AIR is now from HARMAN, what to expect?

Report · May 30, 2019

It is good or bad?

Report · Jul 08, 2019

Andrew from Harman has posted several times that they are going to take Air as a serious business and that they intend to pick it up from where it was left by Adobe.

That is not necessarily a good thing. It's empty pr talk.

"Serious business" also means that they will ditch it if it doesn't generate revenue after 1-2 quarters and that they won't EVER open source the sdk.

As they also don't have full access to the source code, and even less so for Adobe Animate, you will never see a fully fledged IDE to support the sdk - which will drive away every possibly new developer these days.

And there are not that many serious developers left. I could e.g. look at subscription numbers at Distriqt to take the wild guess that there won't be more than a moderate 4 figure customer count without new users.

We will move away from Air for our future projects, the risk involved is just not acceptable for a "serious business". We can't risk leaving our apps (which generate 6 figure sums) stranded. We can also no accept royalties just for the sdk usage.

Even the 20-30$/Month comparing a sdk with Unity without any provided eco system (it's all done by the community...) is kind of ridiculous.

P.S.: Let's be realistic: Harman doesn't even manage to put a simple website online in over two months. To me that is some trust lost right there and looks like a company with top-down management and massive bureaucracy. You think these managers care if they pull the plug in half a year?

Report · Jul 08, 2019

Right on, the long term goals are published by HARMAN and they are so few and so not ambitious, it can only mean they have a very small team and very low budget to start with. So that's why there's no website really, no budget, that's why the long term goals are so few, no budget. Everything will be done according to how much income they get from AIR.

Now for the people that want to say "that's how a business works" I'd like to remind you in business you can also INVEST money in hope to get a result, flutter was a HUGE investment it it still didn't get 1 cent of income, Unity was a HUGE investment before it started bringing income. HARMAN starts with minimum investment, minimum risks and minimum vision and will count every quarter the bucks to see if it's worth keep going, the chances of this "strategy" saving AIR in the long term are almost null.

Report · Jul 09, 2019

Whoever expects a medium-term plan to turn AIR into the next Unity, forget it.

The objective of Harman is to continue developing the SDK, not to make it the next technology to destroy the rest of the technologies on the market.

Google, Adobe (for the type of products they want), Apple, Microsoft ... are companies with resources capable of creating complete "Unity's". The Harman division that is developing AIR is a "small" company (its not "Samsung"). The same can not be demanded from them.

Creating a "Unity" from scratch, it's an immense job. Years and years to get where they are (editor and language status, marketplace, extensions, etc). In addition to this, I believe that the Unity company has had great luck at the moment with the decision of Adobe to leave Flash aside from 2011 ... It's not just that they have invested in that technology. Everything has been perfect in time for them.

If Adobe had supported Flash the right way (2d and 3d at that time with a lot of potential) right now Unity probably tended to have a niche market, because Flash would also approach 3D to a very advanced state (editors, frameworks, etc).

So it's not just a matter of investing a lot of money in a technology. You have to have some luck in the market.

What Harman is doing is the right thing to do. They are adequate plans for a medium term (1 - 2 years).

I always say that you have to be realistic, and if someone do not like where AIR is heading, change technology (and do not waste more time in AIR forums, because it does not make sense ...)

Report · Jul 09, 2019

Regarding other technologies, we will see how long they last in the market. What I see, most are half-baked solutions that last only a few years at the top. None is perfect.

Give me a language like as3.0 (with its limits, like all) as well as the consistency and completeness of the ecosystem, and I leave it for you (those who want it), more "modern" solutions, which "neither" have any guarantee of continuing at the top market five more years (which is the concern of many) ...

Report · Jul 09, 2019

Agreed, I think the 1-2 year roadmap is realistic for a small team just just picking up AIR. I'm sure they're just starting to realise the many issues within the source and having a conservative plan is better than promising the moon and not delivering. In 6-9 months, perhaps the list will grow a bit to reflect other wishes.

