Hacker News Reader

Comment by echelon | Hacker News Reader

> Flash as a technology should never be used now unless for artistic purposes.

Nothing filled the gap left by Flash. It was easy, accessible, and drew so many young people to create.

Flash had security issues, but they could have been fixed. Macromedia and/or Adobe could (and should) have published an open source standard and realized their tooling would always be ahead of the game. They didn't see the incredible opportunity and dropped the ball.

Flash was so much simpler than the Javascript "standards" layer cake we have today. I'm not deriding the web -- it's the last free and open space we have left in an ever-eroding tech land grab. But Flash missed a chance to contribute to its defense.

If Flash had been in decent shape at the time smart phones arrived, Flash support might have been a required feature. Lots of people would have launched apps as Flash apps instead of iOS- or Android-specific apps, and we might have gotten much more open and cross-platform devices as a consequence.

I kind of blame Flash's lack of follow-through for the App Store hell we have today.

The web that could have been...

View on HN

The mid 2000s era was a really magical time for me growing up.

Just the sheer amount of interesting games and videos available on places like NewGrounds and Kongregate was something else. It was kind of crazy how easy it was at the time to get at least some attention to any proof of concept project simply by posting onto said sites.

Something that has definitely gotten much harder over time despite the more advanced tools that are available to hobbyists these days.

I think it was a mix of how easy flash made it to go from some vector art drawn in app, to hooking up some key-frames together with logic for a game etc... But it was also IMO strongly influenced by the internet culture at time, where there was (at least it felt like) a larger audience of people available to try out wacky often no/low-budget things.

I haven't quite seen the same environment emerge on places Itch.IO and to an extent Steam.

I think today's creative youth are on Twitch and TikTok.

I'm not sure if they're building the same kind of future capital and skill sets that we did, but maybe it'll turn out well for them. Especially if there's an explosion in the creative economy.

Still, it's incredibly hard to beat a technology that teaches kids to program as a side effect. Flash and early web raised up a generation of engineers.

Seems reasonable that being a content creator is a viable career for young people as automation and globalization continue to eat jobs.

TikTok is probably teaching vital skills for their futures.

Content creator is a vanishingly small lottery winner occupation now and in the future unless it’s connected to a product in some way.

How much of every corporate job is communicating / consensus vs doing?.

Arguably "content creation" is core to every > middle class job now - you have to get people onboard to get things done. Personally I still think writing is more valuable but perhaps the knowledge creation and sharing (_inside_ companies) of tomorrow will be done by making and watching videos.

I very much do not want to be part of that future, but I recognise it a likely possibility.

>will be done by making and watching videos.

Arguably, it's the case today to some degree with company meetings/team meetings/etc. typically recorded for people who couldn't attend live. But it's pretty low bandwidth for most purposes to communicate basic facts in the absence of interaction. I for one rarely listen to recordings of meetings I miss and I assume I'm pretty typical. For many purposes I'd much prefer a 1 page email.

It's like all the how-tos on the Internet that don't really have a visual component but are a video anyway.

> How much of every corporate job is communicating / consensus vs doing?

From experience as the corporation gets larger communicating and communicating-adjacent activities approach 90-95% of the time. The remainder is doing.

'Communicating' and 'performing' are two different skill sets. I believe on TikTok, you are doing the latter.

"Content creators" almost universally only make significant money when the content they create are ads or porn. It's also about as viable a career choice as "becoming a pop star" was a while ago.

I sponsor a couple of folks on Patreon and Github, and none of them create porn.

I think there's a future where long tail art can be funded on an artisanal level by patrons, and the network will connect consumers to artists that satisfy their interests.

Are you interested in an alternate history Oregon Trail where an alien invasion forces westward migration? Or perhaps a time travelling Jacques Cousteau serial where the protagonist searches for love across the ages?

Somebody will make something weird that satisfies your interests and you will pay them directly. And it'll be better than canned Star Wars spam. It'll be real, and the creator will know you by name.

ML and automation will bring down the cost of content creation dramatically and make it look and feel more compelling than the studio stuff we get today.

