For Widevine, it looks like the following would be required:

    Completion of legal agreements
    Delivery of a Widevine documentation package
    Technical discussion to understand the device type, chipset, and review the Widevine robustness rules
    Exchange of libraries, SDKs, and integration documentation
    Keybox request and fulfillment (as needed)
    Client integration testing

For piggybacking on the Firefox binary, maybe that would work but sounds like something you might get a cease and desist for.

Anyhow, if everyone agree to make the binaries 100% free as in beer (like Cisco did with H.264), then that's better than nothing and makes writing new browsers easier.

But the future is in new devices, operating systems, and virtualization. Being provided binaries limits freedom to make new things. And of course who knows what backdoors are in them.

(Edit to add: this is a great discussion, http://www.reddit.com/r/debian/comments/25kbi7/next_firefox_... -- note that the tracking/privacy/backdoor issues with this are huge.)