Our company is happy to see the items increased texture memory, 64bit and possible desktop web browser update ( A really a big item it terms of functionaility as one can integrate many newer browser based techs into an AIR app )

Report · Jul 09, 2019

So first of all, I was talking about longer term thriving of the whole ecosystem, and how to start getting attention of new users (namely "designer/coders"). I do understand that in the next say 12 months, the priority should be to stabilize AIR runtime and prove there are still some users left to pay for it. Also maybe thats everything that Harman wants forever and is ok with that. But:

"That is not realistic is that you can think at this point to convert AIR into a complete solution with an editor that exports directly to Starling, gpu, 2d, 3d, etc. Forget this. In fact there is no company that has that so "complete solution", except Unity (mainly for 3D). "

Why not? I wouldn't expect an excellent system, like Unity editor where you could add custom user extensions for the IDE. Its about having one simple visual workflow, rather than having nothing. Or having Animate CC, which is more like adding to the confusion about what AIR really is. The key is focus. 3d is definitely not needed, that is Unity territory. Also you write as if Starling, gpu, 2d are three separate things to make it seem more complex, lol. It's just 1 single thing - a simple visual editor with layers, keyframe animations and some boilerplate code to publish the thing you create straight to apk. You don't need to build AS3 editor IDE from scratch as well, just use whatever is available (IntelliJ, Flash Builder etc.), like Unity uses MonoDevelop or MSVC.

There is a big legacy of Flash users who were firstly designers, that's another reason why I think it would be a smart thing to do for attracking new (or new/old) users. I had a designer client like that, he used Flash / AIR for a while to build simple mobile games for his clients.Then wondered why those simple games with couple sprites were so sluggish(he used filters for example..why not, when the Animate IDE offers that?). Its just confusing! I tried to explain about Stage3D and the GPU acceleration and how filters are really no good anymore..but that made him even more confused. He didn't care about knowing about Stage3D or programming from scratch, he just wanted to build simple stuff with couple sprites added visually + some code, done. Eventually, he migrated to Unity, which is IMO more complex/awkward for 2D but it seemed like more "complete" workflow to him. And there was the asset store, so yeah.

Sure, Harman should firstly focus on keeping the current hardcore AIR devs, because those are probably the only ones left at this point. But after that, I think if they want to start appealing to their former userbase or people similar to that and grow...the visual path is the one I would take. Its really not as complex as many of you seem to think. I developed a simple visual system in HTML5 / JS / jQuery for Photoshop, as a hobby project on the side. It took couple months. You could publish apps straight from Photoshop to your phone, with visual code(translated to JS/jQuery), with keyframe animations, sounds etc. I created hidden object game room, calculator, piano app...very quickly. It wasn't perfect, but I was alone and had to use the weird poorly documented Adobe extension language.

The hardest part BY FAR is developing the tech underneath - the actual AIR SDK. Its like the engine of the car (and also the wheels etc.). The thing we're missing is like the metal on top...try to sell a car without that to the masses.But that is what pays for the development of all the hard /costly stuff.

Report · Jul 09, 2019

Google extends the deadline for Adobe AIR 12 months....

Android Developers Blog: Get your apps ready for the 64-bit requirement

So this way we have one more year to port, to wait Harman or to remove the apps, we can still update our games/apps with AIR 32 until August 2020.

It's ironic that for the guys that already have a 64 bit solution have more time to solve this imposition, imagine that Windows force every developer to update their apps when started to support x64, even today are released new 32bit apps without any problems.

Any good 64 bit OS is capable to execute 32bit apps without any problem, I don't know why Goolgle is imposing this architecture for ALL apps, if this is not going to bring any significant improvement to the Android Users, on the counter, it will consume their storage capacity faster, because 64 bit apps are HUGE.

Report · Jul 09, 2019

Care to elaborate a bit more? We have 64 apps (native android) which are about 2mb. How 64bit apps are huge?

