Flash on the web only worked if you had Flash player plugin installed. Both browser and Flash updates themselves would break the Flash plugin constantly. That was across all browsers and platforms. Keeping plugins updated and reliable was a mess, for a long time there weren't even auto updates.

Flash was a mess the whole time. Let's not pretend that it ever really "just worked" because it didn't.

Nobody had a standalone Flash player installed that wasn't in a browser, and when I moved to newer versions of Flash, I would regularly run into problems where someone would have a Flash player that didn't support the stuff I was trying to do.

Flash was really important for its time, and I agree there are things I still miss about it that the web hasn't been able to replicate. But my goodness, as someone who uses Linux today, I can't express how glad I am that the websites that never worked on my Linux machines now do work on my Linux machines. Cross platform my butt, Flash was only ever "cross platform" on Windows, and occasionally on Mac. Everybody else on mobile and Linux got ignored because Flash was a proprietary blob of insecure DRM-riddled code and Adobe couldn't be bothered to give anyone any attention unless they were big enough that they were impossible to ignore. Yes, I'm incredibly disappointed that the web hasn't captured everything Flash did, but it is good that we have gotten rid of Flash because mobile and Linux users deserve to have a version of the web that works on their machines.

These people who are lauding Flash as a universal environment either just have rose-tinted glasses or they lived in an even smaller programming bubble than the one they accuse HTML advocates of inhabiting.

C'mon I never emailed people SWF's, seriously. We uploaded them to flash portals and sent links to our friends. The single-file aspect was important because it made it easy to share (for both you and the flash portal). Also crucially it made it easy for competing flash portals to "steal" your game and spread it even further.

> as someone who uses Linux today

I get what you're saying and I love Linux and go out of my way to support it, but I don't think the Linux audience in 1999-2008 was exactly who most of us were targeting, nor did it have particularly much market share among desktop computers then or now.

Well, but... yeah. I know people weren't targeting Linux in the early 2000's, that was the problem. That's what I'm saying.

One of the biggest advantages of dumping Flash was that it got replaced with a runtime that worked on every OS even if people weren't deliberately targeting it. I'm saying that's a big enough advantage that it's worth the tradeoff. The Linux market share on desktop computers was tiny back then, but it was big enough that the community deserved to have a web in its entirety that worked for them. Not just a couple websites, they deserved to have full access to the platform that claimed to be a democratizing environment, and they were never going to be a part of that platform as long as Flash existed.

I disagree that Linux compatibility back then was a non-issue. It was not. It was a serious problem that held Linux back as a platform and as a daily driver and we all hated it, and it feels very dismissive to say that it was just a tiny side effect or that it's irrelevant to the broader point. Flash was never cross-platform and it held back the entire web from being a universally accessible platform. And it only got fixed because developers were forced to use a runtime that supported Linux. That is the only reason that Linux browsers are usable today.

----

I am kind of skeptical that distributing SWFs is actually that much easier than distributing zip files. Even downloading/stealing and re-hosting JS code is not that hard. This is coming from someone who agrees with many of your points about the overall accessibility that Flash brought. But I do not believe it's harder to distribute content on the web today than it was back then.

I uploaded SWF files to those portals, I put them on flash drives. My process today for distribution is pretty much the same. And once this stuff is on the web, the web is vastly better at distribution than the Flash Player plugin was.

More succinctly, it's unhealthy for any company to 'own' the web. The standards process we have today is far better.

I also recall having a stash of standalone SWF gamęs on my shared network drive at school, because all EXEs were restricted and SWF files could be opened by drag-and-drop on the Internet Explorer window.

The same is still true of a lot of other "niche" application areas.

This is my view as well. Windows on x86 was the golden child. Linux was the hated bastard at best (anyone remember switching between OSS, ALSA and PulseAudio to eke any sound out of the plugin?).

And even that only on x86. ARM? What, you mean that's a processor architecture? That's not even going into the utter dead silence when it came to supporting other Adobe plugins outside of Windows. Adobe Air? Shockwave? I can still easily feel my blood boiling at the absence of these plugins, after so so many years.

"Cross platform" to many meant Windows 98, XP, 2000 and NT5. You could work across that, you had the 'enterprise' market apps people using 2000 and NT which could then also be used by employees at home using Win XP. What more could you want?

I'm guess I'm being only half-sarcastic here.

In my experience, cross-platform is less enjoyable to develop. Yes, it might be more valuable to users for a program to be cross-platform, but it is definitely more enjoyable to develop something that just has to run on a single platform.

For cross-platform work, you have to use languages, runtimes, and libraries that are supported across multiple platforms.

When you are doing single platform work, you can use all the cross-platform languages and libraries and tools, plus all the other languages, libraries and tools that are on the single platform.

I suspect the difference was even more stark in 1999 because Windows was even more dominant than it is now.

What? Literally everyone who downloaded the flash plugin and ran the installer got a standalone player installed as well. It was right there in the start menu for anyone to find.

Flash only ever worked properly if you were running it in Internet Explorer for Mac OS 9. Any other browser/OS combination and you were looking at low framerates, audio/video desync, or both. Linux was worst of all.

Flash was the worst thing to support in those labs - some windows machines, some macs, some Solaris... gl trying to tell non-CS majors to move to a different computer...

I’ll admit, the authoring tools were great, everything else about flash was terrible.

It seems weird, but when you consider the context at the time, it makes sense. Back in the late 90s -- when Flash was on the rise and being rapidly developed -- IE5 was the browser to have on Mac just as it was on Windows. The reasons had to do with the infusion of cash from Microsoft into Apple to keep the latter company afloat while Jobs restructured it and pivoted it to Mac OS X. At the time, companies such as Adobe and Macromedia -- which hadn't been acquired by Adobe yet -- developed and tested on Mac first and most extensively, because they catered to the design crowd. As for browsers -- Safari wasn't around yet, Netscape was boomer tech by that time, Mozilla was broken, and Firefox wouldn't come out for a couple years. That really left only IE and boutique browsers like CyberDog and iCab. So the environment that got the most thorough testing was -- once again -- IE on Mac OS 9.

