In a recent post Unity said webGL is still at least a year out for real deployment. IE simply won't support it (period) and the new IE on Win 10 will, but its adoption rate will be crap for a while. Chrome and Firefox handle it ok but it soaks up system resources and has severe memory limitations.

No good answer really. Unity Player is dead for sure. Flash may be (but I doubt it). WebGL may be years before it reaches stability and hardware accessability. Gonna be a bumpy ride for web based content!

???? IE supports WebGL in version 11 on all Windows versions. Check https://msdn.microsoft.com/library/bg182648%28v=vs.85%29.aspx?f=255&MSPPError=-2147217396

Microsoft even have Khronos Group membership these days: https://www.khronos.org/members/contributors

Playing videos work fine as html5, in fact it works better than the flash player, takes less ram and less cpu.

I don't know about the RAM or CPU usage--I haven't looked into either--but my own experience with the HTML5 version of the YouTube player was enough to convince me to install a Firefox extension that enabled me to switch back to the Flash version. I'm not sure of whether I had performance issues--I think that I may have--but the main issue for me was that for some reason the player lacked some of the resolution options offered by the Flash version, including the resolution that I find works best for me (480p), being of acceptable quality for most videos while streaming via my connection without buffering.

In all fairness, I'm using a fairly old machine, and running Ubuntu.

In a recent post Unity said webGL is still at least a year out for real deployment. IE simply won't support it (period) and the new IE on Win 10 will, but its adoption rate will be crap for a while. Chrome and Firefox handle it ok but it soaks up system resources and has severe memory limitations.

No good answer really. Unity Player is dead for sure. Flash may be (but I doubt it). WebGL may be years before it reaches stability and hardware accessability. Gonna be a bumpy ride for web based content!

Everybody is applauding the death of Flash but there isn't a really good alternative out there yet for web games. It's going to be a bad time for a lot of indie developers in the near future. This is going to end up killing the Unity web player (already dead on some browsers) as well. For as crappy as Flash is you can pretty much expect a Flash game to run the same on any web browser. Same can't really for HTML5.

I feel like this sums up my current situation. I spent a few years learning Unity and developing a tool kit that I was really happy with, only to have the web plugin killed and the crap-tacular WebGL exporter be the alternative. After a few months of HTML5 engines and all the frustrations of cross-browser support and canvas/rendering/audio/scaling quirks, I switched to haxeflixel.com for it's Flash support. It does have a html5 exporter and I find that when it works it works great, but if you get any errors post-compile you really want a pure-js engine like Phaser to debug.

I feel like I'm stuck between a rock and a hard place right now. Use Flash and risk it all going up in smoke in under a year? Or move to HTML5 and just hope it gets better and not worse over the next few years, putting up with all it's annoyances and quirks?

Make a mobile app, and use HTML5/WebGL for the desktop browser if you must. There is no "Write Once Run Anywhere" solution.

Unfortunately "HTML5" (whatever that really mean) remains a mess of a non-standard and WebGL suffers the same kind of cludgeyness that has plagued OpenGL.

And on the other hand you have Flash, which is dead. The question is, why are you so insistent on browser-based delivery? The majority of the addressable audience is already on smart phones, and even with the hassle of developing both iOS and Android, it's got to be less than handling all the browser edge cases for "HTML5."

Honestly, about the only good thing about them you can say is that they haven't had their security flaws discovered yet or at least not to a degree where 'the sky is falling!' is declared anyway.

It's also unlikely to ever happen because they aren't a parallel or superimposed environment to the browser, so any vulnerabilities that affect them will be intrinsic to the browser.

Have any of you looked at Adobe's Edge authoring tools? It's interesting to note that games are not an area that Adobe is pushing at all with Edge; perhaps that's a sign that you better get your native mobile on.

I feel like this sums up my current situation. I spent a few years learning Unity and developing a tool kit that I was really happy with, only to have the web plugin killed and the crap-tacular WebGL exporter be the alternative. After a few months of HTML5 engines and all the frustrations of cross-browser support and canvas/rendering/audio/scaling quirks, I switched to haxeflixel.com for it's Flash support. It does have a html5 exporter and I find that when it works it works great, but if you get any errors post-compile you really want a pure-js engine like Phaser to debug.

I feel like I'm stuck between a rock and a hard place right now. Use Flash and risk it all going up in smoke in under a year? Or move to HTML5 and just hope it gets better and not worse over the next few years, putting up with all it's annoyances and quirks?

