Roon 1.4 request: rethink UX for consistency and convenience

First, just to establish the tone, I love Roon and I looove 1.3.

That said, I have concerns about the user interface. Didn’t bring it up before because I knew the team was busy with deep features.

I will illustrate with specific problems, but I encourage you not to think in terms of fixing individual bugs. I think these issues require serious rethinking. I think they reflect the organic growth of the product: ever since the first, brilliant design, stuff has been added a bit haphazardly.


People have raised several inconsistencies in appearance. The heart, mike, three-dot, repeat and shuffle buttons are in varied groupings and arrangements and positions on the pages they appear. Some things are clickable and some are not, with no consistent indication: a gray button, blue text, white or black or gray text. Some clicks change the visual presentation, some change the metadata, some take action, with partial but incomplete indications. These issues matter and should be addressed.

And radio on the playing page: where transfer, a quite complex concept, is just a button, radio gets a whole little panel of explanation. Jeez.

But there are more fundamental inconsistencies, in how information is brought up. Consider the album page:

Tracks and Credits are tabs, one is visible at a time and nothing else is affected. No need to dismiss.
Other versions is in a modal pop up, brought up with a button, dismissed by clicking outside.
Review text expands by clicking on the down arrow, and it is partially modal: the visible stuff in the upper page remains active, but the tracks or credits are gone. It is dismissed with the up arrow.
Currently playing page is modal, and is brought up by clicking on a button, and is dismissed by clicking the same button.
Currently playing full page is modal, and is brought up by clicking a button along the right edge, and is dismissed by clicking an X in the upper left corner.
PDFs open by clicking a button, in an external viewer. Dismiss through the operating system.
Pictures open by clicking a button, in a modal gallery view, navigated or dismissed in a funny way: if I have two images and am looking at the first, clicking to the right of it advances to the second, clicking to the left, above or below dismisses the image (clicking on the image magnifies it).

None of these are recorded in the page history, none are correctly dismissed with the back button.

Other inconsistencies in the artist page: Performer and Composer work as tabs but are selected with buttons and look like pages. They behave quite differently: on the Performer page, categories (Main Albums, Appearances, Tidal Main Albums) are sections in the scrolling page, but on the Composer page, the categories My Library and Tidal are tabs.

On the Performer page, each section can be expanded to show All Albums by clicking on a button, which brings up an entirely new page and is dismissed with the back button.

On the Composer page, the My Library tab just scrolls, but the Tidal tab has multiple pages which you navigate with buttons and dismiss by selecting another page.

This randomness is more disturbing than icon placement and similar details.

I have suggested that we can pick up ideas from some existing news readers, including the New York Times iOS app. This proposal is not exactly like those, but inspired by them.

An album is shown in a long continuous page, divided into sections. Each section can be closed, partially open, or fully open. Imagine a default layout where each is partially open, like the review page is today. I can scan down past all of these sections. If I really want to read the review, I open it, but I can still scroll down and get to the pages below. If I don’t care about the review at all, I can close it and it wastes a minimal amount of space, a single line.

When fully opening a page that is partially open, the existing stuff is not repainted: I see part of the review, and when I open it more text appears underneath and the lower stuff is pushed down.

Now, let’s modify this idea in two ways:

First, there is no default layout (beyond the first time). The system remembers the way I have it. So if I close a review on one album, other albums that I open will show up with the review closed; if I open a review to read it, navigating to other albums will show the review open. This eliminates much of the settings in terms of defaults: there are no defaults, the system remembers. And this memory goes across sessions, and reboots: the system always remembers everything, including where I am. (In my view, the idea of a session is an artifact of an implementation detail, RAM memory is more expensive than disk storage. But I don’t want to care: if I do something in a program, and go away, and come back later to continue, I want the machine to remember everything about what I was doing. And I don’t want that memory to depend on what the machine did in the meantime, run or sleep or patch-reboot or power glitch. We begin to see this behavior in some systems: even Office is cautiously suggesting that it should take you to where you were last in the document.)

Second, I think this model can apply to many of the random behaviors listed above. But not quite all: some things should perhaps be modal. But not many: modality is abhorrent, it interrupts the flow.

So this model still requires some careful thought. And there are plenty of design decisions to make it pretty and convenient, consistency is not the only consideration.


I agree. I think UI should be the number one priority in the next major update. (+ DSD iso support :slight_smile:)

1 Like

I definitely think there’s room for some UI redesign for consistency and polish - as you say it’s evolved rapidly and incorporated a lot of new features…

1 Like

I would not like to see cosmetic operations the number one priority of a next release.
See the feature request thread for more urgent needs.

