Removing Flash from the web was done with good intentions. It was a propitiatory system plagued with security holes. Instead of replacing it with similarly capable open standards, we've got a propitiatory Encrypted Media Extensions (EME), flaky HTML5 calls that differ between browsers, and webassembly.
It's a shame that HTML5 never really reached the ability of flash. And it's also a shame that Adobe didn't simply release the swf spec.
There were a number of tools that compiled to swf.
It sucks that Adobe never opened the Flash Player. That was it’s death nail. They had little to gain from it. It was free. The money maker was Flash Studio and other Macromedia tools.
If Adobe had opened Flash and had a sane security Sandbox it would have survived iOS. Adobe was like the old Microsoft though. They just didn’t get the power of Open Source.
Flash died because of Apple, and before Adboe bought Macromedia, it was an incredibly small runtime; way better than Java at very specific forms of 3D rendering. Old Flash can be upscaled to 4k. A youtube video is pre-rendered and stuck at that resolution forever.
Hehe. Not really. Jobs may have put one nail in the coffin, but Adobe killed flash. And flash was already dying because of abusive ads and HTML5 before Apple said anything. Here’s one of the Flash developers backing that up: https://m.slashdot.org/story/324267
Can confirm. I was doing flash game development when flash "died." One of the guys I worked with created the Flex compiler. He told me Adobe had a seriously bad combination of brain drain and code ossification. The motto on the flash team was "don't break the internet."
For years, they were terrified of making any changes, especially ones that might break backward compatibility. This lack of ability to change anything meant their brightest never stuck around very long. This only made those who were left even more terrified to make any changes because the institutional knowledge was gone. It's a vicious cycle.
By the time they realized they had to make serious changes and started work on stage3d, they didn't have the expertise necessary to pull it off.
> For years, they were terrified of making any changes, especially ones that might break backward compatibility. This lack of ability to change anything meant their brightest never stuck around very long. This only made those who were left even more terrified to make any changes because the institutional knowledge was gone. It's a vicious cycle.
That's an interesting comment. I actually wonder whether this eventually happens for any successful product. At the end this gives competitors a chance to surpass the product. So it's definitely a cycle.
It happens to a lot of successful products over time, but it's not inevitable by any means. It's possible to make software that is viable long term. But, there are many ugly realities regarding the current state of Corporate America that encourages this cycle.
For example, the fact that most engineers need to job hop in order to get promotions and raises. That makes it nearly impossible to build up long term institutional knowledge at any one company.
Another example: everyone in a publicly held company, from the rank and file programmers to the CEO, are rewarded for short term sales at the expense of long term stability. We had a thread here not too long ago[0] about how being a "maintenance engineer" was career suicide.
I hear this a lot, but where was the Android or BlackBerry or Windows phone that ran Flash acceptably fast, and without weekly security holes?
Mobile killed Flash. It was a poorly implemented technology which was acceptable only during the era where electric power was unlimited, and people were in the habit of frequent reboots to update their OS anyway.
It's worth noting that the Flash platform originated on mobile devices, and pivoted to a Netscape Plugin when the mobile market proved to be premature (early nineties).
For those of you half-considering building a Flash player replacement, I suspect the majority of Flash content (especially pre-AS3) can run performantly on modern mobile browsers with a canvas Context2D shim. Maybe even on 2010-era mobile devices.
Windows CE and Ipaq. I created mobile tour apps for museums like the Louvre, LA County Museum, SFMOMA and the Whitney that successfully served tens of thousands of people daily in the mid/late 2000s.
"Flash died with good intentions" is false. Flash died because it was the perfect tool for making mobile content, if not for the fact that it was sluggish on early iphone hardware. If Apple worked with Adobe to optimize flash for mobile, it would still be alive and well today. Instead Jobs decided business wise he could lock devs into developing native apps in the iOS ecosystem, scapegoat performance and security for the ban, and gain a competitive advantage at the same time.
> If Apple worked with Adobe to optimize flash for mobile, it would still be alive and well today.
Why is it on Apple to save Adobe’a software? Why not Microsoft, BlackBerry, Google ... or Adobe?
> Instead Jobs decided business wise he could lock devs into developing native apps in the iOS ecosystem
iPhoneOS lacked support for Flash long before it had a native SDK. Jobs decided the answer was HTML, which is exactly the opposite of a locked-in ecosystem.
