AIR is now from HARMAN, what to expect?

[clouds40936465]

It is good or bad?

[Leo Kanel]

Don't even bother. A lot of people post they want web assembly, game consoles, linux and all crazy things from Harman. They are so sure, this will become the next unity or unreal engine, that is painful to see.. Probably most of the people without concerns about this, never managed a big team or even worked in one. Most of the people who are super excited / positive, never had to deal with shareholders, and answer why Adobe is dropping AIR.

[hferreira80]

Support for Linux is a crazy thing !?

Are you serious ?

[Leo Kanel]

Business wise it is crazy. In order to move forward with it, it would mean that there are a lot of requests for it, and people would pay to get it. Right now, it makes no sense.

[PippoApps]

It's a while that some of us (active or ex hardcore AIR developers) interact regularily on this forum. Somehow, even if we don't know each other personally, we get to feel each other personality by what we write. I have to say that I enjoy enormously our exchanges here, even the toxic ones. I am from a latin country, maybe that's why I am used to accept and understand occasional emotional bursts (including my own), without judgement or hard feelings, but trying to laugh about it, like I would do with a family member. Writing online can bring to impulsive communication, it might happen for any reason, also just because we are struggling with the same bug for 30 minutes or we are worried for a deadline or some personal issue... I like to think we are all on the same side anyway, and whatever happens it's just an inevitable part of the beauty of human nature. Long live to AIR and to you my friends.

[hferreira80]

I agree.

[PippoApps]

ASWC is not only right, he is extremely competent. One of the few I follow on this forum, since his posts, although enfatic sometimes, are full of precious knowledge.

If we hadn't lived scattered around the planet it would be fantastic to share a few drinks amongst air devs. Leavers or remainders (fantastic metaphor).

[KevinF]

Harman mentions that AIR 33 will be available on a "commercial basis", which probably means charging money for the update. After all, why else would they take over development?

AIR had a good ride, but it's outdated, less mature, and horrendously slow compared to modern cross-platform frameworks. I enjoyed working with AIR, but probably won't miss it.

[ajwfrost75]

Hi

Just to address a few of the points raised here:

- Samsung, and the wider HARMAN companies, weren't a factor in our decision here. We have been porting AIR and Flash Player for years, so in terms of a choice of who to take this on, we're better qualified than most as we already have a team with experience of the thousands of C/C++ files that goes to make up this runtime..! And we do know and love it.. I definitely appreciate the effort that Daniel, Jack, Josh and others have put into this technology and ecosystem and want to do something to try to leverage their experience, but it's early days still for this.

- it's not something we "bought" or something that we've done in order to support any existing internal products or apps using AIR. Afaik we don't have any of our own products using AIR, although it would be something I'm keen to try to make happen! But the way the company is organised means that entities such as Samsung opcos or our Connected Car division would be treated like a customer to us, just like any other customer.

- our 'legacy' work - i.e. the existing ports that we maintain and support for embedded platforms - are not and would not have been impacted by this. We already had the agreements in place with Adobe to support those customers regardless of what Adobe decided to do with the AIR SDK. But we see this as an extension of that work, so we now work to address a much wider base of customers and switch our support into a 'product' mode rather than being services-based.

I'm interested by the "less mature than modern cross-platform frameworks" comment though, as I see it as a lot more mature than some technologies that keep changing so dramatically and that come and go quite a lot. Anyway, we want to encourage your feedback as to how we can make this product more relevant for you all! From our perspective, we want to make this attractive for people to still use and for new people to pick up. So, please keep commenting, as even the so-called negative opinions are useful for us to understand where the ecosystem's concerns are so that we can try our best to address them!

thanks

  Andrew

[Leo Kanel]

Andrew,

I am always happy to hear from you Concerning feedback, and I think a lot of people here will agree, one of things that has a great impact on new people picking up the tech, is marketing. Adobe did none for it after 2012 (no idea if they did before as I was always on the flex people). When I tell people that our products are built with AIR, I always get the same question. "What is that?"

Then I talk about the AAA games and apps made with it, and they are shocked they never hear about this.

For me, if you really want to get new people using it, you need loads of Marketing. A good website, showcases, advertising, maybe events. Good tutorials are also welcome, as we are in an era, that a lot of websites with great tutorials are closing (devgirl.org and many others, were turned off this year..).