Same for me. It's much more natural for me to pay an actual person ten bucks a month for what they do then to subscribe to a streaming service for the same amount of money. It's nice to know one's contribution actually makes an impact to someone whom I appreciate, that it's not just a drop in a gigantic bucket like a Netflix or Prime or Disney+ subscription.

> I sponsor a couple of folks on Patreon and Github, and none of them create porn.

...that you know about. :)

> "Content creators" almost universally only make significant money when the content they create are ads or porn.

Don't go into creating porn with that mindset, it's not a guarantee you'll make significant money. It's still better than creating advertisement though.

That's like saying in 2013 that being a Soundcloud rapper was going to be a viable career choice.

0.1 per cent of those on TikTok, or any other sharecropping platform will be able to extend their online persona into something that makes money.

Content creation is an elite

Only a few make a living out of it

And there's a limit on the amount of tik tok like videos that will get attention, once everybody will start making them

Comparing the creativity of youngsters doing voice-overs in TikTok to that of early 2000's teenagers creating full-fledged games and entire films with Flash is... I don't know.

As you said in your last paragraph, Flash _raised up_ a generation of engineers by _lowering_ the entry barriers to a minimum. TikTok is just a showcase for attw's.

Don't be an old curmudgeon. :) There were a couple of absolute gems in the Flash era but there was an absolute boatload of crap as well. The most creative TikTok-ers are absolutely innovating storytelling and filmography just as the most creative Flash creators were innovating in their own medium.

It's just that when we look back we remember the gems but not the river of crap, but when we look at the present we see the river too.

Where was the majority of user and creator attention dedicated to on the 2 platforms though?

> There were a couple of absolute gems in the Flash era but there was an absolute boatload of crap as well.

Don't 90% of Youtube videos get less than a thousand views [0]? There are millions of videos on TikTok, I'd imagine the algorithm results in the same 80-20 ish sort of balance.

Yes, there were bad Flash projects. But just like everything else, we remember the good ones.

[0]: https://9to5google.com/2020/08/10/almost-90-of-all-uploaded-...

Personally I can't help but think that this is our fault, as an industry, for abandoning the idea that users should be able to create programs. In the 8-bit micro era computers booted into a programming environment, then we had things like QBasic coming standard with the OS, then things like VB, HyperCard, Delphi, and to a lesser extent Excel and Access. For a while, enabling users to create their own tools to suit their needs was considered valuable in personal computing.

That's not how we do things anymore. Now everything is an appliance and the user is cattle as far as the developer is concerned.

I don't think it was the developers who made that call though.

I think it was. Talk about how horrible it is to pile so many abstraction layers together and developers will come out of the woodwork to talk about how much time it saves them. In a lot of cases, developers are rewriting things over and over again in the latest framework-du-jour simply to pad their resumes when they abandon their current employer in two years. Then they turn around and put all the blame on other people like children.

To expand on this, developer culture is to view good human factors almost as a badge of shame. If you press the issue with "this is a horrible experience that gets in your way in a thousand tiny and not so tiny ways", the response is almost always "if you can't handle tiny inconvenience #386, you weren't cut out to be a programmer". Witness the sheer scorn heaped on GUIs.

It's a kind of gatekeeping behavior - "I suffered through memorizing a bunch of git commands by rote, and now you have to as well."

here's some nuance: there's plenty of kids showing off programming stuff/tricks/apps on tiktok and plenty of kids getting inspired to try something themselves because of it. It is the very fact that it seems 'doable' and fun/cool that way.

I was inspired by my uncle’s copy of Flash 5 and his little projects. That eventually prompted my downloading of Game Maker 6 and lead directly to my being a software engineer.

It's definitely not quite the same thing, but I think some of that on-ramp is happening in places like Minecraft and Roblox, and in creation environments like PICO-8.

Macromedia and/or Adobe could (and should) have published an open source standard

They did open it up...

https://en.wikipedia.org/wiki/SWF#Published_specifications

Until May 1, 2008, implementing software that plays SWF was disallowed by the specification's license.[35] On that date, as part of its Open Screen Project, Adobe dropped all such restrictions on the SWF and FLV formats.

https://www.adobe.com/content/dam/acom/en/devnet/pdf/swf-fil...