Why are you rewriting history? Apple didn't just decide not to embrace Flash, they actively decided to kill it. Apple banned flash based apps in their review guidelines and Adobe didn't see a way forward since Job's had decided everything but native apps were to be purged from the app store. Html 5 isn't a replacement for what flash offered, it wasn't then and it isn't now.
Here is the Jobs quote from when this happened: "He takes a while to get there, but eventually he puts the notion into a nutshell. "Flash is a cross platform development tool," he says. "It is not Adobe’s goal to help developers write the best iPhone, iPod and iPad apps. It is their goal to help developers write cross platform apps.""
""Applications must be originally written in Objective-C, C, C++, or JavaScript as executed by the iPhone OS WebKit engine, and only code written in C, C++, and Objective-C may compile and directly link against the Documented APIs"
You seem to be doing a little re-writing of history yourself right here:
> Instead Jobs decided business wise he could lock devs into developing native apps in the iOS ecosystem, scapegoat performance and security for the ban, and gain a competitive advantage at the same time
Steve Jobs never even envisioned the App Store model when he decided against supporting Flash in Safari.
>... prior to its unveiling in 2007, Apple's then-CEO Steve Jobs did not intend to let third-party developers build native apps for iOS, instead directing them to make web applications for the Safari web browser. However, backlash from developers prompted the company to reconsider, with Jobs announcing in October 2007 that Apple would have a software development kit available for developers by February 2008
Interesting tidbit: Apple was actually trying to work with Adobe to iron out Flash's security problems but was rebuffed.
Anyways, the fact is, the App Store was actually an afterthought.... An accident of history that even Steve didn't see coming. The decision to forego Flash came way before that.
>... prior to its unveiling in 2007, Apple's then-CEO Steve Jobs did not intend to let third-party developers build native apps for iOS, instead directing them to make web applications for the Safari web browser. However, backlash from developers prompted the company to reconsider, with Jobs announcing in October 2007 that Apple would have a software development kit available for developers by February 2008
Let me translate that:
Steve Jobs did not intend to let anyone else profit from iOS, instead directing them to make web applications for the Safari web browser that would provide subpar experience.
They released a publicly accessible SDK in less than 4 months? And you believe it wasn't in their plan at all to allow anyone else to develop on it? They clearly planned to have a way to sideload app on it. Itunes already existed since 2001, it wasn't anything new for them. The only things they decided to change was to allow everyone to access it.
Flash would have been a way to allow everyone to access much more powerful feature. This is why they avoided it. The app store was an alternative that allowed them to keep control of it.
I'm referring to were jobs went on a 1600 word rant about flash while not once mentioning the shortcomings of his own objective-c authoring environments and the many good things about flash that lead to the enormous community it had.
https://www.apple.com/hotnews/thoughts-on-flash/
This combined with updated app review guidelines banning flash on iOS accelerated the adoption of html5.
Afterthought or not, apps quickly became the biggest selling point of smartphones from the second gen onward and many technical architecture decisions had the intended effect of locking developers in. Ports to android were often delayed by months and years if they happened at all.
You accused someone of "trying to rewrite history", when what they said was factually correct.
It's also been established that Apple tried to work with Adobe on addressing the issues it saw with Flash but was rebuffed.
The issue of Apple forbidding non-native code is one I disagreed with but has many layers and complexities. It's not nearly as simple as your implication that Jobs seemingly woke up one day and decided to kill Flash for no good reason.
> rant about flash while not once mentioning the shortcomings of his own objective-c authoring environments and the many good things about flash that lead to the enormous community it had
I'm just going to re-phrase what the poster above said - Why is it on Apple to evangelize Adobe's software (and criticize it's own)? That just doesn't make any sense.
In case you missed it, he was discussing Flash in the context of a mobile device. There is simply nothing positive to say about it. Nothing.
Everything he did mention, from power consumption to terrible security and poor usability has since been proven true. I don't know of a single reputable technologist, or tech journalist, who would dispute that. Do you?
Here's a write up of why Flash failed on Android. Notice any familiar themes?
