to be saved and become again relevant. Time will tell.

[William Spence]

This sounds a little exciting to me.  From my understanding, Harman has been involved "behind the scenes" since 2006, so this gives me the impression that the transition won't be horribly bumpy, they already know the machine.  I also like that there will be a pricing structure to AIR, a paid product is much more likely to be able to deliver the kind of support and development that we will need to remain relevant years into the future.  My only concern is that I don't know if it will be associated with the Adobe brand anymore.  I don't know if people will be as excited to sign up for Harman AIR as they were about Adobe AIR.

I know that Andrew is gone for the night, but I'm waiting with baited breath to hear all the answers.  What IDE's will we be able to use, will Starling still be in the picture, what will the pricing structure be, will Animate be supported, will the Adobe brand still be attached or will this become Harman AIR???  Sooo many questions, but I feel a little excitement behind all of this.

I always wondered how a company profited from an unpaid resource that required so many highly skilled programmers and support staff.  It has to be a huge drain on a company's finances with little reward.  If the price is fair, AIR could become a relevant player again because it allows us all onto so many platforms from one code base.  I love this about AIR.

[ASWC]

Yes there's potential for a complete rebirth. At least imo under Adobe the future of AIR was doomed. Now hope can remain but we will have to wait and see. It's either gonna be awesome or it's gonna be disappointing but at least now nobody can know for sure. With Samsung behind for sure they have the financial capabilities of doing something great at least in theory. I don't want to get too excited cause it's still a long shot but this move gave me back something I lost years ago for AIR, hope.

[ajwfrost75]

Hi

It will remain as Adobe AIR, we don't want to (or have the right to) rebrand it.

And Samsung probably don't know anything about this, they bought HARMAN a couple of years ago but we're relatively separate operationally, at least in my part of HARMAN..

IDE-wise there shouldn't be any difference: most IDEs will use the AIR SDK under the hood, and we're intentionally keeping the structure of this the same so that there's minimal updates required. Some IDEs may need to add an 'arm8' option in a dialog box perhaps..

I'm fairly hopeful, and certainly encouraged by the responses that I have seen around this announcement. We've perhaps not been 'behind the scenes' in a way that you expect: we've not been working for Adobe on this as suppliers, but instead we've had/got a license from them to have their source code and port it to platforms that they don't support themselves. So for example Flash Lite powered the user interface of over 5 million vehicles, AIR was on a large number on a QNX-based product, and we're still supporting AIR for an automotive supplier on an embedded Linux OS. Likewise we support Flash Player and AIR on a number of Set Top Box products... less of these now, as companies had been moving away from this platform, but it means that we have a very good knowledge about the AIR software.

While I would love to work on an "ActionScript 4", and had previews of this prior to Adobe canning it, I don't think that's going to happen. I've been following the "language improvements" being made by the folk working on Apache Royale with interest, some of these could have value - but we're not supporting the ActionScript Compiler, just the AIR runtime and its SDK. The developer websites and language reference site will continue to be maintained by Adobe for a while, we'll be discussing what happens to them post 2020 though, if necessary we can take those on.

Starling and Feathers are great projects, but currently we're unable to commit to any funding in the way that Adobe had been doing this. It wasn't factored into our calculations.. I would like to be able to support them in some way, perhaps via some sort of freemium approach that we could help with, we shall have to see.

Hope that gives a few more answers!

thanks, Andrew

[PippoApps]

Thank you andrew, very informative.

What do you mean by not supporting the ActionScript Compiler? Isn't that supposed to be part of AIR SDK?

cheers

[KR-san]

Umm... I think you're honest and have good news about 64bit support.

But I can't accept positive.

Reason:

1. Herman focus on automatic driving.
2. They have migration buisiness from AIR to JS.

I guess Herman bought AIR cheaply to help their business.

Therefore, they will do maintenance but they don't want to evolve.

I am disappointed with Adobe. In near future, Herman or Adobe declare EOL.

Thank you read poor English.

[ajwfrost75]

Linux is an interesting one - we've supported AIR on Linux the whole time 🙂 but for our OEM customers who pay for the integration services. If there's a business case for reinstating Linux support for the AIR SDK, then we would consider it..!