If Flash had been in decent shape at the time smart phones arrived, they might have been a required feature.

...but the feature bloat that Adobe created with the platform was part of the problem, introducing not only security issues but inefficiency. SWF started out as a small and efficient vector format, and then accrued far more features than were really necessary for its core purpose.

(HTML+JS+CSS is even less efficient, but have somehow taken over thanks to Google and the like.)

> (HTML+JS+CSS is even less efficient, but have somehow taken over thanks to Google and the like.)

Adobe themselves pushed for using open web tech. I was working at Adobe in AS3/Flex during the transition and they were going to great lengths to promote and encourage development of open standards.

You are romanticizing Flash. One legitimate reason for it's eventual demise was due to the fact that it was very power hungry to run. On laptops that was bad, and on smartphones it would have been a non-starter. Though curiously, I remember that some smartphones did used to come with Flash like the Blackberry Z30 which had it through 10.2 and then it was taken away out of the hands of the device owners in 10.3

The Flash runtime was not power hungry. It was the terrible code that was written that ran on it that was power hungry.

As HTML5 was emerging someone posted a HTML 5 demo on hacker news. It put 1 core of my machine to 80%. I wrote an equivalent version in Flash and it used 5%.

All the people writing terrible code in Flash simply started writing terrible code in Java script instead.

I never liked Flash, but the browsers were very far behind initially. Animation was slow and stuttering.

At some point the browser vendors started addressing that, and things got better.

But I think the other part of Flash is the tools used to create it - I never understood why they didn't port those to HTML5 and worked with the browser vendors to get kinks ironed out, instead of doubling down on - what was the name - Adobe Air? I think they still had some market power they could have wielded.

Abobe's failure to switch the flash tools to targeting HTML5 is one of the most inexplicable decisions of all time.

In late 2000s they owned the Web App space ( what was called Rich Internet Applications in their speak). Flex (their business GUI components library and markup language) was absolutely dominating corporate Web app development.

And they just abandoned it.

In an alternative reality we are all building HTML5 Web apps with Flex Builder right now.

Sorry I don't understand what is the connection between HTML5 and Flash? Isn't flash an object that runs inside HTML5? What would you like Adobe to have done exactly?

Instead of outputting a flash object file the Flash tools could have output actual HTML.

The Flash Player is, at the end of the day, a proprietary Canvas widget with a ECMAScript based VM powering it. With HTML5 they could have targetted the actual real HTML5 Canvas and completely eliminated the flash player plug in whilst still retaining the tools.

> All the people writing terrible code in Flash simply started writing terrible code in Java script instead.

And then some genius decided "What if we ran JavaScript on the server?" and Node was born.

Then came the node_modules black hole and the left-pad debacle.

Sometimes I think JavaScript was a mistake.

The things that were trying to replace it (i.e. web standards) were even more power hungry to run or didn't work.

For most of Flash's lifespan, laptops and smartphones were not really a thing.

Smartphones maybe but laptops?

Flash came out in the mid 90s. I'd say there was a decade before they became practical and cheap enough to outnumber desktops

Flash grew up in a very weird time where nobody cared about hardware acceleration at all. Go back just a few years and even basic multimedia tasks like playing music or decoding a JPEG require some amount of accelerator support. Go forward a few years and hardware accelerators are suddenly crucial to energy efficiency. Ergo, everything in Flash Player is done in software. Vector rendering & compositing? Software. Video decoding & playback? Software. This wasn't at all exclusive to Flash, of course, but Adobe was uniquely unsuited to realize this problem.

Apple realized very early on that you needed hardware acceleration for common computing tasks in order to have laptops that last more than an hour on a single charge. It took Adobe several years after the launch of the iPhone to actually support some amount of hardware accelerated video decode, and as far as I'm aware vector rendering (the core of the platform) was never meaningfully accelerated. As far as I'm aware, pre-iPhone efforts to get onto phones involved a special cut-down Flash runtime that was perpetually out-of-date and was only really popular on early-2000s Japanese handsets. This wasn't at all built for the "computer in your pocket" that Jobs wanted.