I can tell you've never used the flash authoring tool or you wouldn't be making snarky remarks about this. Nobody is disputing that the runtime had problems and it was never good enough for prime-time on mobile on older devices. But a blanket ban on Flash was anti-competitive behavior that means Adobe never got the chance to prove it could work.
>Adobe never got the chance to prove it could work
Wait, what? They had years to demonstrate it working well on Android. They failed. Spectacularly. Why are you ignoring widely accepted fact of tech history? Are you implying that is also Apple's fault?
Look, if you want to discuss Apple being anti-competitive I'm all ears. I more or less agree. But using Flash as an example does nothing to help your argument. That's because everything Apple claimed about why Flash wouldn't work on mobile was later proven to be true. Everything.
Flash on Android was a disaster. Ignoring that fact to claim it could have worked on iOS (because... magic?) is nothing but revisionist anti-Apple zealotry.
Never let facts get in the way of a good Apple-bashing I guess.
Flash performance was always terrible on all platforms. It span up the CPU on my windows computer from the beginning and never stopped until its eventual demise.
That's from bad programming practices. It was very easy to write a terrible inefficient program with it since it was so accessible. There was nothing inherent in Flash that used a lot of CPU. The VM was pretty light weight. It also had only a software renderer for most of its lifetime which would naturally struggle with lots of crazy overlapping effects.
HTML5 video had significantly better performance in terms of CPU from the beginning, and I'm fairly certain the Flash video players weren't written by amateur programmers with bad programming practices.
Except that it can. Youtube lets producers (channels) post polls, let's consumers (viewers) post comments, and if channels want to, they can do something interactive- over a video series & etc.
Youtube has no interactivity within videos. There used to be a little bit of it with annotations, but those were killed off ages ago.
Compare that to Flash video series like Homestar Runner, every episode had little Easter eggs where you could click on background characters to unlock a bonus scene about them. It really helped you feel engaged with the medium in a way unmatched by anything but full-length video games.
Sure it didn't make much sense in the HTML5 world, but if Adobe hadn't killed the upcoming new version of the language and engine (I think it was called Flash Next) it would have been an amazing tool for making apps and games.
I wasn’t a big Flash user back in the day but I think our product Construct 3 (https://www.construct.net) can do just about everything Flash did, especially with the addition of the Timeline editor and scripting we’ve just added. I’m a little hesitant to overtly market Construct as the “next flash” as that I think would receive a mixed reception depending on your experiences with it!
Tom Fulp from Newgrounds has nice things to say about Construct, and we’re getting very positive numbers from other portals and platforms (some very big) that a lot of volume of games and good quality ones at that are being published with them using Construct 3.
I’d like to hear from more experienced Flash users if there’s anything Construct 3 can’t do that Flash could as it would likely be an important feature we’d be keen to add!
It needs to be able to name a platform, media type, IDE, all at once. E.g., "I'm going to use Flash", "It's a Flash game", "Hold on, let me animate it in Flash". The name should be able to go in domain names; e.g., "superfunflashgames.com".
Basically, it has to have a name amiable to becoming an umbrella term. That gave Flash such a "universal" connotation.
I hope this makes sense - trying to convey the _peculiarities_ of "Flash" which made it (as I see it) successfull and distinct in character.
Flash was fundamentally just a file format. You made the format using the program in the same way you make a Photoshop file in Photoshop. It was a noun and a verb because you make thing and both the part that makes and the part made are called the same.
Due to the nature of present browser APIs you cannot name your editor WebGL or something that produces WebGL. But if you made an editor called Sparkle or something that produced .sparkle files that had native browser support you would have the same nomenclature going on. But native browser integration and containerized file formats with external editors aren't a well made match.
A game development camp I used to teach at here in Austin, Game World's, uses Construct and it is a fantastic tool that's readily accessible, even to children. Once you get them going with game design concepts, they can really knock out a solid game within a week. I think it's a fantastic platform and seems comparable to flash in ease of use.
With the timelines editor theoretically yes, although there's no good easy way to export in a single shareable file (some technical hurdles). Definitely something it's lacking right now.
This comment is pretty misleading. The parts you complain about have little to do with what made Flash a creative sandbox. The parts that do matter have long been replaced with Canvas/WebGL. But the web as a whole has changed. Just like we don't have as many fun quirky geocities-like websites. The issue isn't the lack of Flash, it's that the web as a whole has grown up and changed.