When Mac OS X finally did come out, it would take a number of years for Mac developers to fully embrace it. Remember, to the Mac teams inside Adobe and Macromedia, Inside Macintosh was the fucking Bible. Classic Mac OS was The Right Thing; Mac OS X was something weird and wrong and inherently anti-Mac, as parts of it came from Unix, which was known for having user-hostility baked into its design philosophy. It was also the future, but it was a future they'd take years to reckon with. So, for a while, the native OS X version of Flash chugged no matter which browser you used it in. Meanwhile, IE5 for Mac OS 9 still worked perfectly well provided you launched the Classic environment, so if you wanted top Flash performance it was still available.

> boomer tech

Since we are talking about events ~25 years ago isn’t all of this “boomer tech” because the boomers were at the peak of their careers at the time? Aren’t they the ones that created, like, everything?

Seems like a strangely modern phrase to use to describe what is now ancient history.

Also I’m not that old but I already see the ageism in tech. I see ideas dismissed based on age rather than merit. The normalization of this type of derogatory language makes me uncomfortable.

Does “boomer” not mean “person born between 1946 and 1964”?

Does it just mean ”old” now?

https://www.bondcap.com/report/itr19/#&gid=1&pid=1

Doesn’t that make Halo or CS “boomer shooters”?

Maybe I’m older than I think...

Perhaps you you look/act younger than others.

Or maybe others act/look older than you.

Or, maybe there’s people who exist who exaggerate the first difference between themselves and someone else to put distance between them for reasons like ego/insecurity.

Instead of a lens of doubt and putting the burden of others of having to educate you, it’s also possible to meet half way and ask.. what else could this mean that I’m not seeing? Our brains tend to find what we put attention towards.

Either way I hope that helps and no offence intended.

Are you suggesting that “boomer” simply means “old” now?

Navigator was already effectively dead in 1998, when Netscape announced plans to open source the code and launch the Mozilla project. Navigator 4.0 was a buggy mess, and it did not support the W3C standards of the day. It was only a matter of time before the Web broke enough to force the installed Navigator base to switch.

The kids of the time already knew by 1998 that IE was the future of the web. Netscape had already made their open source announcement and spun off Mozilla in 1998. Navigator was effectively dead.

But the aspect of Flash that hasn't been fully replicated by HTML5/JS/whatever is the fully integrated workflow. That took away much of the underlying woes with environments and configurations. All you did was install Flash, make your game or animation, and a SWF was produced. This ease of beginning surely was a springboard for so many creators and animators, like the Newgrounds community.

I'm also partial to the idea of not everything being done by the browser, and instead plugins specialising in their respective area. The effect of moving everything into the browser only contributes to the Google-led near monopolisation of rendering engines.

It'll be a shame to see Flash go, it really was quite a technology. Couldn't click a link without some cutesy little flash something-or-other popping up.

My memories may be tainted by time, but I'm hesitant to say that modern web apps have surpassed flash yet. A web browser simply isn't up to the challenge of creating a nice interface for an app and I got the impression Flash's tooling was a long way ahead of the game. But Flash stands as a soon-forgotten internet monument of closed source winning battles and open source winning wars.

it was an awful and wonderful time to experience the constant breakage and delight that was the www.

Nokia adopted it and it was quietly embedded in far more devices to provide custom functionality than I realized.

Flash lite even appeared to power the internet channel on the Nintendo Wii.

https://en.m.wikipedia.org/wiki/Adobe_Flash_Lite

This honestly got a laugh out of me.

This lead to my boycott of flash. Sorry flash websites!

Flash was an interesting experiment, as you said, important at the time, however I started distrusting websites, that used it, heavily and avoided them.

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.

The main thing about the single-file thing is that it afforded the flash portal ecosystem. It was really easy to share your work as a creator without a lot of technical experience.

> 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.

Yeah but the problem is creating that canvas-based animation is insane, so there aren't any canvas-based animations, compared to the output of flash cartoons.

Nowadays everybody bakes out their animations to 100x the size with compression artifacts and stick them on Youtube with LESS features than flash had, a direct consequence of the decline of Flash and HTML5's inability as a platform stack to fill the gap.

I love Linux and the open web, I'm glad they exist and I want things to work on them. I just wish HTML5 as a platform had actually delivered on its promises by not missing the forest for the trees.

I think we might be arguing past each other.

I fully, 100% agree with this point. I just don't think it has anything to do with the web.

When people brought up replacing the Flash runtime with the web, there were two things being claimed:

A) that the platform itself could handle all of the technical requirements of Flash

and

B) that tools would exist that replaced Flash.

I believe that the first claim mostly came true, that the web today is about as powerful as Flash was. The second claim never even came close to happening. A lot of platforms exist that export to the web, but even the really popular ones like Unity just don't feel the same. Unity is, unfortunately, not an adequate substitute for Flash.

And I think that's a real tragedy, but I feel like we're at a point where the web should be left out of the conversation. There are game-related APIs that I want on the web, but for the most part there is nothing that browser makers can do to solve your problem.

Someone has to make a development environment. It's a separate problem. The web today is capable of doing the stuff you want it to do, but there is no development environment that replaces Flash.

And that's kind of revealing because it's not just that there isn't a web-compatible alternative to Flash, or there isn't an Open Source alternative. Proprietary game engines that aren't even targeting the web are not as good as Flash was for rapid content prototyping. Why can't anyone, commercial or Open Source, build a Flash replacement for any platform/runtime? That's the conversation we need to be having, not whether or not the web is mature enough as a runtime.

All of the APIs that are necessary to replicate Flash exist. The only failure of the Open web was assuming that the games industry was going to be competent enough to author new accessible tools. Heck, I was critical of Flash deprecation at the time, but even I thought that Adobe was eventually going to seamlessly export to HTML5. I don't know what to say beyond that. On behalf of HTML5 advocates everywhere, I'm sorry we thought Adobe might be a competently run company.