Plenty of Android smartphones ran Flash. I remember browsing the desktop (Flash) version of Youtube on my LG phone in 2009. Flash support continued for at least a few years, but once HTML5 video became common, it dropped off.

Yes, there is nothing like the flash IDE available today.

I had a manager in the late 2000's who was giddy with the possibilities of HTML5 "It can do everything we do with flash!" Whilst that may be technically true, in practice bloated libraries like D3 are required for even the most basic vector manipulation and coding every tiny detail of an animation feels like groping in the dark when there used to be a source of illumination.

I'm not sad about the demise of Flash, but I think there is a gap in the market for a more graphical keyframe based approach, where things can be rapidly prototyped, resulting in clean code.

I'm using Tumult Hype, it's SVG animation capabilities are pretty good IMO, and it's all GUI-based.

OK cool, will give it a shot, thanks for the recommendation!

I tested every application I could find a couple of years ago. Lots of pretty prototyping apps, but generally the end code is somehow garbled or proprietary.

> I'm not deriding the web -- it's the last free and open space we have left in an ever-eroding tech land grab

No it isn't. It has degenerated into being only browsable in its entirety by Chrome, with all other "browser vendors" except Apple having given up due to sheer complexity and infeasibility to keep up - Opera, Microsoft, and now Firefox. That's what you get by putting Google (ie WHATWG, consisting of Chrome devs) into the role of defining so-called "web standards", with W3C (a self-declared "standardization body") depending on Google's and other big player's money having done nothing meaningful for years except brittling and foobaring CSS.

The Web is gone, ChromeOS has won.

Remember that Jobs banned it to kill it.

https://www.cnet.com/news/jobs-why-apple-banned-flash-from-t...

I don't think this is the fatal reason Flash is stopped. Flash is only forbidden in mobile web browsers.

Flash is mainly self-cancelled by Adobe itself. It looks they couldn't get expected revenue from this product. However, I think they are short-sighted at this point.

It was also disallowed from publishing flash apps in the app store. Once the developers left it was as good as dead IMHO and all that kind of content moved to mobile.

Definitely. I remember at the time Adobe Air was still recently new and there was some buzz around in the tech scenes because everyone like Balsamiq Mockups, which originally was only in available in Adobe Air.

Then Apple announced they weren't going to support flash and overnight everyone realized flash was dead, and Air and Silverlight were both a dead end.

I was working on a Silverlight product at the time and it got canned soon after too, with Microsoft soon losing enthusiasm for it too.

Some oldies from here:

2007 - Steve Jobs saying Flash is bad for video:

https://news.ycombinator.com/item?id=130496

2008 - John Gruber being prescient about Apple banning flash compiler:

https://news.ycombinator.com/item?id=320431

2010 - Apple bans Flash-to-iPhone Compiler (2nd comment thread is pg saying he's against it ideologically):

https://news.ycombinator.com/item?id=1250799

This is untrue. Apple attempted to ban all third-party development tools, but never actually enforced the rule due to the threat of a lawsuit from the FTC. They dropped the rule about three months after Jobs' famous "Thoughts on Flash", but the fact that developers still remember the ban is probably a good chunk of why people forgot about the exporter.

That was much later and too late for flash.

https://appleinsider.com/articles/10/08/05/ftc_believed_to_b...

Flash abandoned their own product only after it wasn't allowed on iPhones anymore.

I remember before this thinking that Flash was a viable way to make cross platform apps that also covered mobile

So at least from my perspective this had a big impact

As someone with a solid HTML/CSS background but weak JS, I gave up on trying to figure out GSAP and other animation frameworks, and just bought Tumult Hype.

It's a GUI-based web animation tool that exports to HTML and MP4, with the ability to write custom JS if you need it. Being able to visually look at timelines and set tweening by dragging a slider is much more my speed, and reminiscent of Flash, which I haven't done anything on in more than 10 years.

Best of all, it's a one-time payment. I considered Adobe Animate, but monthly subscriptions are a total no-no for me.

While I agree there is an unfilled need after the end of Flash, is the problem really a matter of technology or is it just that nobody has produced a vector animation editor as easy/accessible as the Flash software was? Clearly all the pieces are in place to do everything Flash can do if you can emulate Flash at full speed entirely in the browser.