Canvas/WebGL may replace Flash to end users. I'm not aware of creator tools that allow people to put things together as easily as they could using Flash.
Literally everything about animate.
It is hugely bloated and awful to use and crash prone. It takes upwards of 30 seconds to save a file. It has two different kinds of tween to support backwards compatibility. And the "HTML" export is a garbage fire.
Flash used a form of vector graphics that were incompatible with svgs.
Adobe has been trying to maintain backwards compatibility with flash (because most studios were still using the last version macromedia put out over 10 years ago) So adobe animate was always doomed to try and please everyone and end up pissing everyone off.
It's also nowhere near as easy to distribute these games compared to Flash. All you needed with Flash was the swf and as long as you had a flash player then it would work.
What? Everyone these days has a browser. How is it more difficult to share a web game than sharing a flash game? IF anything, back then you often had to install macromedia and other plugins since they didn't always come by default on every computer.
Discovery. Discovery. Discovery.
Sure you can host it on your personal site but who is gonna find it? It has to be on a 3rd party website and that 3rd party website wants some way to monetize it (by showing ads). It's harder to manage many files than one file that a VM can run as is.
Also Flash had something like 98-99% installation base during like 2000-2010 so it was very much not a problem. Flash had a much larger market share than any one browser.
The reason discovery is down isn't because they're hard to find, it's because there's much more interesting "entertainment" on the web these days, such as Youtube, Netflix, Reddit, and other time sinks. Back in the days, Flash games and geocities were the only forms of entertainment, which is why they were so big. Nowadays, we've got amazing free content and games everywhere, which is why this kind of entertainment isn't as sought after anymore. It's not lack of discovery, it's lack of demand.
eh, no it hasnt, and even if it did it would have nothing to do with flash. Flash was a great domain-specific tool for somthing that is significantly more cumbersome to write (and difficult to debug, and less wysiwyg) in canvas. Plus technology is supposed to move forward and 20 years is a geologic timespan in internet years. We expected to have much, much better tools by now, not a barely equivalent
Again, the issue isn't the lack of tools. It's the lack of interest. Making games is easier than ever with Unity, yet people don't spend all day playing random small Unity games. We have limited time, and as the types of entertainment increase, we have less and less to spend on small silly things like Flash games.
Back then, a Flash game was the epitome of web entertainment. Now, we have thousands of amazing Youtube videos, Netflix originals, memes, and so on. If you were to give people the same flash games now, they'd find them ridiculously lame.
Farmville and Candy Crush were literally flash games. Also many genres like physics games were born from flash.
On Unity I agree. It's never going to have 50kb games that download and start instantly. It's also not particularly easy or good for working with 2D games. I've used both professionally.
But it gets really pedantic arguing the semantics of web flash - fundamentally it was only showing on a web page because you installed the browser plugin or it. Today browsers might have stopped supporting arbitrary library plugins like that but now you can write them in webassembly for the same ends. Its just that nobody has actually gone and done it yet.
Except no one plays Farmville anymore, and as Candycrush implied, most flash games were replaced by mobile games. My point was that people are finding their entertainment elsewhere (mobile, youtube, reddit), which is the real cause of Flash's demise (and similar content).
Unity impose hardware constraints that are difficult to overcome in poorer places too. I have an old Dell at home and I won't even try to install Unity on it because it struggles to Windows 10. It's a perfect fine machine for coding with a simple Linux distro.
>Just like we don't have as many fun quirky geocities-like websites. The issue isn't the lack of Flash, it's that the web as a whole has grown up and changed.
We have tons of memes uploaded each day, with crude animations, that could have been 10x easier with flash...
Is there really a way to do DRM that doesn't involve proprietary pieces somewhere along the way?
I think the modern browser is technologically superior to what and how Flash did things (with the exception of native sockets, which is probably for the better). It's really everyone using Adobe's content creation tools (which were well integrated/designed) that made things so low friction.
It took a long time for the browser technologies to catch up but I wonder if in the end it integrates as one system better than bolting on an open flash spec would have been.
Do you think we really need DRM, though? I've been using the web without it for a while, and I just boycott sites or services that require it.
More to the point, I think that the current DRM scheme is inherently and completely broken. I could use it and let websites pretend it's useful, but it's not worth the increased quantity of proprietary code to me.