[Frédéric C.]

Thank you Andrew for your commitment and continued communication (especially on a Saturday ! )

[ASWC]

Attracting new people to this tech is gonna be your challenge number one. I agree at this point AIR is not especially "bad" in comparison to other similar tech but it lacks clear commitments/means for its future (from Adobe) and so people have been dropping out and new people are not touching it. It's also hard to find any mention of AIR in any cross platform dedicated forums/articles. How do you turn this around NOW is clearly a difficult question but one your company is gonna have to find a clear answer to if your intention is really to jump start this tech again. Imo you won't be able to do this without clear roadmaps (probably very ambitious roadmaps) and aggressive marketing (if nobody knows what you are doing with the tech then nobody will pick it up).

[Frédéric C.]

@ASWC, would you have some examples and suggestions about what an AIR roadmap may contain, based on your needs?

Because I may lack imagination, but I don't see that many lacking points that may make AIR unattractive for newcomers. I mean, if I read your last post correctly, the new roadmap should bring AIR to another level. Let's be realistic, it may not be first Harman's goal: From my point of view, Harman seem to want to maintain / polish / add some new features, but not make AIR another Unity for example, at least not immediately. Does that mean it can't be a viable solution that may attract new people? I don't think so. There are many techs and engines that are available now, and many of them are finally maintained by a small team, or even some volunteers. Roadmaps are not that common for even popular engines. So I can understand that you or your manager need a product which evolves a lot, with big roadmaps, large communities, high investment, and clearly AIR is not for you then. But big players like that are not that many, and I think AIR is definitely not ridiculous as an alternative for middle sized companies and indie developers.

Personally, I think that AIR as a tech is still quite attractive, as it has many powerful features. Recently, since the AIR future was uncertain, I looked for alternatives. Here were my main criteria for finding a challenger:

- Complete set of features (with ability to call native features on mobile systems)

- Good range of targeted platforms

- Ability to develop on Windows

- Nice, active and supportive community

For now, it seems Unity is the closest to my needs, but not as good as AIR. However, I asked myself: "if I was a total newcomer looking for a crossplatform solution, would I pick AIR?" Honestly, I wouldn't. Here is why:

- I would have probably found AIR through an obscure article, without knowing much about it

- I would have made extra research but would not have been able to find a single page listing main AIR features. So for me, it would still remain very unclear about what this tech really allows to do.

- Then I would have come to those forums, and with all the negative messages about doomed AIR's future, it would have been the last nail in the coffin. I would have run away, trying to find something else. Too bad, as AIR would have fulfilled all my needs, and beyond!

I agree with you about the fact marketing is very important. And enhancing AIR would be great, of course. But one step after another. If Harman manage to create an attractive website, where newcomers could clearly understand what they can get from AIR, with centralized resources, it will be a great starting point. And let's take our part of responsibility: as a community, we also have to show more enthusiasm. Who would feel like giving AIR a try by reading the last 2 biggest threads on those forums? Not me. So I think that if we want to support AIR and Harman in their efforts, we also have to show that what remains of this community is not just a bunch of depressed devs who are just here to argue with each other...

[ASWC]

Well Andrew said they want new comers, they want to attract new developers to AIR. I think he's correct, this is the goal number one if they want this tech to survive and have a future. Now as I said, it's not impossible but clearly it's not gonna be easy and it's gonna be challenging.

First, marketing of course and I don't really qualify to give advice on this.

Second, don't focus on AIR features, focus on AIR attractiveness and by that I mean reassure new comers: yes AIR is in good hand, yes AIR has a future, yes AIR will get bunch of nice new features, etc ... You do this by pushing regular updates, adding new interesting features, a framework that is very ACTIVE is a framework that is attractive.

Third, new feature? Personally I think the loss of web publishing option was a big deal (yes you can still publish to Flash but it's useless now), so option to publish to : webassembly/html5/webgl (any really) just like Unity or flutter does. This would boost the interest of AIR immediately.

Now can they do it? Can they pull it off? I really wish they do, but it's not gonna surprise anyone I'm not very optimistic right now.

[PippoApps]

Web target would be nice. Still I believe today the web evolved in a way that using a screen development environment such as AIR, aimed at building apps that run on a screen, and not in a scrollable browser, wouldn't make too much sense as a web port. Unless for gaming, but in that case you are forced to use unity webGL (buggy and with problems we all know to well). The bad guys won, and we lost the browser. Imho, as an apps developer and designer, AIR is STILL the best, without web export.

