JRiver usually doesn't keep a list of features they want to add in a major version, as such lists only lead to high hopes and expectations. There's always that chance if they publish a "to-do" list for MC27 and they're not able to add a certain feature or features announced during that release cycle, some users might get upset about that. Instead, they usually compile the list of new features as they're being added. Since they're early in the MC27 development cycle, they've likely decided not to publish the "What's new in MC27" list of features yet until it's ready to be published (and/or closer to the first builds of MC27).
Yes, taking advantage of the early bird upgrade is a leap of faith, but if/when JRiver adds a feature in the MC27 release cycle and you didn't take advantage of the early bird upgrade offer, you might come to regret it (or you may not care and would happily upgrade if you want). Another way to look at it is upgrading to help fund development and support JRiver regardless of what new features are added. Personally, this is why I upgrade yearly, to help support development and JRiver in general.
Also, keep in mind guys...
1) Upgrades are optional. If you're happy with whatever version of Media Center you're using and you don't want to upgrade, then don't. There are a few exceptions to this though, mainly on Mac and Linux where older MC versions may have issues on newer Mac and Linux distro releases. Windows in comparison is very forgiving in regards to backwards compatibility with older software.
2) People say there's not many audio-centric features these days, where the past few major versions focus more on video and images. This is true, but remember JRiver Media Center already can output bit-perfect audio, and it's pretty hard, if not impossible, to improve upon that. Existing audio features like improved DSD-to-PCM low-pass filtering options, DFF-to-DSF (and DSF-to-DFF) repacking (conversions) without resampling so no PCM step in the middle, etc. could be improved in newer MC versions. Also support for things like VST3 and other DSP additions/tweaks could be added as well. But remember, it's called JRiver Media Center so it focuses on audio, video, images, television, etc. But to say there hasn't been audio-related improvements over the last few years really should read the MC changelogs more closely. Things like the WDM driver (and improvements to the WDM driver), memory playback (including additional changes/improvements over the last few MC versions), the SoX resampler, TPDF dithering, HDCD decoding, waveform seekbar, etc. have all been added. Afterall it is called JRiver Media Center, so it focusing on video and images makes sense, as audio support is pretty mature at this point.
So that's all fine and dandy. However, for those potential users with certain expectations and are holding out for specific features (mainly these listed below), keep in mind...
1) Integrations with streaming services like Tidal and Qobuz are likely never going to be added. This is something that has been discussed many times before (do a forum search, Jim explains why JRiver chooses not to pursue this) however there is a workaround, using MC's WDM driver alongside the Tidal/Qobuz apps will work, but it requires using two apps (and I don't think multichannel works). Jim has said they're willing to support Amazon Music if Amazon themselves funds the development, so you guys who want this should go bug Amazon about them giving JRiver API access (and the funding for development) to add Amazon Music integration support to Media Center.
2) MQA is likely never going to be officially supported. You can already unofficially passthrough MQA through Media Center if a) you're not using any DSP whatsoever and b) volume is set at 100%. Personally, I find MQA to be an audiophile scam, hardware and software DRM and like Jim has said in the past a solution looking for a problem.
3) People saying they want JRiver to rewrite the entire graphical user interface (GUI). This isn't feasible for multiple reasons, mainly because it'd cost a lot in development time and money - it could take a year or two to completely redo the GUI. Are you guys really willing to give up all other features so they can focus all development on a new graphical user interface? Not only that, if they do change it some users may not like it and would complain about it and want the older interface back. This is a kind of battle where there's no way to win regardless what they do, so I wouldn't expect this one any time soon. That said, JRiver is willing to consider ideas/improvements/adjustments in the existing user interface as long as it's not complicated. You just have to post them in the appropriate topic(s) and explain as good as you can the changes you'd like to see.