I've seen what it could do with a Clip, iPod, and Zen. I wasn't impressed as it made them actually look less polished and more "hacky". The other issue is that it causes more problems for the OEM, as people then start demanding the manufacturer implement things they didn't intend on. It takes them away from their in-house roadmap in having to field questions from users that end with "But RB has done it..." and as a developer myself, it gets frustrating when users start killing your timeline.
I'd rather let the OEM maintain their roadmap, and have a cohesive userbase drive the innovation with requests and head counts. If there is a feature that is needed, and enough users demand it, the OEM will put it into their roadmap. That way, new users as well as those who aren't tech savvy get the features directly out of the box. In turn, that makes for a better user experience and drive good sales for Cayin, which in turn makes for more and better future products.
Once you get the third parties involved, the manufacturer gets pushed between a rock and a hard place. The customer base gets diluted and split between those who want to see it in the real firmware, and those who "Will just get the RB team to implement it". With less demand, the OEM's firmware ends up getting less attention, and the product usually suffers. Remember that the OEM has licensing and royalties for their platforms and firmwares. If they lose demand for this firmware and it gets "opened up", what's to keep them from making a "thin" firmware for future products because they know that it's just going to get hacked and not care anymore (or worse, can't afford to pay the full dev team). They'll still sell the hardware, but the out-of-box experience suffers. In the "old days", it was fine to push an OEM, because there was little competition so OEMs could just say "too bad". But in today's industry, there's a lot of other companies and I'd rather see one that makes good, solid stuff out-of-the-box and can afford to pay for a dev/support team to maintain it, not a bunch of hobbyists and part-timers.
The N3, *I feel*, is perfect where it as, for what it is intended for. People I've recommended it to (as a sound engineer for those who want something to audition mixes on; as a developer for those who want something nice to listen to while working; and as a photographer for those who want something to have a lot of music to listen to while on set) all love the N3 - straight out of the box. It becomes a lot harder when you start having by saying "well, buy it, then go here, download this, install this to the card, then do this...". Look at Magic Lantern for cameras. Yes, it opens up features, but I shoot all 1Dx and 1Dx-II cameras, which Canon has said they'd legally go after anybody who back-doors the firmware, so we get a very stable firmware directly from the OEM. I can go on a shoot knowing that the camera works perfect every time, without hacks, and with the full support of the manufacturer (that I can call right then and there if an issue arises). There hasn't been a single ML feature that I've found needing, as a lot of the features are for making the lower cameras do things that wasn't the primary feature of the camera - video. They are primarily still cameras, not video, as the C series cameras are designed for video with the capability to take stills. Yes, it's fine for the hobbyist to ML their 5D series to do video, but people who do nothing but video either 1) use the OEM firmware just fine with the included video functions or 2) get a camera design for video (C-x00, BlackMagic, Arri).
The Fiio X1-II (as well as the X5-III) has good hardware too and sounds nice, but the poorly written firmware and massive bugs (that still exist on the current release), have left a bad taste in a lot of peoples mouth. If any platform (Linux based or otherwise) need some third-party hacking, that(those) would be a better candidate. I'm not suggesting that anyone stop from open-sourcing the N3. I wouldn't use it, but that's me. I just feel that the time would be better utilized forming a cohesive userbase that can give Cayin good, solid feedback on what they can implement into their timeline.