So, realistically, I guess maintaining and improving mobile targets (desktop already works well enough), and advertise the real advantage of AIR, reassuring about the timespan, it will be more than possible to attract new users. Businesses mean profit, and i's just matter of letting businesses know what can be achieved with AIR wih little time and budget. I am talking about communication, advertisement,  gaming, art installations (art sector is a larger market than most expect), museums, integrated systems, automotive, prototyping, closed systems (I build a lot of stuff for i.e. malls, theatres, exhibitions, museums), enterprise apps, marketing suites, events, educational, etc. etc. These are big markets, where a lot of money goes. Of course if Apple wants to launch their new killer app they will go native. They have cash and angineers and must support their own platforms.

I do believe AIR is mature enough to offer a lot of advantages for a whole lot of business sectors, and just maintainance/update roadmap with assurances of lifespan will be more than enough to market it. Also, the Flash bashing days are far enough and people got bored of that hype, so that AIR is not associated to Flash anymore. Let's try to stay opositive, and see what happens in the future, also welcomeing Frederic suggestion, trying to avoid to do bad advertisement for newcomers if they stumble on these posts

[ASWC]

You need to look at the big picture, web option could make a big difference for AIR surviving. For example, Flash banners used to be everywhere, today it's all HTML5/Webgl and it's not done in Unity but it's possible some will start doing those in flutter. Anyway the point is there's not many option when it comes to make banners but what if AIR becomes another option? Now you'll get more people looking at AIR for developing their marketing banners. AIR will get mentioned more in forums as an alternative dev cross platform. This is one way marketing can work, little steps, become more relevant for more specific jobs so that people talk.

So yeah you can do without web option but that's a lot of people, tons of people, that will have no reason to talk about AIR in their forums/articles. Tech success depends on people talking about it so give them as much reasons as possible to talk about AIR.

[Frédéric C.]

@ASWC It seems Unity is actually also interested in this market, by the way: Instant Games | Unity

[ASWC]

Yes it's a market and it's a cross platform market as well. I think AIR must get in there to be more relevant so for me web publishing option is a must.

[ajwfrost75]

Awesome discussion, thanks everyone!

To talk about the web target, there are two sides to this, or rather two ways in which it could be achieved. The first is in the development tools, where you can write your ActionScript like normal, and then publish it for a web target rather than publishing it to SWF format. This is what things like Royale are doing, and whilst the generation of JavaScript (or WebAssembly) is possible, the main issue then is in how the APIs are implemented i.e. flash.display.Graphics, and all the rest. Getting them all to work would perhaps be impractical, if not impossible, given the difference in rendering models between the Flash Player and a DOM-based browser.

The second option then is to have a "runtime" that is written in WebAssembly and loads up a generic SWF file and runs it. So in this sense, you're still creating the SWF files as usual with the open tools like FlashDevelop etc, and then it's loaded into the equivalent of the Flash Player, sitting within a browser. Which would be great, but again is an awful lot of work and you're looking at all those APIs again. Trust me, the Flash Player (and AIR) code is not simple, there are complexities in there which are pretty astounding which are behind things like the API compatibility matching and security, sandboxing etc, not to mention hooks for supporting Scout and the Flash debugger.

But yes, it would be great if we could do this! I think we may be getting ahead of ourselves at the moment though.. I agree on all the marketing side of things, we're looking at a gradual handover of some of the material from Adobe's development centre and yes should be looking at promoting this and encouraging the existing community as well as then trying to promote it..

[el111]

Thanks for the detailed explanation on WebGL/Webassembly.

I have often thought similarly.

Browser games are an ever decreasing share of the market. It is 2019 not 2005. Most people walk around with a super-computer in their pocket. There's practically no money in targeting the browser. For Desktop there is Steam and Microsoft are now even allowing Win32 apps in their Store (UWP simply hasn't caught on).

As for ads. They are built in GUI-tools which already output to HTML5 (such as Animate). They have a shelf-life of 30 days. They are not coded but churned out.

I concur with PippoApps​ that installations, media screens, kiosks etc are growing markets and well suited to AIR.