So I wish more of these kinds of threads were subtweeting people like Tim Sweeney, or Unity devs, or Adobe itself, rather than trying to blame the web for what is arguably one of its biggest success stories: the evolution from being only suitable for toy interactions into a generally accessible application runtime. The web succeeded at that, beyond almost everyone's expectations.

Almost everything that the web can provide to replace Flash it is providing. But web advocates can't build an IDE. Somebody in the games industry has to do that. So maybe those people should be the target of this ire instead of HTML5. I get why people are frustrated, but web advocates can't fix any of this. Somebody who builds game engines needs to fix it, the industry needs to actually make tools.

Like, to be kind of blunt, I wish people would stop criticizing the web for the problems that the games industry created.

Web standards evolve slowly and compatibility is often even slower. Developers still can’t target WebGL2 on iOS for example and only just got it for desktop Safari. We’re basically getting to the point where the web is a viable platform for games although there are still significant performance differences between browsers. No one’s going to seriously build tools for a platform that’s very slowly coming together. But the future is brighter there now just not in a realistic timeframe for a Flash replacement to be made.

The expertise to build Flash didn’t actually come from games. The next Flash might not either. It was also a long journey if you look at the development history of what became Flash. Building something at this scale takes a lot of time.

Then at the same time the market changed and a lot of the web game eyeballs became mobile game eyeballs. Hence a lack of industry interest in rekindling web support.

On the plus side I work for a web based game development platform and work pretty much exclusively in Chrome all day building games for the web. Our aim is somewhat different to Flash but we’re in the same ballpark in terms of wanting to make development intuitive and fun.

In the broader scheme of things I think the next couple of years are going to be really interesting for web games in general.

To be fair to the original thread, Apple has always tried to protray that it's doing something brave or prescient by dropping technologies or introducing proprietary standards, and it's pretty clear that they were not actually interested in the Open web when they dropped Flash. I do think Flash was dropped in a clumsy way, and I would try to separate Apple's clear push to try and monopolize the mobile games market and kill web games from the actual push for the Open web as an application platform.

I think Apple has made it pretty clear on iOS that they don't care about the web as an application platform.

----

> In the broader scheme of things I think the next couple of years are going to be really interesting for web games in general.

I... hope so. What bothers me about the long-term future is that I'm now seeing less investment on the web side of things on APIs that I think are going to be essential to pushing web games forward in new ways. Even simple stuff like web audio, good controller layout support (also, controller rumble support please), better timing for fixed game intervals. I think the web could easily replace Flash, but there are definitely more APIs I want in general for games.

My worry is that it takes so long for the tooling in those areas to catch up that the actual APIs start to lag behind because there's not enough pressure on browser manufacturers to care about them. We already kind of saw this with Chrome breaking audio on tons of web games because they just didn't think to test them.

Best of luck to anything you're working on, I'm rooting for you.

All I needed to do was export the flash document as an exe and boom: it worked. The portability was amazing. I would just cut a release, put it on floppies, and take them to school. I never had anyone complain once about having an issue. Oh, and no one gave a damn about Linux outside of server admins back then, so yes an .exe was "universally" portable across whatever windows version people had running. There was also a fairly easy process for creating versions for Macs then too, but I don't remember how I did it.

Also, Flash was very good about having backwards compatible exports, so for as long as you weren't using the latest features you could reliably export to older versions of Flash.

Flash was a blessing in the late 90s and early aughts too precisely because it was the only reliable runtime when the internet was a flaming trash heap of incompatibility during the browser wars. Nothing worked everywhere /except/ Flash. The only other app platform available at the time were Java applets, and those were a nightmare. There's a reason why you used to be able to buy entire websites that were Flash based: it just worked.

Yes, Flash only worked if you had Flash installed, but honestly, who didn't? Was this really any different than making sure you had whatever God forsaken Windows service pack installed on your computer? And yes, I know Flash didn't work on Linux, but almost /everyone/ ran Windows back in the day.

I loved every product Macromedia created and owe a lot to them for having made it so easy to make engaging dynamic content. It's how I got hooked on developing software as a kid in elementary school.

It's a shame that there's nothing as simple anymore. I died a little inside when Adobe bought them, and I feel like the next generation is missing out on what was a really fun, flexible, and simple tool...

Making a ball change shades and move in Flash was a few intuitive clicks on a timeline. Now we have a only have a canvas API with relatively low level primitives. If you want things to be simpler, you have to go find some other package, oh wait now you need node, npm, we pack and 20 other dependencies to build the damn package, did you download the proper CSS file too, oh wait you need to import the script at the bottom of the page not the head, blah, blah, blah.

I miss Flash. Yes, the internet is better without it, but I feel like kids of today don't know what they're missing.

The support for using simple tools to make HTML5 animations and interactive content has been there for years (since at least 2013, when CreateJS was released: https://www.createjs.com/ ))

Adobe Animate: https://www.adobe.com/products/animate.html

Back then, one of the arguments against Flash was that it was closed and proprietary, both authoring- and runtime-wise. Browsers nowadays are arguably open-source runtimes, but we still don't have a good open-source authoring tool that replicates the magic of Flash.

I imagine there are some open source authoring tools out there. I doubt they have the polish that Adobe products have. If you find one, or make one, post it here.

That kind of scene is not present now.

Yes there were huge Gaming communities around it. Flash at the time was like Unity today. And may be even better than Unity from ease of use perspective. Lots of interesting Indie games.

Sometimes I just wish Adobe had open sourced Flash.

My fans spinning up every time I open a random website definitely disagree with you. Wasn't that the #1 argument for getting rid of Flash?

