Hacker News Reader

Comment by roca | Hacker News Reader

What reason is there to believe that Ruffle will succeed where other Flash emulators (Shumway, Swiffy, Lightspark, ...) have (apparently) failed?

View on HN

I wouldn't say Lightspark has failed at all, they're progressing nicely. Others are too.

I heard a rumor Shumway was abandoned as a project by Mozilla because ex-Adobe engineers were steering the project and Adobe issued concerns; they probably could have been worked out, but you know, Mozilla. Since then, Flash has been end of life'd more aggressively, so I think Shumway could return.

Flash files should be like music or graphic files - playable in a variety of players and items. Even VLC someday, we could hope.

Hi, original author of Lightspark here. There is certainly value in OSS implementations of Flash, but the reality is that reaching accurate support for all the SWFs out there is a titanic effort. The sheer size of the runtime library of ActionScript 3 makes it a massive project, and on top of that the documentation/specification is most often vague or missing altogether.

Lightspark is something I started 10 years ago, back in my University years. Nowadays, I am working on the problem of Flash preservation and life extension from the perspective of compilers and virtual machines at Leaning Technologies. If you want to read more about our approach:

https://medium.com/leaningtech/preserving-flash-content-with...

https://medium.com/leaningtech/running-flash-in-webassembly-...

https://medium.com/leaningtech/announcing-cheerpx-for-flash-...

A Flash player is a lot more like a Java virtual machine + libraries or a Web browser than it is like a video codec.

I have not heard that rumor about Shumway. As a former Mozilla distinguished engineer, I think I would have.

Thanks for pointing out Lightspark is progressing. I thought it had stalled, but I was wrong. Still, if Lightspark is doing well it's not ideal to split efforts into multiple projects.

I’m a program manager at Mozilla and I helped manage the Shumway project for a year or so. A couple jobs before Mozilla, I was an engineer on the Flash Player team at Macromedia and then Adobe. Since I was not writing Shumway code or guiding the team’s implementation decisions, Mozilla’s lawyers did not feel there was a conflict. If Adobe really wanted to shut down Shumway (or Ruffle or Lightspark), they could cite numerous obscure Flash patents regardless of who’s working on the project.

Mozilla stopped working on Shumway primarily because:

1. Flash compatibility was going to be a lot of work and full compatibility was impossible (e.g. matching Flash bugs, closed-source codecs, system APIs not available to JavaScript)

2. Content creators were already moving on from Flash. They were slowly heeding the “Flash is a dead end” messaging from browser makers and Adobe itself.

So Mozilla saw multiple engineer-years of work ahead to support a proprietary technology that was shrinking and decided the Shumway engineers’ time would be better spent on other Mozilla projects. Even if Firefox bundled a Shumway with great Flash compatibility, content creators would choose a technology that is going to work in Chrome and IE.

Fun fact: before working on Shumway, one of the engineers built another JavaScript Flash Player called Gordon. The Shumway name is thus derived:

Flash -> “Flash Gordon” -> Gordon -> “Gordon Shumway” (the alien ALF’s real name) -> Shumway :)

As a counter example, VLC does support the Java-based menus of BluRay discs if you have Java and so on available.

I recall being able to play .swf files in Media Player Classic a decade ago.

Sure, you can embed Adobe's Flash or a JVM in a media player.

But what I said is still true in terms of implementation effort.

Reply on news.ycombinator.com