Report · Jul 09, 2019

Every developer in the world knows that a 64bits app technically requires double space than a 32bit app because they use 8 bytes for memory address instead of 4, in fact is not fixed a 32bit app could be between 20% and 50% smaller, but allway 64bits apps are bigger and use more memory, The Character "A" requires 4 bytes to be stored in 32bits and 8 bytes in 64bits, this why don't have any sense force developers to deliver only 64 bits for a mobile world.

If your 64bit apps are about 2mb for sure they'll use about 1mb if they were 32bits, and for sure as well there are no graphics inside, only text and vectors. For Games the reality is other, the same 32bits game could require not only twice the space to be stored but twice the memory to be executed.

Report · Jul 09, 2019

What you say is true, but I wouldn't call it huge though. I thought I missed something lol

Report · Jul 09, 2019

Well if my game is going to use 50MB instead of 25MB, well that's huge for me.

Report · Jul 09, 2019

Working with Unity for games, 50mb feels light as feather to me lol But I get the point.

Report · Jul 09, 2019

juvelez wrote
Every developer in the world knows that a 64bits app technically requires double space than a 32bit app because they use 8 bytes for memory address instead of 4, in fact is not fixed a 32bit app could be between 20% and 50% smaller, but allway 64bits apps are bigger and use more memory, The Character "A" requires 4 bytes to be stored in 32bits and 8 bytes in 64bits, this why don't have any sense force developers to deliver only 64 bits for a mobile world.
If your 64bit apps are about 2mb for sure they'll use about 1mb if they were 32bits, and for sure as well there are no graphics inside, only text and vectors. For Games the reality is other, the same 32bits game could require not only twice the space to be stored but twice the memory to be executed.

I'm not really sure this is accurate. 64bits code indeed uses more bytes for pointers and other data structures, so you can access larger memory offsets. So it takes more RAM at runtime, and the compiled code is also generally bigger. But if you have data files in your APK (let's say a SWF for a JPEG ile), I doubt their size will be doubled because the APK is compiled with 64bits flag.

Report · Jul 10, 2019

Might be good, because it provides amazing support especially business enterprise customers, Also it embeds flash layer with clients SWF (Small web format) applications.

Report · Aug 10, 2019

"I've begun to realize that you can listen to silence and learn from it." The silence over the techonology (the SDK) is so freaking loud!!

ajwfrost75

Report · Aug 10, 2019

We've started distributing AIR SDK v33 - if you're not getting information about this, please email us at adobe.support@harman.com.

There are still a number of delays to us getting our basic website up and running though, legal/marketing/security reviews are still ongoing. Once that's up, I'll post a link and everyone can just go and get it rather than us having to spend all the effort on manually maintaining the distribution!

thanks

Report · Aug 11, 2019

This forum is broken. People who aren't logged in can't view more than the first page of comments. Your comment is on the 4th page, so anyone who isn't logged in can't see it. Go ahead, try it. Open the 4th page here in a logged out incognito tab. It just brings you straight to the first page, and when you try to click to the 4th page it just keeps reloading the first page. Most people who aren't logged in just assume this forum is completely broken, and they probably take it as more evidence that AIR has been completely abandoned as well. Imagine what percentage of developers urgently need to see comments such as the one you just left here to get the slightest flicker of hope that they should stick with AIR, and they won't see it simply because they aren't logged in and don't realize they need to log in. A lot of us are also quietly waiting on the sidelines to see a sign of professionalism before paying for anything. I haven't even investigated who I am supposed to pay or how much to get these SDK updates, and I know I'm not alone. This is absurd. Get your website up. Now.

Report · Aug 12, 2019

For what its worth: The paging is only broken on mobile, I think. Also you can click "latest reply" to jump to the end of a thread. And in defense of Harman, this is an Adobe website and not under their control. But aside from that I agree, an Air website is a must have now.

Report · Aug 12, 2019

Yes I also agree! Sadly there are parts to this that are outside of our control, particularly given the use of different companies' trademarks and other legal issues that need to be resolved. It would make our lives a lot easier if we could just push everything out via a simple website..