I started developing Flash websites in Flash 3 in school, transitioned into Actionscript 2/3 and did some complex projects over the years. Can't vouch for Linux but on my Mac and on Windows it was definitely great, it DID just work and it was cross-platform. People have short memories - but cross-platform back then was also mostly referring to huge JS/website inconsistencies between different browsers. I mean - when you say cross-platform, what are you even comparing it to?

I had a huge offline collection of different SWF files/websites for inspiration and that I could just start up from the filesystem and they also just worked (if it was a single-file project of course). I'd wager that 95+% of the people also had the full Flash Player installed.

Today I do some complex JS/React/webGL stuff - and let me tell you it takes like 10x as much (money, time, knowledge, effort) to develop stuff that's comparable to what I did 10 years ago in Flash. And this comes with more or just as many cross-browser and performance problems.

Blaming Flash for "fixing your computer" or saying it somehow had "hosting issues" I can't really understand at all.

The "mobile era" just started when Flash started transitioning into smartphones/apps - first results were not ideal but promising. Were Adobe allowed to have 10 years of iterations on those - I do firmly believe it would have been a #1 app development environment today. There would have been no need for Unity or even cross-platform frameworks like Flutter, etc...

But it is what it is - sorry, but anyone who thinks Flash dying was not primarily about Apple or even Google not wanting to keep their walled gardens for themselves is delusional. Adobe definitely has a good part of the blame for not recognising the potential and fighting for it some more.

When you read negative comments, it's always someone who didn't really use it.

It worked for developers, it worked for gamers. My guess is that the haters are just very loud in their hate.

This is generally true, and for a simple reason. When people don't like something, they avoid it. And therefore they wind up not really using it.

Take me for an example. All that I needed to know about Flash is that someone, somewhere, thought my computer should randomly start flashing and making noise. And the way that they did it really sucked to experience on Linux. (Not that I wanted the intended experience.) The fact that the plugin that I needed to let them do so was a repeated source of security holes was just icing on the cake.

So I avoided Flash. And disabled the plugin at some point. With the happy result that a lot of particularly annoying ads went away.

Now I'm sure that there were a lot of people whose experience was better. It was popular for a reason. But it wasn't a better experience for me.

Initially it was a strategy to reduce security incidents caused by out-of-date Flash plugins.

Flash was absolutely horrendous. Also, people keep talking about how easy it was to create stuff in it and how JS/HTML involves too much coding. This is more revisionist history - Flash had its own proprietary scripting language called ActionScript and you definitely needed to have some coding skill to do most things in Flash beyond basic animations.

The amount of hoops you have to jump through to run something interactive in Unity is mind-boggling by comparison (without even talking about sharing that thing with another human on another computer). Processing was also pretty good at the first bit, but it wasn't as tuned to games and sharing your stuff was pretty awful (java applets).

Just because I <3 Flash doesn't mean it didn't have security problems, though. I wish someone had bought the tech off of Adobe and turned it into a gaming platform a-la Steam with proper input support.

Flash also got used for a lot of other things and in a lot of cases it was basically to add in functionality that wasn't supported by native web standards yet. This was why entire websites were built in flash. Because it gave you the freedom to do things that you just couldn't do with regular HTML/JS.

If the point the author wants to make is that creating games in HTML/JS is more cumbersome because there isn't one authoring tool that is used by everybody and has a comprehensive toolset, maybe that's true I don't know. But I definitely prefer this reality to a reality where Flash is still being used in 2020.

When you think about all the challenges the web faced early on, with all the attempted browser lock in and stuff like Flash, we got pretty lucky that things are fully open now.

And because HTML/JS are open and not proprietary, the people whining about it are free to take a shot at creating their own authoring software for games without having to worry about getting anybody's permission or getting sued by a company that owns the rights to JavaScript.

I will say, I think the niche of browser games played by kids is now taken up by phone games from a consumer perspective. I'm not sure even if an easy way to create them existed that they'd be nearly as popular.

Where we are now is that the lock in is so wide that basically everyone is running Chrome, including its competitors.

When google announced the "Manifest v3" thing, which does impact ad blockers, people cried, but nothing happened.

Yes, the standards are theoretically open, but concretely nobody else can keep up, Google-the-server launches X and Google-the-client implements it, and than they propose standardizing it but move on anyway, and everybody else plays catchup.

It's not worse than closed source Flash, but it's huge lock in anyway.

That is for me the right way to frame it. Both flash and the web lost out to mobile. Neither managed to get a sufficient foothold there, and in practice people’s primary computing experience is now through fully proprietary apps distributed through vendor-controlled walled garden app stores.

The technical qualities of the underlying platforms didn’t matter all that much. Mobile platform vendors have control, and they made sure native apps made with their tools distributed through their store was the easier path. Apple could have improved safari’s progressive web apps ability to the point that it became a feasible alternative to making native iOS apps, but they didn’t, so people simply stopped trying to shoehorn web apps onto mobile platforms.

Apple didn’t kill flash to replace it with html5, they killed it to replace it with their own proprietary tech. I’m not sure they intended it that way, but they definitely intend to keep it that way (witness for example how web notifications are in safari on desktop but not mobile).

So from a user's standpoint, things have become better not worse – they don't have to compromise security or accept horrible battery life in doing so.

From the developer's standpoint, things have improved too – they don't have to depend on niche toolchains and they have better monetization options today than ever before. For small game studios it is easier to find mainstream developers who can do game development rather than only those willing to master niche toolchains.

The side-effect of using mainstream toolchains is it has more moving parts and hence has more complexity (that is unnecessary at a solo developer scale).

I agree that the learning curve has become harder a creative arts designer who wants to be a solo game developer to express their creativity.

So maybe the real question is what are the accessible mediums in which creative digital artists put out their hobby/amateur work today?

Battery life wasn't really a thing I worried about before smartphones came along, so I'm not sure that's super relevant (although, if you want to pick that fight, Unity on mobile is a real battery killer). Security was rough, but I think I'd trade a bit of security to ditch the walled gardens we have today.