​Andrew, you mentioned Audio/Video . Adobe did have in their beta release notes that they are "in the process of reworking our video pipeline to better use modern hardware capabilities and leverage native operating system API's" It would be good to know if this is still in the works?

I think the key things around AV are stability and support for the latest codecs (eg HEVC on iOS).

[phillipkerman]

I actually did a kiosk about 1 year ago with video recording, and for at least 24 hours, I actually considered AIR.  What was I thinking?  The great thing about Flash and AIR was that Adobe dealt with licensing that various media formats it would support.  The world has evolved.  Browser video is ready for prime time.

Delivered on windows using reactjs (not native), node/express, ffmpeg, and a few other open source stuff.  We used KioWare's kiosk shell combined with Chrome. The technology choices had no impact on business decisions once I started.  Plus, the fact there's like a billion OS libraries in JS it was never an issue getting or extending existing code. For example, while KioWare has an onscreen keyboard software, but when I was in discovery, I could easily compare several open source solutions to this.  In the past 25 years of using "multimedia" tools I've had to create my share of onscreen keyboards.  Using a popular platform ("the web") means you can leverage all that.

I'm not trying to pick a fight with this one, but wanted to share my experience where I sincerely considered AIR.

[PippoApps]

This is not a personal issue but we are trying to help HARMAN staff to get ideas from the developers. If you want to discuss other technologies please do it somewhere else. If we are here is because we don't just believe in AIR but we enjoy coding in ActionScript, if you enjoy JavaScript and the millions of free library, plenty of forums for that.

My respect to Zwetan, a very active contributor to the AS3 community in many ways.

[ASWC]

You are missing the point completely, if there's a time to criticize AIR or talk about why AIR was not considered for a project it is NOW. HARMAN has no use right now for "I love AIR so much please give me more", this will not help them make a better AIR or a more attractive AIR. They need, they want, criticism NOW, they want to know WHY people don't pick AIR NOW. How else are they supposed to make it better? Looking at a crystal ball? So please think about this for 2 minutes and then WELCOME criticism of AIR and ask for more.

[phillipkerman]

I got the email--thanks.

==SUPPORT (occasional emails?)
What mechanism are you going to use for support? ...Adobe's forum?  Asking for direct emails to "ANE" of "FEATURE" or whatever is more lame than the worst Adobe support.  You might think I'm impatient, but "where's the beef?"  So far it's just press releases and emails. Saying "maybe" another release at the end of July...etc.  Man, I've worked on some fly-by-night projects, but it's June 8th right now....  Most of my customers in other tech have clear plans for at least 6-12 months out.

==PRICING (yes, I know "it's coming"):

What's the pricing structure? I can't afford to even take the time to install a (still not materialized) beta without knowing the offer you're making. No doubt the fanbois on this list will dig right in--though it might not always be best to get feedback from such blindly obedient users.

==TEAM (is it literally just you Andrew?)
Please disclose the kinds of resources committed to this experiment.  Is it just one dev?  You say you have 1000s of C++ files, but please tell us the size of the team and their exact objectives/roadmap.  If you're making promises of solutions in the next few weeks I think it's a reasonable expectation for me to try to assess a confidence level.  From 0-5 what's your confidence level that you'll deliver a ready for prime time 64bit solution by July 1?  

FWIW, of the three major clients that had some existing remnants of AIR two have made prudent business decisions (to leave) based on Adobe's inaction and silence... they don't even want to hear what this HARMAN thing is all about.  One glutton for punishment at least still wants to know what the deal is--but honestly, there's nothing I can tell them except excerpts of your comments and emails.  Despite my negative tone, it's actually way more than Adobe has done in the last 24 months--so kudos.

Straight advice: deliver the promised 64bit solution before asking for new feature directions. 

[Leo Kanel]

If it was Adobe and not Harman, you will be crucified by the well known fan boy right now whose name starts with Z =P

That been said, it feels like they are trying to make something work at the very last minute. Adobe dropped the ball on this one, and at least we have someone who managed to pick it up. Any company who value their customers has already migrated, so let's wait to see to what this whole thing will lead. Don't expect too much as AIR is after all a dying art (not a technology people please don't go crazy. I am just saying that cause you can't find any devs any more).

Whatever Andrew and his team delivers, will be a welcome gift, as we at least escaped the sudden death of August deadline.

[hferreira80]

"fan boy right now whose name starts with Z =P"

Hehe, in fact