I don't like DRM myself but there are a very large number of users that like to use services which wouldn't stream content otherwise. Just like any protection mechanism it doesn't have to be perfectly impossible to break to provide value to someone.
The nice thing about the way EME was implemented is if you aren't one of those users you can simply disable Widevine or (if you prefer to not even have proprietary code on your computer) use a build that doesn't include it.
DRM does exactly what it was meant to do: remove rights from the folks who play by the rules.
The public premise that DRM is about stopping "piracy" was always a lie. Obviously, it can't do that. In fact, DRM encourages piracy by reducing the value of non-pirated versions of things.
DRM exists to control and manipulate regular, honest customers, such as making them watch ads or to pay over and over for different copies of the same thing even when they actually have the legal right to make copies for their own use.
Yes, because enough content hosts absolutely refuse to make their content available without DRM, and enough users are willing to switch browsers to one that will show that content, that even Mozilla finally gave in after years of fighting against it. In the end they were afraid that they'd lose enough users that they wouldn't be able to continue the project any more. https://hacks.mozilla.org/2014/05/reconciling-mozillas-missi...
>It's really everyone using Adobe's content creation tools (which were well integrated/designed) that made things so low friction.
I really feel like flash's heyday was before adobe bought it up. Most of the best flash sites, newgrounds and stuff were more active and had tons of content while Macromedia still owned flash.
>I think the modern browser is technologically superior to what and how Flash did things
Depends on what you mean by superior. There are still things Flash could do and modern “open web” can’t, determining keyboard state (caps lock state and current input language) is one example.
I feel like you can do pretty much everything with HTML5 and JavaScript that you could with Flash and ActionScript. The IDE for Flash had some nice UI abstractions but there's no reason that couldn't happen in a web IDE in 2019.
Distributing your website like a flash file/app as a binary is one missing feature I suppose?
As consumers, most people probably only care about what modern browsers can do. As a developer, the mechanisms become front and center. Modern web development is a Rube Goldberg kluge of things glued together with duct tape. Flash was much more coherent and imposed less cognitive burden of tying it altogether.
Not just NPM - the "web" is a diverse set of APIs with various peculiarities. It's not inviting to an artist with little to no programming experience (and even with the potential to learn). That was the advantage/disadvantage of flash - closed system, one (major) IDE, one language/standard library.
I'd need to see some performance metrics to come to that conclusion. "Bloated Flash site loading with a progress indicator for a few seconds" was pretty common and a huge complaint about Flash and its use on the web.
Nothing, of course. Anything that could be done with Flash in its heyday can be done better (in several senses) with modern tools.
But that’s not what we’re talking about, is it? We can ask what’s actually being done in web design these days… and who’s doing it. If you’re old enough to remember the period from the late ’90s to the mid aughts, think about your friends who were putting all kinds of wacky, creative pages on the web. Picture that personality type. What place does it have in the universe of websites that we visit today?
The answer, as far as I’ve seen: user. We spend time on a smaller number of sites/platforms. The design of those sites tends to be (at its best) streamlined and optimized. And people who want to share their creativity are given little frames within that structure.
Even personal websites are usually built through WordPress, Squarespace, or similar. Something has been lost. I’m not going to say things are worse nowadays. That would be too sweeping a judgment, even if I substantially agreed with it, and I think I don’t. But I get this weird feeling when I spend time online, like it’s a tree that died some time ago, but it’s being reinforced, made to stand straighter than ever, given artificial leaves that never turn, etc.
Talking about this to certain devs is like talking to them about sports. They just don't get the learning curve was the most accessable for the average user, not just those who know how to program.
We know there are better technologies security wise and performance wise. It's not always about what is 'technically' better rather than something that people enjoyed using and opened the door to creative personalities for the web. There is a time and a place for both and I think there has been neglect for the other side of the brain.
> can be done better (in several senses) with modern tools.
contestable. html5 games can be choppy in ways flash was not. even streaming video was pretty fast, comparing it to something like webrtc multi-party video. Hell even basic text-based chat like slack is a resource hog in html.
IMO, nothing. What HTML 5 (and more specifically, canvas+js) are lacking is a complete timeframe-based IDE.