I do think things are better in most ways for users. Because Flash portals didn't require a download, it was MUCH easier to try new things. This led to a virtuous cycle of people making weird shit. Because so much game content is gated behind app stores these days, you have to download a thing and then wade through permission/notification screens and it's too much of a cognitive burden compared to "oh, that looks neat." (Toss on that app stores are horrible for discovery). I think the way most people discover (mobile) games these days is through ads, and if a game is running ads they're a large company with plans to extract $5-$10+ out of you...

So we've traded big catalogues of free games with crappy banner ads for walled gardens that depend on ads for discovery of exploitative IAP skinner boxes.

I miss flash games and scene that existed around it. Including free aspect of it and ease of development of it. I do not care about complex games game studios put out.

It struck a perfect balance between static and dynamic typing, had all the OOP you wanted, but FP too.

Also, e4x and data binding in Flex was magical.

It could be a combination of both - the tooling for creating that stuff with just JS/HTML is not as good and the diminished demand prevents the creation of such a product.

The simple format made it easy to upload content both for the portal and the developer. The tools made it easy to get started and to grow. The funding made it easy to start, find your footing, and stick around long enough to get good and eventually great.

As the format was attacked and HTML5 didn't work well with the old tools, and as Steve Jobs pushed app stores forward, it all came crashing down.

I saw it happen in real time, these forces evolved together.

The internet is a much less interesting place without it, and the amount of ceremony required to replicate in HTML5 what any Tom, Dick or Harry could slap together with Flash is insane.

We have regressed here.

And then reflect on why there are zero tools for the great collection of technologies collectively called HTML5 that have Flash's timeline tools, independent keyframe animations etc. etc.

It's nonsense to say that "HTML5/JS has [...] gotten worse". It's not a vendor SDK where platform version X doesn't support feature from version because the platform hired a new head of the design team who wants to shake things up by ditching the previous gen for the new shiny that you have to use instead. Web platform standardization is additive. If the new stuff sucks, don't use it, and keep doing things the "old" way. Don't do dumb stuff like trying to stay on the bleeding edge for no good reason and then complain when you get cut.

It's also a total contradiction to double back after all of this and try to claim that HTML5/JS is "programmer-centric". To go further and say that "Flash was content-first" is something all on its own. Standard web tech is content-first in a way that Flash never was, or more accurately, it's user-first. Flash on the other hand was always creator-first, prioritizing authors'/designers' pixel-perfect vision over users and accessibility, and the tacit endorsement of the worldview that the ends justify the means and that it's okay to favor blobs over, you know, content—which is ironically exactly what's wrong with the current world of SPAs driven by inscrutable minified bundles on the order of megabytes just to do something like put half a dozen form controls on a page that still doesn't manage state correctly.

I suppose the point may be that no one has written a tool to package a HTML5 app up like that? Seems possible to build under certain constraints.

EDIT: I got downvoted for telling the truth? The website is here if you don't believe me: https://webpack.js.org/ It literally does what the post I replied to asked for.

I try to make a point of upvoting comments I disagree with, but that are clearly made in good faith in order to counteract this trend.

I suppose this problem is a normal low entropy state in community moderated systems.

It takes a lot more mental energy to moderate based on the value of the conversation rather than whether you personally agree.

As for size, browsers really should allow gzip encoding for html files from your desktop. Would pack all the inlined crap down to a reasonable size and then you could transfer things around better.

I remember a point where just about every time anyone logged into a XP machine, the Flash player would start begging to be updated. There seemed to be multiple updates a week sometimes to fix stuff. When Chrome started embedding Flash it was just about revolutionary since you didn’t have to keep as much of an eye on whatever CVE popped up that day.

4chan's /f/ would disagree, they're posting stand-alone flash games and animations just fine and those also worked locally.

The author's points hold true at least somewhat if you're talking about a single cohesive Flash application, but much less so if you're talking about a website built with flash.

Point I didn't mention clearly enough was that single-file-as-default was REALLY good for flash portal distribution. Made it easy for creators to share and easy for portals to ingest files, AND to "steal" them from eachother, which was something us developers actively encouraged, it's how you got your game or cartoon played/watched by millions of people.

And even if this had problems as you rightly suggest, for its time the alternatives were worse. Comparing via the benefit of hindsight in 2020 isn't really relevant to the point of the article.

Apple's ".app" distribution format works just like that, while allowing the developer to have a directory structure for their convenience. App stores solve the problem in other ways, but there is no good app store for the web.

Also as soon as you have more than one file you can run into path issues.

So why doesn’t that tool exist?

Well... as far as I know it does. I’m pretty sure this is exactly what Adobe Animate is meant to be:

https://www.adobe.com/products/animate.html

So why isn’t it causing a Flash-style revolution?

In my opinion... Because of YouTube. Twitter. Etc.

Before everyone had heaps of bandwidth, you needed to consume smaller files. A full 1080p video was unfathomable with the codecs, compute and network bandwidth available in the early 2000s. Swf files however were small and easy to distribute. The perfect storm occurred due to the structure of early internet, and communities like Newgrounds flourished.

I honestly think Flash just got a bit stale. It stopped being new and exciting, and for animations video sites were probably going to win out anyways. For games, a lot of indie devs now target PCs and game consoles. Unity and Construct export to HTML5. People wanted to do more advanced things to the point where Flash itself started doing things like 3D rendering, at which point the appealing aspects of Flash started to make less sense anyways.

Many seem to argue passionately, probably from an emotional place, that Flash was killed. Honestly, I just don’t agree. Flash was dying either way. Everyone just gradually decided to cut off the life support.

The era it ushered in will be missed, though.

The author doesn't argue a tool like Flash should exist. He just states the fact that HTML has still not caught up to fill the blank caused by Flash exiting yet, which is true IMO.

These arguments get confusing because HTML, SVG and JavaScript are human readable and writeable. But that doesn’t mean the way someone should make animations in HTML5 is by typing coordinates into a text editor.

If there really is a gap in the web runtime that Flash fills, what is it?