Report · Aug 12, 2019

Not just mobile. Paging is completely broken when logged out in Chrome on my iMac Pro.

I am mostly angry on behalf of Harman. If it weren't such a precarious situation where Harman is forced to play nice with Adobe, then I'd suggest they go ahead and sue Adobe just for the state they left this forum in. Surely part of their deal must have involved Adobe doing the bare minimum necessary to maintain good relations with developers and help retain the very people who are expected to pay for the SDK. They didn't, at all. We got one confusing blog post and that was it.

Harman should find all of the following unacceptable from Adobe.

- The broken Adobe forum driving a large percentage of confused developers away, even though fixing paging probably just requires changing a single line of code on the server. I mean seriously, it's probably an "if" statement checking if you're logged in and then doing a 301 redirect to the first page if you're not. Just remove the "if" statement. Take five seconds and do this. Jesus christ.

- Not taking five seconds to add a CSS rule to the forum to give ajwfrost75 and other Harman employees some sort of official-looking badge icon

- Their insidious frightening opaque messaging of "we're killing Flash, which implies we might kill AIR too, but we're not going to tell you what we're doing with AIR until the very last minute"

- The blog post where they say "Harman has experts who will help you to migrate away from AIR" which confuses people and implies that even after this transition AIR will definitely be abandoned even by Harman which just encourages everyone to abandon it now rather than later

- Not letting anyone who is logged out find the comforting counterarguments to any of the above that are necessary for anyone to continue to have faith in AIR

- The radioactive PR disaster that comes from all this that prevents developers from being able to even talk their bosses or partners into agreeing to use AIR for anything

If Adobe set out to deliberately sabotage Harman here and make them fail at the starting line on purpose, then they could not have done a better job than what their mere tone-deaf incompetence accomplished. It's astonishing.

Report · Aug 12, 2019

Adobe has always been the problem, who knows what Flash and AIR could have become if Adobe never was in charge of their future, probably something amazing AND lasting.

Report · Aug 12, 2019

And Harman will be the last lunge, the beginning of the end of AIR, they are a small team thinking more in the money they will earn, that in the future of AIR, if they had been smarter, they would have welcomed the entire current community and allowed them to update their apps to Android x64. So they will build the enough confident on them and retain a huge potential community to earn the needed budget to be rich and to maintain AIR properly, but today they are only earning enemies, detractors and massive desertion, with that evident ambition that are pushing away and forcing to search alternatives. But they must be rich now or nothing, a pretty blind business vision.

Report · Aug 12, 2019

What are the top alternatives people are moving to? Asking for a friend.

Report · Aug 12, 2019

It depends what market you have and what performance you need. Unity is the last truly good perf all platform framework with google flutter trying to cover all platforms as well. You can do cordova for all platform but perf is not that good. After that you have other solutions but they don't cover all platforms, xamarin is a good one for Ios and Android only and of course after that native dev.

Report · Aug 12, 2019

Agree with ASWC. Aside from Unity the open source alternative Godot is a quite nice option as well. Runs on Win, Mac, Linux, BSD, and Haiku, and deploys on iOS, Android, Win, Mac, Linux, UWP, html5 WebAssembly. One-click deploy and export for most platforms. For console support you will need to contact a company, though, to do the conversion.

It offers a full IDE with GDScript (Python-like), C# 7.0, and C++ (without the need to recompile the engine), debugging, and it uses a friendly file system approach for its project files, making it fully compatible with version control systems.

Full 2d and 3d support, and a built-in GUI framework with theming. Anything can be animated with its built-in timeline (comparable to Flash's timeline with nesting timelines, etc.)., including function calls.

https://godotengine.org/

Godot is gaining more and more traction, similar to Blender. But of course, depends on what market(s) you are in. Like Unity, Godot is aimed foremost at games dev, but I have done interactive children's books in Godot. The scene/instance/node system and its 'animate all' option is attractive. I find it a pleasure to work with. The built-in scripting language is a doddle to learn (GDScript).