A while back I wrote, “MQA may offer SQ value but it is number 87 on my list of priorities”. @hifi_swlon asked what was 1 – 86, and @Danny wrote “another topic for you to explore on the idea of what’s important to audio and music lovers? I’d love to read it … and oh my, the debate it would create!”
There are not actually 86. There are several scenarios that come together into one core functionality, with 85 sub-bullets: Roon needs to be a truly distributed system, with support for mobile devices.
The Roon team is of course already aware. Nonetheless, here are my thoughts, in the hope of inspiring debate.
First, some observations about us users, and the non-users, and market reality, and the evolution of society.
The world is not centered on affluent middle-aged white guys whose lifestyle revolves around a giant hifi altar in their homes. The world is not, and music enjoyment is not, and while certainly the hifi industry has been, it must expand because that is not a growing demographic. We have to offer value to younger music lovers, and we have to fit in with their lifestyles. I say this even though I am a middle-aged white guy with a giant hifi altar at home, but that doesn’t mean that I spend all my time in front of it, or that my lifestyle is centered around it, and regardless, Roon should not just satisfy me and my ilk.
This discussion is about lifestyle. It is about being away from the home hifi altar. This is not just a youth attribute, or if you will, “young” is not about years but about mindset. I am often away from home, and I want my music, in circumstances like
• A second home, owned (long-term) or rented (medium-term)
• A resort or hotel room (short-term)
• A boat, anchored in a quiet cove (yes, this is my personal favorite)
• An office
• A hotel room on a business trip
• An airplane or a train or a bus or a taxi or an Uber – a vehicle I’m not driving
• A car that I am driving
• Running or biking or…
• Exercising in a gym
Yes, some of those places seem less relevant for serious music listening. Perhaps. To some of us. But let’s leave that debate be. We should not differentiate “serious music listening” from the other kind. I think we can agree that several of those matter, we have seen most of them appear in various contexts in this forum. Each of us will be interested in some cases, and when we look at the whole community and market, all of them matter.
I say it is about lifestyle. But if you want to look at it from a technology perspective, or from a market and business perspective, what are the hot trends today? It ain’t the PC, baby. It is about mobiles and about the cloud. It’s about embedded devices. Small devices. A more high-falutin’ term is “ambient computing”. “Ambient intelligence”, “ambient connectivity”. It isn’t about a “computer”, in the sense of a box that has the purpose of being a computer. It is about every device being intelligent and connected. Which means we will have access to information and intelligence wherever we are, using whatever device we are touching, or even no device. This is the future. The world will look like that. And quickly — it isn’t science fiction. Music has to fit with that. Roon has to fit with that. Otherwise Roon will dominate a dying field, will be the coolest codger in the retirement home.
HOW IMPORTANT, COMPARED TO OTHER IMPROVEMENTS?
In my view, this is far more important than any other feature requests discussed in the forum. Such universal, distributed access will change my life.
Sound quality? We already have stunningly good sound quality. The entire digital chain, including Roon and its ecosystem, is far better than the speakers and the room. Quieter power supplies, better cables, upsampling, all of that is unimportant. MQA, HQPlayer, DSD, all unimportant. Because without any of that stuff, if you just put Roon in an off-the-shelf computer and run a cheap cable (or wifi) to a mainstream DAC, it is already mind-blowingly good. Not just good compared to what some of us grew up with, but compared to how good it can be with the most ultimate tweaking. Any tweak that raises questions about whether it is audible is uninteresting. Uninteresting doesn’t mean it isn’t real, it means I am not interested because there are other things that are more interesting.
Improving sound quality would not change my life. User interface improvements? Convenience? Better random play? Yeah, yeah, all of that is good. But it wouldn’t change my life. Having ambient access to my music library, with all the power of Roon’s library exploration, wherever I am, that would change my life.
Think of it this way: which Roon improvement would make you go running to a non-Roonie friend and say, you have to see this new thing I’ve got, it will totally change the way you enjoy music?
I want my music in all those places. I know that because I do bring music to all those places, but it isn’t Roon, and it isn’t automatic or convenient. I don’t necessarily need all the Roon functionality in all those places: I can’t navigate the social graph of artists while driving or running. But I want the same music, and I want it sorted and organized and labeled the same way: I don’t want an album named for another “album artist”, and I want to be able to look up an album on any one of the primary artists, the way I can in Roon (and can’t in my car). And I want to leverage convenience features such as tags and playlists: I have defined my own category “Italian Jazz”, a tag into which I have explicitly included certain stuff and excluded other stuff, it isn’t just based on nationality and genre, and I want to be able to choose that when driving or running.
And I most certainly want all the Roon functionality when I am in a vacation home or hotel room or office. Because it is useful, and because I don’t want to have multiple user interfaces and paradigms. And dammit, because I have chosen Roon for my music.
And I don’t want to have to manage content across different systems with different technical constraints.
And this is dynamic stuff. The content changes dynamically, of course. But so do tags and playlists and favorites and other metadata. I am not satisfied with sitting down to create a few playlists, and exporting them from Roon into a file in some external format, and then leave it be. I may be listening one evening and tag an album, or “heart” it, or exclude it, and the next day in the office I want those choices reflected.
