Hacker News Reader

Comment by perardi | Hacker News Reader

Let’s pretend that Apple allowed Flash on the iPhone.

Would Flash still have continued to be viable? It’s really difficult to thought experiment that out, but, given the need for responsive designs, and the maturation of JS and the whole React/Angular/etc paradigm, would Flash still have stuck around?

(I realize that a lot of the JS improvements came about because you now could not guarantee Flash, so you had to flesh out the native browser APIs. Remember Flash file uploads? Or using Flash to copy to clipboard?)

View on HN

Creating a throwaway account to comment as I worked with the Flash team when things came to a close...

Who knows if Flash would still be around today, but there was a great deal of innovation happening on the Flash engine up until the point the Flash Runtime team was effectively dissolved, because of Apple's decision. Up until that point, Adobe worked very close with Apple to recreate the Flash Platform from ground up with an entirely new engine. The new engine was vastly superior, and passed all of Apple's requested criteria for performance and memory usage on iOS. The runtime overhead compared to native was astoundingly very close, and effectively negligible for game development. In fact it was so efficient that most people internally thought that the new Flash engine could outcompete Java for general purpose cross-platform runtime based on several internal benchmarks. Even the language was improved with ActionScript4 (including Generics support) and things like threading/parallelism, simd, etc support was a focus.

However, everything went south when Apple when, despite the new Flash engine exceeding their requested requirements, Apple decided at the last minute that they are pulling out of the agreement. While it was not explicitly said, it was implied that Apple thought Flash support may hinder native game development that then could be sold on the iOS store. Once Apple pulled out, it was the straw that broke the camel's back as the massive investment into the new rewrite of the platform couldn't be justified. It was really sad as we had brilliant engineers and researchers working for years on the rewrite. But, sometimes that's life.

It's not well known, but ActionScript 4 language specs was publicly released: https://github.com/adobe-research/ActionScript4

More info: http://www.rivellomultimediaconsulting.com/actionscript-4-re...

I worked at Palm on WebOS. If this is the same engine we got for WebOS (which is also what Android phones had afaik), it was still enormously power hungry, slow, unstable, and ripe for security exploits. I’m not sure I’d buy the claim that you met their requirements as security would be a big one, especially since Safari already had a huge attack surface for quite a while on iOS and Flash integration would effectively be increasing that many times over.

No partner was sent the new Flash Next engine afaik, so Palm would have had the old mobile port of the existing Flash Runtime, which was pretty slow on mobile.

I mean, I'd wager that if this person says they "met Apple's requirements" then that's probably because they actually had Apple sign off on a long list of requirements that were indeed met, in a formal fashion. When you're working on a project of that size/importance, with companies that large, you don't typically casually say you "met the requirements" without that having been validated by the customer/partner.

That said, I could indeed imagine that behind the scenes Apple had reservations about security which contributed to the canceling of the partnership. Heck, I bet there are lots of reasons that we probably won't get to hear about unless someone involved with that on Apple's side posts about it.

I mean no one from Adobe’s side has posted anything. For all we know it could be someone trolling with an anonymous account. By now I might have expected some actual reporting about this if it was at the point where Apple had formally given Adobe a list off requirements. That would be different and you wouldn’t expect to see Steve Jobs blasting Flash continually into 2010.

Let’s take it at face value though. I’m also clear that in large companies like these, rumors abound. It could be a junior engineer who misunderstood and these weren’t actually requirements from Apple and it was internal requirements. Could also be stories part of the team believed but missed that they missed other requirements (eg they met the perf criteria but failed on security).

Do you know if there is any reason (other than simply not wanting to) why adobe couldn't open source the Flash Runtime? It would likely have been very successful if they had (iOS or not, and it would have put quite a bit of pressure on Apple to support it), and Adobe would have stood to make a lot money selling the Flash IDE.

There was a lot of support internally to open source Flash, but the reality is that Adobe used a lot of partner solutions that have patents to improve Flash for various needs-- video DRM is one example, but there's other more practical features that used patents too. The work to remove all the patents would have taken at least 6m to a year, and that's far too much investment for no ROI.

Was this before or after the launch of the iPhone? I'm primarily curious because I was under the impression—based on several reports I've read—that at launch Apple actually had no intention of building an app store or allowing third party iPhone apps, which means "native" game development would not really have existed.

> I was under the impression—based on several reports I've read—that at launch Apple actually had no intention of building an app store or allowing third party iPhone apps

I've seen contrary rumours that Apple always intended to do so, but that it wasn't going to be ready for iPhone OS 1.0 so they promoted web apps until 2.0 was ready. But I'm not sure we'll ever know which is actually true.

Let’s not sugar coat history.

Adobe claimed that they could have gotten Flash working on the first generation iPhone if Apple had allowed them.

When Flash finally was running barely on Android, it required a 1Ghz professor and 1GB RAM.

The original iPhone had 128MB RAM and 400Mhz processor. It wasn’t until 2011 that an iPhone had those specs.

Flash is a blackbox, and is a completely closed environment, and it's only software, there is little hardware relation. So no.

Even today, windows is able to survive because of hardware.

Flash could have gained from opening their format and making it a little more standard. I'm really happy that it's dead.

Although the Flash authoring and Flash Player software have always been closed source, Macromedia, and later Adobe, have for a long time opened various formats used in the Flash ecosystem. Some examples:

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

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

https://en.wikipedia.org/wiki/Tamarin_(software)

No.

It was designed for desktops, as Jobs stated in his thoughts, it really requires a keyboard and a mouse for most SWF's to work properly. It also became feature rich and power hungry at a time when Mobile devices were taking off, so it was never going to run on the mobile devices of the day. This is all besides the fact that its custom networking features were proprietary, highly exploitable and notoriously buggy.

It was great for it's day, and still has features and advantages in animation and networking that JS Canvas or Node can't equal today; but it had to go. I loved Flash and worked as an AS3 dev professionally, so i wanted it to succeed as much as anyone, but it did not adapt or fix its issues so it had to die. Good riddance.

HTML/JS was already improving, wrt better handling of the kinds of things Flash was specifically used for (graphics based interactive applications) before the iPhone was released and became popular. The canvas element was introduced around 2004, WebGL started with Canvas 3d around 2006. At that point, I think the writing was on the wall, Flash was a proprietary format that had many of the same issues as Java Applets. Something better integrated into the presentation experience (JS and canvas/webgl) was going to overtake it as it created a more uniform developer/content producer target. And with improvements to JS performance, and the general improvement of computer performance, it was inevitable.

Additionally, you couldn't guarantee Flash availability even before the iPhone. Non-Windows/Mac systems didn't guarantee it, and many corporate networks eliminated it from their computers (though, ironically, now the main reason I still deal with Flash is my employer's training system).

Cutting it from iOS, and the popularity of both the iPhone and later iPad, just hastened the decline.

Flash continued to be available on iPhone, despite urban myths.

Adobe was quick to react and added support for AOT compilation to native code for iOS.

So plenty of casual games before the rise of Unity, Unreal, SpriteKit, Cocos, were actually AOT compiled Flash games.

I am certain that, colloquially, when people say “Flash”, they mean “the web browser plugin that runs animated content inside of a web browser” and not “that Adobe AIR thing”.

Reply on news.ycombinator.com