Flash Animation was made possible by technical limitations as much as by technical possibilities. 480p Flash clocking in at less than 100kB vs RealPlayer needing several MB for the same amount of content in 240p is a huge difference on a slow connection.

1080p streaming in real-time is "good enough" and YouTube will not only host it for free, but even share ad revenue. You could deliver vector animation in HTML5, but a pure animation platform will have a hard time against something that can do both animation and camera recordings.

There are actually a large number of apps that did launch on Flash and were on iOS and Android. Adobe had app exporter functionality for Flash back in 2010. There's even drama surrounding it. If you remember Steve Jobs' famous "Thoughts on Flash" - that wasn't a mea culpa for not supporting browser plugins. Apple was seriously considering banning all third-party development tools to stop Adobe from getting Flash games on iOS, and "Thoughts on Flash" was written and published to justify that ban. The FTC threatened to sue and Apple backed down a few months later.

I should also point out that you seem to be remembering the days of Flash 5-8 where development was primarily on the timeline and AS2 classes were fancy new tech. By the time Adobe and Apple were feuding over Flash, most developers were using AS3, which was a lot more like modern JavaScript development. Hell, Adobe was specifically pushing to make AS3 the next version of JavaScript. You could still develop AS3 on the timeline, but it was less powerful than AS1/2 and most developer resources assumed all your code lived in separate files using Adobe's APIs to interact with the display list directly.

Agree with that sentiment! But would take it one step further. I argued in 2010 that Adobe should have (could have, would have) launched their own hardware platform. This would have given them & the thousands of loyal Flash devs a more permanent 'final destination' for their apps & games - despite Apple's intentions.

Easier said than done, but fun to think about how things could have played out if Flash was given a dedicated hardware platform and follow on generational iterations ie- today maybe we'd have one more option other than iOS or Android. In fact, an allegience with Microsoft probably would have been best and saved both companies' endevours in the dedicated phone/tablet space but now I'm just spitballing.

> Flash was so much simpler than the Javascript "standards" layer cake we have today.

For basic animation and graphics that's true but not for creating anything interactive. ActionScript was based on ECMAScript / JavaScript and was very similar. If you knew JS you could easily write AS and vice versa.

If you mean it didn't have the complexity of today's JS-based apps that may be true but that's a choice, not a technical limitation.

I feel like Unity3D could be turned into an animation tool like Flash with some extensions. It already has HTML5 export.

Unity’s HTML5 export is super heavyweight (unless you do some super complicated things getting it to work with the ECS stuff). Flash was...though abused a lot, much more lightweight, able to slot into a randomo webpage without sticking out too much/causing performance issues.

Also as a direct substitute, Unity has much inferior font rendering, and no built-in vector graphics capabilities.

If I was doing a 2D animation tool for browser export, I don’t see that unity would offer me much, and would come with a lot of downsides.

The Flash authoring tool (now called Adobe Animate) has HTML5 / Canvas output via the CreateJs library.

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

Agreed. Flash a quality that's hard to identify.

Is the source code for Flash available? That would be brilliant.

Media authoring-wise, the Wick Editor [1] is attempting to fill the gap. It's still pretty young, but looks promising.

[1]: https://www.wickeditor.com

There are game libraries that try to be similar to flash, such as http://www.openfl.org/learn/.

Canvas should aim to be as close as possible to this. I think with a bit better animation support they would be there, but I really don't know much about the domain.

With frameworks like PixieJS and Three.js it is there, but requires more coding to start, instead of using a simple GUI to get started and then add code as needed

The flash player spec was published and the ActionScript3 engine was open source.

What replaced flash? Was it SVG + JavaScript?

Video replaced flash movies and nothing replaced flash games, there's just a hole.

> Nothing filled the gap left by Flash

Flash should have never been used to create websites. end of story.

Nothing filled the gap left by flash?

I see that as a good thing. Flash was a power hungry, slow and obnoxious technology. Not to mention it was a security nightmare.

I'm glad the barrier to entry to make pointless animations is atleast slightly higher

Reply on news.ycombinator.com