The author is right. But Jobs was also right, Flash was fundamentally to resource intensive for mobile and arguably still is. Without mobile, Flash was doomed. Flash also has a bunch of other issues in terms of security and performance on PCs which were never really addressed.
My experience was that Flash ran like garbage on anything but Windows/Intel. Linux/Intel? Trash. Mac/PPC? Trash. The web at the time was unusable on my PowerBook G4 without FlashBlock. Maybe Flash would still be around if Adobe bothered to make it run well on anything but one platform.
HTML5 still doesn't replicate what mattered about Flash
61–70 of 372 posts
Re: HTML5 still doesn't replicate what mattered about Flash
#62Why didn't https://en.wikipedia.org/wiki/Adobe_AIR take off as successor to Flash? It is available for all platforms, runs ActionScript.
Re: HTML5 still doesn't replicate what mattered about Flash
#63And I'd like to argue the fact that there is no such (integrated) tool existing indicates it's a niche market.
Re: HTML5 still doesn't replicate what mattered about Flash
#64Earlier quoted context omitted.
There's been lots! I've funded and contributed to many of them! The most famous is probably OpenFL, which uses the Haxe programming language. It does exactly that. You can even use Adobe Animate to author content for it. The main missing pieces, however, are: 1) The abstraction inevitably leaks. HTML5 is just a fundamentally fiddly bundle of tech and there's always this or that little thing that can shake loose. 2) T…
> The tech burden is fundamentally higher with an open source project no matter how you try. With Flash you never needed to even touch a command line or download anything other than the Flash authoring tool. What does open source even remotely have to do with that?
Re: HTML5 still doesn't replicate what mattered about Flash
#65Honestly, just based on what I used Flash for, the big thing I miss has nothing to do with Actionscript or any of its code or distribution. Distributing SWF files was no easier than it is today to distribute an HTML directory in a zip file, in either case the only reliable way to share your stuff was to link it on a website. And obviously, the web works for that.
So what I miss is not the code or the platform, what I miss is the animation pipeline. I feel like Flash was a relatively unique animation tool and content creation tool and the web hasn't really tried to replicate that, in part because it really has nothing to do with the web's capabilities.
What we're missing is probably not some kind of giant API effort on the web that would make Flash animations possible, what we're missing is a standalone program that generates animations for the web in a format where they can be manipulated with code. As a pure prototyping tool for game development, I would already rather be working in a program like Construct instead of Flash. I just don't want to use a separate program to generate sprite sheets, I want them integrated really nicely into the timeline and the coding environment.
But I don't think that has anything to do with the web at all. The web definitely isn't perfect, but with WASM and WebGL we have more graphical/processing power than Flash ever offered, and aside from some problematic areas like web audio, controller input, and other game-focused APIs that browsers have been slacking on -- what we're really missing is a development environment. Flash's development environment wasn't self-hosted in an SWF. You don't need a development environment for the web that is self-hosted in a web page. You just need a thing that spits out WASM. It's not the web's fault that nobody has made that tool.
----
If all you're worried about is programming and Javascript, then Construct 2/3 already has more accessible programming environments than Flash ever had at any point in its lifecycle.
Construct isn't as powerful as Flash was. But setting up a quick prototype game in Construct is fast, and it has been fast for a long time, and that's the important part. The important part of Flash was not how powerful it was, it was purely about the accessibility. And Construct 2 is an extremely easy environment to learn to code in. Even as a seasoned programmer I liked Construct 2's drag-and-drop visual interface; it made it easy to visualize the overall shape of the code.
But again, Construct lacks that animation pipeline where you can draw, animate, program, and export all in one kind of unified environment that allowed you to jump around a lot to different tasks.
And of course, Construct also isn't Open Source so you don't know what the company is going to do in the future. That's a real problem, that's why I don't use Construct at all, even as I'm singing its praises. I will not use a proprietary development environment. But it at least outputs to an Open format, and the people who miss Flash are probably not going to be losing sleep over using another proprietary engine anyway.
Re: HTML5 still doesn't replicate what mattered about Flash
#66The author is right. But Jobs was also right, Flash was fundamentally to resource intensive for mobile and arguably still is. Without mobile, Flash was doomed. Flash also has a bunch of other issues in terms of security and performance on PCs which were never really addressed.