Multi-master: it is not a centrally controlled environment. It is not a star architecture, a master-slave content distribution system. It is not sufficient that I add and manage content at home and make it available elsewhere. I may be adding or curating content in many of those places. I may certainly be downloading or ripping content in my second home on vacation or a hotel on a business trip, but that’s only the most basic case. The world is more flexible than that: while flying through the sky, I may be reading the “What are you listening to now” thread in this community, respond to some recommendations, sample some albums on Tidal, and add them to my library for future reference, and I may tag them accordingly (New Stuff). While running, I may click on the Like or Dislike button on my portable device. So the library must be managed in a “multi-master” approach, where any component in any location may make any change.
Opportunistic, lazy sync: if I am in the office or on a business trip, and read about an album and download it and add it to my library on the local device, I may want it to show up on my home system so I can tell my significant other, you have to check this out. And it isn’t just from remote to home: my significant other may be traveling as well. But such synchronous distributed update is too brittle: if the system insists on that, it would refuse the update if the connection is down (airplane), or the home server goes down while I’m on vacation. Such distributed transaction processing was all the rage in the 80s, but in the modern world it was a dead end. Ok, the immediate sync is a luxury feature, if I’m away for a few days the update can wait. But when I come home or to a place with better connectivity, I don’t want to go through a laborious manual sync, I want it all automatic. So we need lazy, opportunistic sync, of both metadata and content, in both directions.
TECHNICAL CONSTRAINTS AND OPPORTUNITIES
In order to achieve this, we have to consider some technical constraints, and take advantage of some technical opportunities.
Bandwidth: in some of these scenarios the device is mobile and connected through cellular. Technically, state-of-the-art cellular supports very high bandwidth, but the cost may be high, and some service providers cap or charge exorbitantly for volume. Some devices may be occasionally connected through unlimited wifi, and at other times through limited or metered cellular. Sometimes I may have no connectivity at all; the main example is an airplane, this problem will certainly go away, but how fast? When will it be available but expensive? We can’t be cavalier about this.
Storage: the storage capacity of a mobile device may be limited. It is growing at a furious pace: an iPhone can have 256 GB, the new iPad goes to 512 GB. But those are costly, not everybody has that much. And even those numbers are smaller than some of the library sizes described among this community. And we shouldn’t limit the solution to only the top of the line devices. Granted, the limit moves upward quickly, so deciding what demands to make on the device is a tricky business decision. In my view, we should not aim only at the peak. The world is not just affluent, first-world geeks. On the other hand, we should not look too far backward, Roon has been commendably disciplined about demanding modern devices. But where to draw the line?
Remote home connectivity: setting up a remote connection to a home system is complicated and fragile, there is a long list of single points of failure, at each end of the connection, the mobile and the home: the internet itself, ISPs, modems, routers, WiFi, computers, NAS… We do not have good ways to troubleshoot remotely. The vendors update components when they feel like it: the reason for the updates Is often security, but fixing security often breaks the network. Does the whole chain always come back after a reboot? A power outage? If you are away from home, is there anybody in your home capable of troubleshooting the connection? Capable and willing? How is Roon going to guarantee that this works, for most non-expert users?
Among the opportunities, we have streaming services, Tidal and others (soon, I hope). This reduces storage demands and bandwidth (by limiting resolution). This is great. But the streaming library doesn’t contain everything I have – many of us have eclectic libraries and care about them. Do we limit the remote content? Spotify and Apple have bigger libraries than Tidal, but they don’t want to cooperate, and some of my eclectic library isn’t there anyway (e.g. ECM).
One solution is that I upload my music to the cloud and Roon provides streaming of it, my own private streaming service. This introduces several business issues. Because of copyright it must be truly private. We have to address the cost of cloud storage: if the user has to provide the storage, that adds complexity and new failure modes; if Roon provides it, it complicates the business model. Can we reduce costs by deduplicating content while preserving the private licensing of content, or will the copyright holders block that?
There are several possible tradeoffs:
• With less local memory and more bandwidth, use streaming.
• With more local memory and less everyday bandwidth, do local caching of content.
• With moderate local memory and less bandwidth, downres the content and cache it locally.
• Identify streaming equivalents of local content, and use streaming as proxies
• Private streaming from the cloud
• Selective inclusion of content, based on automatic algorithms or explicit marking
• Combination of those
For remote locations like a second home, where I want the full library at full resolution, we probably need to support FedEx sync: ship a USB drive with a library snapshot to the remote location, and then use cloud sync for the incremental updates. This hits a cost sweet spot. (After the big stack of CDs for Father’s day, you can ship another USB drive…)
One way or another, because of the challenges of requiring home connection to play, I think routing through the cloud is necessary. Bite the bullet and solve it. Business and copyright and all.
This is exciting. And it is necessary for important market segments. It is challenging, technically and from business and legal perspectives. But we don’t need it all at once. This prioritization is the Roon team’s challenge.