Edit: also, I believe they are directly and indirectly arguing for the existence of such a tool:

> 1) For 95% of applications you can just distribute a single SWF file

> 2) You have a robust authoring tool that is animation/graphics-first and newbie friendly

The first one is 100% technically possible with tooling; though admittedly better embedding of assets would be nice. The second one is just literally directly asking for such a tool.

The runtime and the tooling go together. The reason why tooling doesn't exist for HTML5, SVG, and Javascript is it's just so damn complex and fiddly that it's impossible to paper-over and abstract with a robust authoring tool. Both sides of Flash were designed for the ease of this kind of development.

There hasn't been a good high-level authoring tool for HTML in decades and the situation is getting worse, not better.

Indeed. The fact that Flash is proprietary meant that all browsers had to use the same software (up to versioning and OS API implementations) to run SWF files. Now with HTML5, developers now have to deal with differences between different browsers (Chromium/Safari/Firefox are the most popular ones).

It was very interestin. Among the vastly underrated things it had was the insanely fast drawing of vector shapes with insane things like blazingly fast flood fills. It was extremely resilient in the presence of self-intersecting shapes, shapes of almost any complexity, open plygons etc. etc.

The tooling seems to be what most people miss. That's not trivial. Adobe Animate is supposed to fill that void, but from what I've seen, it's a long way from a drop-in solution.

An authoring software just like Flash could exist for the HTML runtime -- it isn't the nature of HTML which is the problem. Thus it's not HTML which has to "catch up", but vendors of authoring software.

We don't live in that era anymore (and I sort of miss it :( ). We live in an era of SaaS and major platforms. For 90% of "custom webpages" wordpress covers it. For the 9% not covered, flash wouldn't really be a good fit. The loss is for that 1% of creative fun junk by hobbyist that's just missing all together. Flash made it easy for a novice to sculpt something. However, it didn't make it easy for a business to maintain something.

The “days of flash” included the rise to dominance of Facebook (Facebook games were largely, maybe entirely, Flash), and even moreso included the entire MySpace era of social media.

So, yeah, I'd question why Animate isn't good enough to fill Flash's role.

Yes, this is what I meant. Sorry, I didn't clarify this in the replied comment.

> it isn't the nature of HTML which is the problem

It is in some degrees. The same html/js code might run some differently in different browsers, from either behaviors or performance aspects.

On the other hand, Flash runtime only presents some small behavior differences on different OSes or with different CPU architectures. And for different OSes/browser combos with the same-power CPUs+GPUs, there is almost no performance differences.

Really? Last time I used Flash, a few years ago when it still mattered, it used 100% CPU just to play video on Linux. The Windows version didn't seem to have this problem and I deduced that Adobe just didn't care about Linux.

From the animators I've listened to, the answer is that Animate sucks to work with and is a royal pain in the butt, so they either use different tools or stick to older versions of Adobe or even Macromedia Flash/MX.

So ironically the designer no longer just asks 'is Flash available on this platform?' but instead 'which codecs are available on this browser and platform?'.

Can you give some examples of what's wrong with HTML5 video streaming and these implementation issues? There's a huge community of video engineers working to make streaming video better, and we'd love to hear feedback.

Piracy was pretty normal back then.

In truth, the fundamental capabilities of the modern web platform matches and typically exceeds anything that was offered by Flash. What these people are really lamenting is the ease of authorship that Flash provided to non-technical users.

Macromedia/Adobe provided a suite of powerful tools for authoring Flash animations and applications. There is no free equivalent available targeting HTML/JS. This is not a problem of capability in the platform technologies. It is because there is no economic incentive to make and give away such tools. What equivalents do exist in this sector cost money.

When it comes to Flash, the old adage, "if you're not paying for the product, you are the product" holds true. The Flash authorship tools were free because 3rd party authors creating Flash content were assisting in the promulgation of a proprietary platform that was gradually replacing open web standards. Had they succeeded, Macromedia/Adobe would inevitably have looked to squeeze and monetise their domination of the web.

Programmers are often accused of not understanding what Flash offered to artists. Maybe that's true, but I've not encountered many artists who conversely understand the above.

I commented to the same effect elsewhere, but I would take this a step farther. There is no alternative, free or paid, Open Source or proprietary, targeting any platform (even just iOS or Windows) that is as good as Flash was at rapid prototyping and accessible game creation.

To me, the focus on the web is really frustrating. Unity is not as good as Flash was for rapid prototyping. Gamemaker is not as good as Flash was for animation. It's not the platform's problem that literally no one has found a way to merge animation, artwork, and programming in the same way as Flash did.

It's not that there's something awful about the web and that it ruined the ability to make that tool. Literally no one for any platform for any motivation has ever replicated Flash even just as a content creation tool.

So why are we talking about the web at this point? WebGL, Canvas, and WASM are powerful enough to replace Flash. That is not the issue anymore. I think when people touted the web as a replacement for Flash, there was an expectation that some company or Open Source initiative somewhere was going to make effective tools and that never happened. And instead of talking about the fact that the entire games industry has been unable to replicate Flash, we're attacking the Open web instead.

I agree that Flash was prematurely killed online, I agree that HTML advocates were too dismissive about the problems. But we are far past that point. The web kept its end of the bargain, the APIs exist, but nobody else built the necessary tools. We should be talking game engines right now, not the web.

If your thesis is true that the web/html api could've been just as good as flash had there just been a suite of tools to create the content, then why doesn't one already exist? I think the true underlying issue is that the expectations of the users have grown - "simple" animations that can be created in the flash suite and replicated using a html suite cannot satisfy users any more.

I think that's the big question that these threads need to be asking, because it's not just that it doesn't exist for the web. It doesn't exist anywhere for any platform.

Coming in and just saying, 'HTML5 can't replicate Flash' -- not only is that inaccurate, it's unhelpful. We need to ask what modern developers need and why we seem to be unable to supply that. What are our modern tools missing? Would Flash still be relevant for indie development today? Do we need something more extensive or powerful? Do we need to rethink developer workflows? Is the problem funding, or finding a way to monetize, or is it a community problem where we just can't get indie devs to standardize on one tool?