I call a bit of BS about this. In the end, the code that creates js or HTML5 equivalent (poorer versions in fact) end up being far more resource intensive and buggy than flash was. Importantly, this is after millions of man hours were put into optimising the hell out of the js engines. Further, apple itself ended up optimising it's processor architecture to allow faster JS execution. What would have happened if Steve…
It never played well on any system I cared to use. The tools to generate it were unavailable to the likes of me and a lot of people here, who back in the day were second class citizens.
Also it tried to make the the Web TV before there was anywhere enough processing power at the ends nor bandwidth in the middle.
Designers on their expensive powerful machines built magnificent content that nobody who respected the technology in their personal stack, and paid a reasonable price for their infrastructure could use.
Well, back in the day we may have been second class citizens. How the world has changed. Good riddance to bad rubbish!
Re: HTML5 still doesn't replicate what mattered about Flash
#67I have some sympathy, but not a lot, because I don't feel like this is a problem with the web. Honestly, just based on what I used Flash for, the big thing I miss has nothing to do with Actionscript or any of its code or distribution. Distributing SWF files was no easier than it is today to distribute an HTML directory in a zip file, in either case the only reliable way to share your stuff was to link it on a website…
It PAINS me that when somebody wants to upload a friggin' cartoon on the internet they have to render it out to 100X the native size and introduce compression artifacts. We were more sophisticated than this in 1999!
Re: HTML5 still doesn't replicate what mattered about Flash
#68The tweet mentions that you could 'just' send an SWF file to someone and it played. Well, if anybody had to fix a computer, or hosted anything Flash-based during that era, then they would know that's bullshit. Sure maybe with the standalone Flash player it worked allright, but that's not how people really used Flash. Even in my experience, a single SWF running locally didn't always work across all versions of Flash p…
My experience was to upload to Newgrounds and Kongregate and send it to my mom on an ancient computer and she'd run it just fine. I am sorry for oversimplifying but I do think many people had the same experience as I did.
If your only requirement is that you should be able to put your code on a website and your mom should be able to visit a link and run it, then the web is already significantly ahead of where Flash was.
Of course, the accessibility of creation is a problem, but if you send someone a link to a canvas-based animation, they are more likely to be able to play it today than they ever were in the days of Flash. They can even play it on mobile. Unless you're doing something very experimental, they'll likely be able to play it even if their computer/browser is out of date.
I mean, if we're simplifying this down to the common experience, the way people consume HTML games and Flash games is literally the same -- they visit a link. I have self-hosted remote playtests of the current game I'm working on by putting the game on a static page, and that is just as accessible to playtesters as anything that I was ever doing when I was building Flash games. More accessible, because now I don't have to send people to a wiki to get the game playing on Linux.
Re: HTML5 still doesn't replicate what mattered about Flash
#69I suppose you can distribute HTML5 as standalone binary with Electron? And I'd like to argue the fact that there is no such (integrated) tool existing indicates it's a niche market.
I think something more like Web Components would be the direction I'd look in for a solution to this.
Re: HTML5 still doesn't replicate what mattered about Flash
#70Earlier quoted context omitted.
> My point is this is not standard or even typical behavior and so you rarely see it. It could easily become a standard if someone writes a convenient JavaScript library or composer tool to do exactly that. My point is that the underlying platform is not the limitation.
Understood. My point is that whatever convenient library/composer tool that can do that doesn't exist yet (or isn't widely adopted), and it's been 10 years, and 10 years ago a bunch of people said that everything Flash was offering would be perfectly replaced in short order.
Because - the flash monopoly is basically "there for the taking". The market's demonstrated a huge amount of interest. There's a runtime that can do it (the modern web). All that's left is we just need someone to build an authoring tool, and they'll instantly win over that huge group of users who've been demonstrated to exist (not just legacy users, but tons of new folks with a similar mindset). Building an authoring tool of course is incredibly difficult, but - given the potential payoff, you'd really think someone like adobe would be all over that, if only out of fear of being disrupted.
I'm just really surprised nobody's done it. I expected it 5 years ago. Hell, I'm even surprised how sluggish flash emulators have been to show up - we're finally getting what I think is the "right one" in the form of Ruffle ( https://ruffle.rs/ ), but I'm surprised the older Shumway project didn't take off. Then again, maybe we simply had to wait, because I don't think anything prior to WebASM would have really been appropriate for the job.
I think the best case scenario would be some sort of "commercial open-source project", like how Krita structures themselves, taking up the mantle of being a flash-like authoring tool.