HARMAN itself, yes lots of automotive and audio work, but that's all independent of this initiative. We don't have "AIR to JS" migration work, but we do have Flash to JS migration business because of the browser limitation coming in at the end of 2020.

This definitely isn't about "helping our business" - it's tiny, in the grand scheme of things for HARMAN - but we want it to be successful, and for that we know it needs to be properly maintained and updated. But yes, it also has to be commercially viable: the money coming from the product will fund the team that develops it. I'm hoping this has plenty of years left in it, personally I love this technology and use ActionScript/AIR for any little utility that needs to be knocked up.

Re the comment above about the ActionScript compiler: yes it's distributed with the AIR SDK, but we don't have the source code for this from Adobe...  There are other tools that we had discussed e.g. ATFViewer, and still some open issues, so we will see how this goes. It's close collaboration, it's not like Adobe are just throwing the whole thing over the fence at us!

thanks

[Frédéric C.]

Thank you Andrew for your answers and transparency. I'm sure this partnership has required much more efforts and discussions than we can imagine.

About the future of AIR, I have a question: what do you plan on the marketing level? Is your approach to just update and maintain the technology, so it fits your own needs, being funded by people who already use it? Or do you plan to invest in marketing AIR as a great cross-platform tech, to also attract new developers? Many of us have always been depressed by Adobe's lack of communication about AIR, so people are not even aware of its existence. Do you plan to make AIR more popular by promoting the tech?

Thank you.

[federicor30206780]

https://forums.adobe.com/people/Fr%C3%A9d%C3%A9ric+C.  escribió

Thank you Andrew for your answers and transparency. I'm sure this partnership has required much more efforts and discussions than we can imagine.

About the future of AIR, I have a question: what do you plan on the marketing level? Is your approach to just update and maintain the technology, so it fits your own needs, being funded by people who already use it? Or do you plan to invest in marketing AIR as a great cross-platform tech, to also attract new developers? Many of us have always been depressed by Adobe's lack of communication about AIR, so people are not even aware of its existence. Do you plan to make AIR more popular by promoting the tech?

Thank you.

I think that for that pay model works, it's essential to invest in marketing and invest in evangelizers who post in blogs and communicate about the usability and potential of the technology. It's the way to attract devs and gain more paid users. I think it's important also to have a good community site ( with projects showcase ) and a clear roadmap.

As many of us think, AIR is a mature and good technology that is capable to do almost everything and compared to the other crossplatform technologies is quite strong in a lot of points. The worse part of AIR has been clearly the support, totally neglected by Adobe and the lack of communication in every aspect. Those has been the main reasons why the people jumped to other frameworks and I completly understand it. They jumped because the confidence not because the technology, when you have customers and need to start over a project you evaluate pros and cons and obviously the feeling was not good at all with the future of AIR.

If Haman is able to recover devs confidence and offers a really solid future path, I think little by little we can recover audience and devs because the technology itself is quite good.

[PippoApps]

Andrew,

as a suggestion for the paying model, I wish to share with you how we use AIR.

We do publish on stores only in 10% of cases, the remaining 90% is to create kiosks and installations, or enterprise/pre-installed iPad/Android apps to deliver marketing tools for people on the go, or to deliver applications that in a closed network in UDP can control larger AIR applications running on big screens.

This means we have to find a way to pay for commercial AIR apps when no royalties are involved such as:

- published on a store

- commmercially used in kiosks

- used for non-distributed commercial apps

Some unusual use cases:

- Android devices with pre-installed app for cinema ticket scans and seat selection (both for commercial movie thatres, or for movie theatres inside Hospitals, so free for charity).

- Wall screen with AIR interactive presentation, and several pre-installed iPad apps used by staff to control and input data in the wall screen, in large shopping malls around the world

So probably a single fee x screen on desktop, or x app on esnterprise app.

Thank you very much.

It is heart warming to finally talk with someone involved in AIR development

Filippo

[Frédéric C.]

