zwetan I agree. As developers, we are often tempted to idealize other technologies, as if they were miracle solutions. You know, the famous "grass is always greener..." Then you try those other techs, and you realize that, like with everything else, there are bugs, steep learning curves, missing features, lacking documentation, confusing paradigms, and so on.
For example, during the last few days, I've been really hyped by all the information regarding Haxe + OpenFL posted on these forums. On paper, it sounded awesome, and the new Graal for AIR users. Today, I spent the day installing everything and trying it. And it made me realize how good, stable, well documented and mature AIR is. I don't say OpenFL is bad, as I'm sure it has nice features and big potential. But for me, it still feels rough around the edges, especially due to lack of centralized documentation and tutorials. I also faced some strange behaviors (maybe due to my lack of experience, but in all cases, it didn't feel always intuitive).
Every change requires time, efforts and compromises, even for techs that are similar to the ones you already know. Nothing is for free, and that's when you have to do things yourself that you are happy to be able to pay for some support. I hope Harman will be able to go on with AIR, and even if I first had mixed feelings about their announcements, I'm now comfortable paying for their support.