1 Like

+1. Including the possibility (MacOS) to swipe the magic trackpad to go back to the previous screen.


My vote is definitely for a review of UI consistency and polish. The “urgency” for new features is always debatable. If we’re not careful we’ll end up with something like the horror that JRiver has grown into.


Thanks Anders for your thoughtful post. I’ve brought it into Feature Requests for discussion and to enable the devs to find it when looking for UI feedback.

1 Like

Roon is for Music only.

Big part of Jrivers GUI inconsistences result from being a Music, Video, TV and etc… software.

I hope that GUI polishing can go hand in hand with feature development.


And that is also my biggest fear, i do NOT want multichannel audio, i do NOT want support for .iso-files etc.

It would be good to see an official agenda where Roon is heading… It’s easy to suggest new features but it’s more difficult to put your vote down for keeping it 2channel music oriented…

1 Like

How is that any different than saying i dont want mp3 support or i dont want internet streaming support …etc. features are endless and i think they are wellcome by many. I dont use all the features of Roon and some i am maybe not even aware of. Howeever i like new features although some are useless for me. I think the important thing is that unless the new features cause bugs or compromise sound quality and my experience of roon, i welcome them even though i may never use them. So more features dont have to be messy or compromise on 2 channel quality of roon. however i think that as so many features are added with 1.3 i would like a more stable, bug free, more ergonomic ( regarding ui and metadata) roon in the first place. Of course features are wellcome. Dont get me wrong i dont say that roon is buggy hard to use …etc. In my opinion it is the best software for my needs and i like it and admire it so i expect nothing but the best. Also when i check the feature request thread, most of them are about the roon experience (ui,metadata and related feature request or bugs)… sound quality issues are not top priority if i am not wrong

It’s already multichannel - came with 1.3.

I’d imagine you could just make a feature request of that specifically…

I’ve never heard Roon say they had any intention of doing video or whatever, but I guess ultimately they’ll add whatever they think best for the product and their user base.

In the community, one of the hardest things is keeping the feature requests clear and on topic. Hence we should probably stick to this being about UX/UI

1 Like

I totally agree… in the end developers will decide …So I think this should not stop us asking for what we think will improve our xperience

I dont mind the cosmetics as much as the user flow. I just think there are too many clicks and mouse movements to do simple things. For example, to play an album from the Album view takes 3 clicks and movement from the album, to the top of the screen to the middle of the screen. Seems totally inefficient. What about constantly having to back to the hamburger menu to go to various pages. There a lot of examples like this where I feel its just inefficient. I know people will say big deal, moving your mouse across a screen a few inches. In my opinion, Roon is first and foremost a music library and playback system. If it cant get these basic features right then I think it has failed in its fundamental purpose.


Minor user interface improvements should be made but it certainly shouldn’t be the main focus of 1.4.

I’ll do six clicks in a user interface if it gets me offline sync and playback to my macbook pro and eventually my iPhone once the Xamarin issues get resolved. That’s a 1.4 worthy feature, IMO!

I actually like the steps required to play an album. It’s make me feel like I’m kind of “go to my bookshelf and pick up that cd, insert it in my player and hit the play button”.
It’s something that makes Roon feel “more real” than other softwares.


Better UI +1


I’m with @AndersVinberg on this. As he describes, as features are added it’s important to keep watch over UI/UX to keep things as consistent, intuitive, and elegant as possible. And there’s always room for improvement in this area … in any application. As stated, without vigilance over time, you end up with a JRiver or a Microsoft Word, and if that’s what we wanted, I don’t think we’d have shelled out our hard-earned cash for Roon.

I can’t see a compelling reason why work on UI/UX can’t proceed alongside new features, nor do I see a bright-line distinction between “cosmetics” and “user flow” —IMO, they’re two sides of the same coin. I get that the development team has been preoccupied with making Roon the best in its class technically (and I sincerely appreciate the effort), but there’s ample evidence that they’re aware of the importance of UI/UX, and I anticipate they’ll move toward a more balanced approach with future versions.

My two cents.


I am in agreement with @AndersVinberg, too, in that the inconsistencies make the product harder to operate.It resembles nothing else (not the Windows or *ix UI anyway), and that increases the need for consistency.

In using Roon, I feel I’m stumbling through it. As one example, I can never figure out the quickest path (or any path) to track credits. I’m still not sure if they are present for all library items
or not.

@AndersVinberg said it well.

1 Like

I agree, UI need a bit of polishing.