Every thread where people complain about the loss of Flash could be a really productive conversation about what a modern animation engine would need and how we could build one. But instead it's just people re-litigating a 15-20 year old fight for no reason.

Because the APIs are there. Any company could build an animation engine today that output to any low-level language: C, Rust, whatever. And the output could be compiled to WASM and the pixel buffers could be rendered to a canvas. We could even do smart updates and partial re-renders and cache large operations on separate canvases. Or that frame compilation could happen behind the scenes in WASM and the image data could be spit out as a single array of pixels. We could do all of the optimizations we need.

But that tool doesn't exist. It doesn't exist for the web, or for iOS, or for Android, or for Windows, or for any modern console. Why? That's the productive question to ask, that's the question that will actually help us make progress towards replacing Flash. The only thing the web can provide is APIs and technology, at some point game devs need to stop complaining about the web in specific and start asking why our game dev tools in general aren't up to snuff.

There are basically two camp of people. Those who use Flash, and stating HTML5 not as good as Flash are coming from a user, designer, creative point of view. Its ease of use, from coding to authoring. The simplicity of getting something done. Like HyperCard.

The other camp are programmers, software and web developers who just saw Flash and HTML from a technical perspective. Of course you can replicate 99% of Flash capability, but that was not the point. Since the first camp sort of rely on the 2nd camp to provide tooling. And the 2nd camp doesn't understand or care what they are missing. Nothing is being done.

Classic Software consultant and development problem.

Again, if Flash no longer provide any value to Adobe, may be they could open source it?

HTML5 is fine. The problem is the lack of tooling. That's what's letting me down as a user, designer, creative person. Not the web, the lack of tools.

But HTML5 can not fix that problem. Only the games industry can fix that problem. HTML5 is never going to be a game creation IDE, it's just an API and a set of technologies. Companies like Epic, Adobe, Unity, etc... have to step in and develop game IDEs that are comparable to Flash.

And they just haven't. I don't know why they haven't, people should ask them that question.

There's this idea that web advocates don't understand that web tooling for games is bad, and that's just not the case. As web advocates we do understand the problem, but Mozilla is not in a realistic position to build everyone a game engine. It's not something that browser-manufactures can do. It's something the games industry has to decide to tackle. Calling out the web is counterproductive, we are doing everything that we can do to help with replacing Flash.

> Again, if Flash no longer provide any value to Adobe, may be they could open source it?

That would be great, but I wouldn't bet on it. The hope was originally that Adobe would migrate Flash to HTML5 canvas output. They never did until Flash was mostly dead, and from everything I've heard their new engines are kind of terrible to work with anyway.

The secondary hope was that they'd Open Source Flash, or at least help with emulating SWFs so people could continue to use old versions of Flash to publish to the web. That never happened either because... I don't know why. Maybe licensing issues, maybe Adobe just doesn't care about anyone in their user communities at all.

To see people giving tearful salutes to a company that literally just sat there and watched its product rot rather than do anything at all to help the community... it's just very frustrating. I was a Flash developer too. But I'm not mad at Open web advocates, there are plenty of other people/businesses for me to be mad at.

My 2 cents: the market for animation on the web collapsed due to changes in Youtube's algorithm, and Flash itself has been more or less superseded by game engines that can export to the web, 3d tools like MMD and game mods.

I don't think the problem is a lack of tools. It's more fundamental.

It's very hard for us to admit that something could just be a string of pure luck, like the brief moment where Flash, Newgrounds, and fighting stickmen were at the cultural epicenter.

In fact, we as engineers seem biased in the complete opposite direction when it comes to understanding social phenomena. Technical advantages and disadvantages do very little to explain the world and why people do things and why things catch on.

Whether or not HTML5 replicates "what mattered" about Flash does nothing to understand how Flash fell out of cultural hegemony because it's not what mattered about Flash. HTML5 can't do anything to roll back the clock and recreate the perfect storm responsible Flash culture.

I think Flash, like Javascript and Wordpress, was simply at the right place at the right time to take advantage of unique characteristics of the early internet that are very hard to quantify with confidence much less replicate or predict.

Yes, that's what we've been saying. And it was HTML5 boosters who said that when Flash died nothing of value would be lost, I'm pointing out their comparing a platform & authoring tools with a narrow tech stack.

And the Flash Authorship tools were absolutely NOT free. You had to pay hundreds of dollars for the Flash IDE (not that it stopped people from pirating it).

FlashDevelop was free, but that was not an Adobe product.

I could easily recreate my games in javascript today (or swift. or C#/unity. or whatever) from an "is this possible" point of view, but I think it's much harder to make money today. You could argue that's more due to the competition level and number of creators than Flash, but I think the siloed app stores are awful for the long tail and smaller developers.

That ramp-up of competition kind of happened even before Flash was dead-dead, though--the big portals existed in a time before Facebook games (which suck now but were amazing for awhile) and mobile games. It's wild now to see my kids (5 & 8) play a lot of online games on modern much-shittier-than-Kongregate-and-Newgrounds portals, but there's basically a short time period between "old enough to play games" and "old enough for their own phone or steam account" that they're in right now, and I suspect that audience isn't big enough to support the kind of innovation we saw before Facebook games (which were also made with Flash, for the most part!).

What? The Flash authorship tools were definitely not free. I remember Macromedia Studio used to cost nearly a grand. They were widely spread because, like Photoshop, Macromedia/Adobe didn't bother cracking down on piracy.

Flash can export to html/canvas iirc. I dont think the lack of tools is the reason why html5 didn't replace flash in terms of output of new games/animations.

I think one reason Flash had such prominence was their effective monopoly. Remember when you had to install a browser plugin to get Flash? They reached ubiquity, but it was basically an impossible job for anyone else to achieve. So if you wanted Flash-like content, Flash was your only real option.

