Flash Player, like most proprietary software, is not Adobe's to give away. There's lots of licensed code owned by companies that sold it to Adobe (or Macromedia) and having a functioning FP runtime would require getting those companies to also license the same code. Remember: proprietary software cares a lot less about licensing arrangements[0] than we do. Everything they do wrong can be papered over with money and licensing agreements, because the entire point of copyright is to make money moves out of people's wallets in a way that improves the health of the state. The FOSS world rejects this principle, so we have to build everything from scratch to get a clean copyright pedigree.
If you're wondering what would actually block a FOSS release of the Flash Player runtime, here you go:
- Sorenson Spark, a half-finished H.263 implementation.
- On2 VP6
- Likewise, their H.264 or MP3 decoders, which were almost certainly third-party libraries as well.
- The "advanced text rendering engine" added in SWFv8 to replace legacy font rendering, which I know was licensed but I forgot which company.
- Adobe DRM
The media codecs would be replaceable with any number of FOSS implementations. I've[1] personally written a clean-room decoder for Sorenson-flavor H.263 in Rust, and the APIs for these libraries tend to be pretty simple and easy to wire up. Font rendering would potentially be more invasive to the codebase. And DRM is basically high-grade radioactive waste to FOSS projects. The main problem is that this wouldn't be "just compile Flash from source" anymore, you'd be making an "OpenFlash" fork with at least some compatibility breaks involved.
And to be clear, Adobe actually had some interest in opening Flash Player. They released a bunch of technical specifications for the SWF format - albeit, hilariously inadequate and inaccurate ones. AS3 in particular was supposed to be a web standard, in the form of ES4, or "JavaScript 2.0". Adobe actually released their AVM2 implementation, avmplus, specifically with the goal of it replacing SpiderMonkey in Firefox. This never worked out[2], but they kept updating the avmplus GitHub right up until the death of Flash.
The underlying problem with both "opening up Flash Player" and "porting Flash to iPhone" is that Adobe management was not terribly interested in doing things that did not have a direct connection to bottom-line revenue. If they were willing to open Flash, they would have also invested into their platform instead of trying to find a way to bilk money out of people cross-compiling Unity to Flash.
I think the best goodwill move Adobe could make now, in 2025, would be to specifically release the renderer component of Flash. They own it completely, the copyright pedigree goes all the way back to Jonathan Gay, and it does a lot of specific things that no other SWF runtime implementation gets quite right. It would be an absolutely amazing object of study.
[1] To be clear, several other interested Ruffle contributors helped me actually get it decoding video (especially some broken files) and doing so efficiently.
[2] I'm not a fan of him but I hear Brendan Eich stalks these forums. He probably has better insights on the backlash to ES4 than I do.
If you're wondering what would actually block a FOSS release of the Flash Player runtime, here you go:
- Sorenson Spark, a half-finished H.263 implementation.
- On2 VP6
- Likewise, their H.264 or MP3 decoders, which were almost certainly third-party libraries as well.
- The "advanced text rendering engine" added in SWFv8 to replace legacy font rendering, which I know was licensed but I forgot which company.
- Adobe DRM
The media codecs would be replaceable with any number of FOSS implementations. I've[1] personally written a clean-room decoder for Sorenson-flavor H.263 in Rust, and the APIs for these libraries tend to be pretty simple and easy to wire up. Font rendering would potentially be more invasive to the codebase. And DRM is basically high-grade radioactive waste to FOSS projects. The main problem is that this wouldn't be "just compile Flash from source" anymore, you'd be making an "OpenFlash" fork with at least some compatibility breaks involved.
And to be clear, Adobe actually had some interest in opening Flash Player. They released a bunch of technical specifications for the SWF format - albeit, hilariously inadequate and inaccurate ones. AS3 in particular was supposed to be a web standard, in the form of ES4, or "JavaScript 2.0". Adobe actually released their AVM2 implementation, avmplus, specifically with the goal of it replacing SpiderMonkey in Firefox. This never worked out[2], but they kept updating the avmplus GitHub right up until the death of Flash.
The underlying problem with both "opening up Flash Player" and "porting Flash to iPhone" is that Adobe management was not terribly interested in doing things that did not have a direct connection to bottom-line revenue. If they were willing to open Flash, they would have also invested into their platform instead of trying to find a way to bilk money out of people cross-compiling Unity to Flash.
I think the best goodwill move Adobe could make now, in 2025, would be to specifically release the renderer component of Flash. They own it completely, the copyright pedigree goes all the way back to Jonathan Gay, and it does a lot of specific things that no other SWF runtime implementation gets quite right. It would be an absolutely amazing object of study.
[0] https://www.theregister.com/2018/03/14/a_dolby_sues_adobe_fo...
[1] To be clear, several other interested Ruffle contributors helped me actually get it decoding video (especially some broken files) and doing so efficiently.
[2] I'm not a fan of him but I hear Brendan Eich stalks these forums. He probably has better insights on the backlash to ES4 than I do.