In the end I'm not that that heavily attached to the web UI either way, as my primary UI is my own app, SubFire, which I built to run on multiple platforms. The Web UI is, for me, just the point of fixing album covers and tweaking ID3 values, plus a little playlist management (and I'm writing my own smart-playlist "radio" generator that's actually going to take over that aspect more and more).
Perhaps it would be interesting to know just how many prefer apps over the web UI no matter what, simply because of where they use it. Indeed, I'm sure most of us who are app developers generally use the app we built, because it keeps us involved in the development and improvements. But that's the kind of demographic question that is hard to get clear data for, just like it can be hard to get a feel for how many use it for video vs just for music (I'm a music-only user - my video cloud is Plex).
Whether or not some of these UI changes are 'right' might also depend on the organization of your collection. I have 2 servers. One has a LOT of very shallow folders (the rock, celt, folk, and kids/disney set - one folder is JUST my 'King Crimson' related collection). The other has only a few folders but they are very deep, as the classical folder is divided up by era, then composer, then works divided up by conductor. What you use the shortcuts for, I use entire separate MusicFolders for instead, based on your screenshots.
[this, btw, means that one one of my setups, having the Index page use pseudo-tabs for the top-level music folders is useless because 3/4ths of them don't even show up, while the other works just fine and I can clearly see the select-option tool being annoying outside of the left sidebar.]
In both, the shortcuts (for example) mean nothing - in the first case, being 'shallow', I can get to what I want within a few clicks; in the second case, since shortcuts seem to only work for the folders under the 'indexes', they're of no use at all to getting to, say, my fav composers. They really do mean nothing to me. I didn't even know they existed 'til someone complained about it during the b1. I just don't need them, so I don't care either way. For internet radio, I had very little use of it on the Web UI, because being just the raw stream, it didn't give me the updates of track, cover art, etc, that the dedicated clients of the radio servers themselves (radionomy, etc) give; anybody else like that would not have even noticed that the radio page wasn't even accessible in b1
.
To be honest there are a LOT of features, and a lot of different ways to do things. Anything Sindre does is either going to complicate it more for someone who doesn't use or need a feature, or simplify it but necessarily remove a feature or access path someone really liked. This is true for any aging stable and successful product.
In the end, any UI needs to make all things possible and the most common tasks easy. The index being its own top-level page meant that the left side had little use as being somewhat redundant. The oversight was that the "common task" is that users switch gears (in effect, change stations) often, and rapidly. If you just want to dig down to the album or playlist, hit play, and let it go, the empty left-side was just fine. Hide it except when you need it for those rare tasks (settings, logout). It seemed to me a valid reason, a valid "user story", to justify the limited content within it. I was fine with it that way, myself, but as I said, my primary use case for the WebUI is not necessarily to play music.
But obviously many here aren't like that: they seem to want to be able to switch from one to another to another without all the work of going back to the top-level index to do it, especially if, perhaps, they're having trouble deciding what they want. (reasons I won't ever use spotify: too many choices...)
So now, in this b2, we're kinda in a hybrid case there, trying to provide the index for those of that latter group, but now it is the only home for those other 4 options and they have to remain accessible. Maybe they'd be better at the bottom than the top of the left? Not sure. For the most part, b2 was more work behind-the-scenes, especially on the video transcoding. The left-side index I see as a work-in-progress (this is why we call 'em 'beta') for feedback, of which yours is I believe valid. I think if you want to know what he had in mind in the UI originally, look at b1: each feature accessed by top-level "tab", and the sidebar only being there for those rarely used features to get them off of the main navigation page.
But that's just a guess. For myself, as a developer, I'm more interested in API improvements.
Joe