I personally find that mode of development distasteful, and I don’t share the author’s affection for this particular breed of "creativity," but anyway it was there and some folks liked it and there’s a real opportunity for someone who wants to reimagine it.
I do not have experience with pure Flash but with Flex4 that is a powerful GUI for applications. I can share here what I miss from those days:
- you had types, performance was important so you had generic types like Vector , the code was compiled and it was really fast
- there were built-in efficient GUI widgets, you could have a list or table connected to a data provider with tousends of items but the components were smart and only created enough UI widgets as needed to render, where in HTML you either have to create a LI element for all your 10000 elements, create pagination(is not built in), create infinite scrolling(not built-in)
- it was super easy to customize or extend the provided things if you needed , as an example in HTML look at the <select> it is not customizable, everyone has to implement it's own using divs, css and js , but everyone gets it wrong in the corner case(you have issues where the dropdown pops outside the screen, missing events, hell even youtube search box sometimes breaks and the dropdown popup remains stuck open, so even Google engineerscan't create a working dropdown component
- performance, remember when a blinking cursor was causing an electron editor to use a lot of resources , the issue with web is that was designed in rush for showing documents.
- layouting components was super easy , theming was easy - with css in large old projects there are so many issues( I know people use less but this is not standard )
I have some solutions too, browsers need to provide more powerful components, if we want to do RIAs or desktop apps then we need the same components that desktop toolkits offer with the same customization. Other issue I seen was in a project where the devs ahd to import a third party library because the scrollbars can't be themed in all browsers , I am not refering to the right side of the browser ones(I know sme people hate if you change those) but scrollbars on a list or dropdown component. So you have projects where you import 1 lib for scrollbars, 1 lib for modals, 1 for fancy buttons and inputs, 1 for drag and drop, 1 for calendar widgets, 1 for dropdown that allows you to have icon + text, and of course a framework like angular or react . Browsers should check what developers need the most and add that as standard( ex better dropdowns) you would still have the ability to make your own shit and use your preferred framework
Low barrier to entry, cross browser compatible, multimedia websites.
Someone with artistic talent could quickly learn flash and get something bizarre working very quickly. Drop the minimally skilled into a modern development environment and novelty is much harder. 95% of the time that’s probably a good thing, but it sucked a lot of the fun out of the internet.
When flash was dominant there was an abundance of vector animations. There isn’t a technical reason this couldn’t still be the case, however we’re in the graphics dark ages right now. Everything is a raster. Everything that isn’t is static or on a very light fixed loop that takes an absurd amount of resources to animate.
The important bits of Flash were not so much the capabilities of the player but the tooling to be creative. Flash found so many niches from animation, games, applications, being embedded for UI rendering and on because it provided an agnostic tool to make all these things in, really easily.
In terms of capabilities though there was much less worry about compatibility.
animated vector graphics? Browsers only have data-intensive pixel graphics codecs. Also, the vector graphics support that they do have is pretty limited and slow. Anyways, I'm glad that flash is dead, because it was proprietary software and constant source for security bugs.
Tooling. I don't think anything fully replaced it. OTOH, while not web things per se, I think Twine, Unity and RPG Maker scratch similar itches for many (but definitely not all) people potentially interested.
ensuring that the user sees the exact same thing on any browser / operating system combination ?
Just the font rendering differences between windows / mac make it impossible to create text that is supposed to fit perfectly in a given box because hinting differences may mean that you are a few pixels longer which can change the layout of a paragraph, etc...
with flash, there is only one single way things look across all platforms.
A consistent display was a nice thing to count on when making little games, applets and ads. However for general web layout it made accessibility features difficult to impossible to implement.
I'd also add that behaviour wasn't always consistent. There would still be OS/environment specific bugs that would crop up from time to time. (Though granted, these would rarely be related to rendering or layout)
Flash made a lot of things related to vector graphics more accessible. The vector engine was quite powerful, although lacking proper hardware support, it was quite CPU hungry.
Still, I don't think we have an adequate replacement, even it would be technically possible, I don't know about similar tools for vector animations..
Note they are nitpicking spelling (propitiatory vs proprietary) not discussing the topic. It took me a second to realize as well but to be fair you should have noticed if you read their link before responding.
It's a shame that HTML5 never really reached the ability of flash. And it's also a shame that Adobe didn't simply release the swf spec.