That's software engineering, though. Platforms rise and fall for reasons completely unrelated to your efforts. Languages come and go. New markets and verticals emerge, and your bosses insist that the entire company has to pivot because Twitter Cards are totally the next thing! In fact, that's life…

The savvy response is not to stew and fume over the demise of your favored target. Sigh, brush off some books/docs and get to cracking on whatever is next.

What I do not understand is why so many of you seem fixated on browser-based delivery. Have you looked around recently? People love the idea that they have their time-wasters and toys in their pockets, accessible at a moment's notice. Gotta wait until your number is called at the DMV? Whip out your phone. Waiting for your doctor/dentist? Whip out your phone. Riding the subway/tram/trolley/bus/street car/Uber to work/date night/wherever? Whip out your phone.

What I'm saying is that you should be making mobile apps. And, let me be clear: the economics of mobile apps are a shit show, but games are about the only vertical that is consistently profitable, even without excessive use of in-app purchases.

So. If you're just making games as a hobby, and you don't care if anyone else plays them, then who gives a fuck what tool you use to make it? Build it in Flash, and take on the responsibility of ensuring that you always have a stable Flash plugin for your hobby hacking. But if you want to make games as a social effort, even if not (yet) a business—if you want people to discover, play, talk about and recommend them—then you need to go where the people are. And that sure as shit ain't Flash in 2015.

They want to kill Flash because of security issues? They do realize that everything with a circuit board is exploitable/hack-able, right? Even if you migrate everything to HTML5...all you have to do is right click to view the source.

Anyway, if Flash were to disappear anytime soon, I don't foresee Flash game developers having much difficulty porting their games to HTML5/Javascript. It's only a matter of changing the syntax, correct? Javascript is pretty straightforward in manipulating elements in exactly the same way as Actionscript (grab an ID, change attributes).

Even if you migrate everything to HTML5...all you have to do is right click to view the source.

Let's not perpetuate the myth that obscurity is a form of security.

One of the major drawbacks of Flash relative to HTML5 is the fact that Flash is proprietary. Anyone can go find and fix security bugs in the Chrome or Firefox source code - not so for Adobe's.

Anyone can go find and fix security bugs in the Chrome or Firefox source code - not so for Adobe's.

Lets be fair to them however, when bugs are found and they are informed they are fixed in pretty short order; and that informing is a pretty key point here, if no one reports the flaws in Chrome or Firefox they are unlikely to be fixed either. (The recent series of OpenSSL issues point towards that particular problem...)

even with the hassle of developing both iOS and Android, it's got to be less than handling all the browser edge cases for "HTML5."

As someone who deals with Android day in, day out, I really doubt that.

With HTML5 you've got maybe 4 or 5 targets to worry about, targets that have different problems but a limited number all the same. They lack tools, although Chrome is improving in this area, when it comes to things like debugging and for large engines they are remain a problem (I see next to the guy who handles HTML5 support in UE4; he does not enjoy this.).

Android, by contrast... well... when I left work on Friday I had 8 devices on my desk - I consider those 8 separate platforms and I have more in a draw. They lack tools and frankly, atm, I do 99.9% of my debugging via log printing and not a small amount of god damn genius. Android is hell.

I guess if you are a developer you could put your trust in a 3rd party to develop your engine and tools (and deal with that pain) but then, to a large degree, we are right back where we started with trusting things which are installed not to have security holes which can be exploited on the devices.

Lets be fair to them however, when bugs are found and they are informed they are fixed in pretty short order; and that informing is a pretty key point here, if no one reports the flaws in Chrome or Firefox they are unlikely to be fixed either. (The recent series of OpenSSL issues point towards that particular problem...)

I don't really buy your base premise there - yes, bugs can occur in any piece of software, that's a given. The question is what methods one has available to find and correct such bugs.

In the case of open-source software, each and every customer has the ability to investigate the source code for potential bugs, including running static analysis software, to establish a base level of confidence in the implementation. In addition, every customer has the ability to vet each additional change submitted to the software, to establish confidence that new bugs are not being introduced. And it's important to keep in mind that open-source software release cycles are typically driven not by the need to push out a new version, but by a 'it'll be ready when it's done' mindset.

In the proprietary case, both end users and commercial customers are left hoping that Adobe's engineers just happen to be incredibly knowledgeable and conscientious about security, and that Adobe's managers aren't pushing for releases before all bugs are ironed out... Neither of which are true, outside of a few specialised software teams in NASA and the defense industry.