Since HTML5 is based on open technologies, there are lots of competing tools on the market doing exactly what some people apparently lament are no longer available. They're out there if you look, and arguably it's a healthier marketplace since there's real competition - although the choice is more complicated now, since there isn't that one option everyone uses.

[1] https://www.construct.net

Flash was offering things that the normal web couldn't do (like video). It took a long time for the web to catch up nominally, and it's still lacking

Macromedia would still be selling Flash if there was an open standard that replaced swf and was equally performant

After all the web is now full of SPAs which are a very bloated version of flash pages with much less functionality

Flash MX was a paid tool.

Luckily browsers don't matter, and newgrounds has their own flash player you can launch content with.

I think a large part of the flash animation scene being as big as it was was simply that it was the only real way to do video at the time, and so it was popular because video is engaging, and animation was the way to do it when video was big, bandwidth was limited, and browser support was bad.

Now that YouTube exists and video hosting is free, bandwidth is plentiful, and when everyone has a smartphone camera and can shoot some video super easily, the cost of animation is much higher than the alternative. If you do want to put in the effort, you go where the eyes are and render the animation to video and upload to YouTube.

Web platform games very much did replace flash games over time. I'm not saying that it being bad isn't a part of it, of course, but I don't think it is the lone factor.

Tech affects community affects affordances affects distribution pipelines.

codepen?

github pages?

We all invested in responsive layouts and dropped the rich, absolute and hand made layouts of yesteryear.

So ... there's nothing like Newgrounds at its peak, then.

The requirement is _far_ beyond simple hosting. The creation process has to be 5 minutes to an animation from zero domain experience, and an hour or two from a game. Flash's tools were _that good_.

One more shout out to the vector graphics editor - the way you did boolean operations with shapes based on their color has not been bested since by any other editor.

Can you expand on this? How did it use colors to help us work with boolean operations on shapes?

It has automatically merged shapes of same color into one. So to add or remove to a shape, you just had to paint in the same (or a different) color.

For what it's worth, I've found it super confusing.

Adobe provided the initial resources (developers and hosting) to get the project into a stable state. The project is now managed by Apache.

The team solved a couple monumental challenges (converting AS3 to JavaScript, and putting a GUI framework on top of it).

https://royale.apache.org/

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.

What would have happened if Steve Jobs hadn't made an ultimatum on Flash? The iPhone might have taken a few more iterations, but eventually they would have figured out how to support flash well in their phones. We might even be in a place where processors are designed to efficiently execute flash (the new M1 has so many specialization this is not even reaching out).

Instead what we have today is a mess of things that still don't add up to technology we had a decade back, and we have really bad, inefficient code running everywhere trying their best to emulate what flash did.

In the end, the reality is that Jobs made a tech forecasting mistake assuming HTML5 (not js) will eventually catch up with flash. It never did. As poor forecasting decisions go this is fairly reasonable, but we cannot move forward if we continue insisting that this was the best decision even in retrospect.

I doubt it. Adobe tried to make a go with Flash on Android to spite Jobs and it was a train-wreck. While Android makers were willing to put giant batteries in to support sloppy code, Apple wasn't.

> The iPhone might have taken a few more iterations, but eventually they would have figured out how to support flash well in their phones.

Apple shipped a phone with barely a day's battery life and it remained that way for years as features and performance caught up. If they'd shipped Flash, it would have cooked the CPU on every site it visited ruining the limited battery life the phone had. This is one of those cases where shipping less made a better product.

> In the end, the reality is that Jobs made a tech forecasting mistake assuming HTML5 (not js) will eventually catch up with flash.

You are splitting hairs. Nobody has made a perfect technology prediction 10 years out.

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!

Adobe had zero F*s to give about anything not mainstream.

No one wanted to do it.

Flash run on PC's hundreds of times less powerful than your smartphone is right now. For the original iPhone, sure, but this resource argument is decades out of date now.

In the years after, all we've done is made HTML/JS/SWF as powerful as Flash and even more resource intensive.

The one ultimately responsible for Flash's death was Adobe, who neither felt it was worth investing in to overcome these issues, nor worth open sourcing to let the community give it a try.

What a loss.

For Adobe I give full marks for cajoñes and long term thinking, but I'm gonna have to deduct points for the closed nature of it all.

It is always difficult to try and second guess decisions which were made by other people behind closed doors. Particularly when you know the end outcome where the people who made the choices wouldn't have.

When Jobs wrote that memo, the iPhone wasn't remotely dominant. It had tiny market share on and Windows Mobile and Android were both very big threats to the iPhone. There could easily be an alternative history where Adobe wrote a really effective version of Flash and Apple was left out in the cold.

Fast forward 10 years and it all looks obvious, but I don't think it was a slam dunk back then.

But I lay the blame for that with Adobe. Why didn’t they just adapt the Flash application to output html and JavaScript? If you can compile modern JS to old with babbel I don’t see why they couldn’t have an interpreter to turn action script into JS and render everything to a canvas. They had a massive community who would have loved it. Instead they just kinda gave up.

As I recall, it tries to naively export your Flash graphics as HTML5 Canvas drawing commands, but they come out all glitchy because Flash's renderer just uses a fundamentally different model (planar maps vs. stacked rendering).

This probably is fixable, but I sincerely wonder whether Adobe still has the engineering skills to solve new problems like this. They've subisted for so long now by just adding new layers of janky, slow, webview-powered cloud-enhanced e-commerce integration to their existing products, that I wouldn't be surprised if most of the serious engineering talent has left.

It needs to be feature parity including the ‘runs anywhere’ of an swf. I don’t even think it’s too late if they could really deliver this.

Just look at the community surrounding Instagram filter development because they made really easy-to-use software (Spark AR studio) that kids and artists can use.

I wonder if something like figma/webflow, that allowed the WYSIWYG nature of flash, combined with an <embed> or web assembly output would be a solution?