An idea regarding the paying tiers. What about making a deal with the current ANEs vendors to include access to their ANEs in an "Extended" tier. Something like:

1. Free Tier

2. Distriqt Extended Tier (with full access to Distriqt ANEs) and MyFlashLabs Extended Tier (with full access to MyFlashLabs ANEs)

3. Pro tier, including advanced support.

I think teaming up with ANE vendors might add great added value to your tiers, and would allow building a solid AIR ecosystem: All actors would benefit from it, and developers would have a consistent and complete offer to rely on. I'm convinced we are always stronger together.

Just saying...

[ajwfrost75]

Some interesting ideas here, thanks everyone for the feedback!

In terms of Android versions, the idea is to make it work across as many devices as possible. If you'd have a preference for using newer features that wouldn't then work as ANEs if the runtime was targeting the older support, then we can look into this...

The compilers we have in binary form and are licensed to provide these as part of the AIR SDK, so no worries on that front. I don't think they've been updated for a while... although of course the 'older' AS3 compiler is open sourced and now managed by Apache Flex, and the Royale guys have been adding interesting features to their branch of this e.g. private constructors, which caught my eye..! We'll see..

iOS, Mac and Windows: Adobe will continue to support AIR v32 on the desktop platforms so those aren't a priority for us yet, but we'll look at bringing in support for those in the next year. iOS yes we will be supporting but we're focussing heavily on the Android platform first because of these 1st August deadlines!

ANEs and including those .. that's an interesting thought, we can discuss further with those guys. Making the whole situation simpler (i.e. so you don't have to pay lots of different entities) would perhaps be useful for developers, but we'd have to look at how this works in practice to make sure that the ANE vendors don't lose out. There's a similar possibility with Starling and Feathers, although of course those are free currently (unlike ANEs) so it would depend on whether Daniel/Josh want to change their commercial model.

I think some of those discussions may have to come later, just from a practical perspective we've got a lot of work to get done here for the Android platform and setting everything up! So extra packages/bundles with third-party software is a nice idea but fairly low on the priority list current...! Keep the ideas coming in though 🙂

thanks

[William Spence]

Andrew, any chance you guys may be able to finally update the WebKit in AIR? The current one is getting quite long in the tooth and is starting to cause compatibility problems. This has been a big concern of mine as I have to use webviews in my AIR apps, and more and more problems have been coming up. I don't want to, but if it gets much worse, I may have to look at other options.

[ajwfrost75]

I think we need to have a virtual "suggestions box"...!

Language updates in AS3 - so for example the 'arrow functions' syntax perhaps? I'm interested in what else people might want out of the language.. To be clear though, there's a distinction between the compiler and the underlying language elements, so for example we can update things like the Array.insertAt() methods and add in new ActionScript top-level functions etc.. for syntax updates - and possibly for new primitive types - we would need the compiler code though.

Updating the WebKit version would be a lot of work! I would be interested to see the number of people who are looking for that. Is StageWebView not a suitable alternative? We can potentially do bespoke development where there's a need...

How about if people have an idea for a new feature, they can email adobe.support@harman.com, and put "[Feature request]" at the start of their email subject..? Until we have our own forum/bugbase set up, thiis may have to suffice...

thanks

[chaky]

Andrew,

You sad:

to be clear though, there's a distinction between the compiler and the underlying language elements, so for example we can update things like the Array.insertAt() methods and add in new ActionScript top-level functions etc.. for syntax updates - and possibly for new primitive types - we would need the compiler code though

I am mostly interested in making many of the AS3 methods compatible with object pooling like for example Matrix.transformPoint

where it would be great to supply another parameter (in this case Point object) where to write the result instead of creating new objects on every call.

Currently Matrix.transformPoint and many other methods create many objects. We do not have option to use object pooling and save on memory allocations and avoid triggering GC a lot.

Thanks

[Frédéric C.]

William, did you have a look at WebView ANEs by Distriqt and MyFlashLabs? They seem pretty advanced, so maybe it would not make sense to reinvent the wheel?

[PippoApps]

Frédéric I believe Willaim is referring to StageWebview on desktop.

Very useful in many cases. I.e. showing a pdf file.

[William Spence]

Frederic, I currently use a MyFlashLabs RichWebView ANE for my mobile projects.  But my concerns for the webkit are with StageWebView for the desktop.