Over the last few months webloggia has been full of discussions about the new musical functionality that’s starting to emerge around the web. I wasn’t immune from this trend – I wrote about MediaUnbound (On MediaUnbound and Recommendations Engines) and linked to the (currently pretty awful) Music Recommendation System for iTunes. Dan Hill has also been talking around the subject, talking about first Socialising mp3-based music listening and then about whether whether recommendations scale. And those minxes over at 2lmc linked and commented upon the views of people who are suggesting better ways that iTunes could handle transitions between songs. And of course the new version of iTunes and the iTunes Music Store also now has the user-generated iMix feature – standard web-native functionality which allows people (and now people in the UK, France and Germany rather than just the US) to put mix tapes on the web where other people can rate and/or buy them. And that’s just the tip of the iceberg…
Then of course there are the staples of this new musical functionality – from the rapidly-becoming-indispensible audioscrobbler (which uses the flexibility and granularity of net-enabled MP3 playing devices to create charts, lists and recommendations) through to the self-generating radio stations like last.fm and launchcast. And then there’s all the little hook-in tools like iChatStatus (publish current listening to iChat’s presence display) and Kung-Tunes (publish current listening to the web) that have slowly becoming integrated into my life without my really noticing how they all hook together, communicate, branch off and build upon each other.
All this new funtionality is emerging at the same time (or at least starting to be adopted at the same time) because we’re beginning to see a world in which a decent number of early adopters are now starting to do a substantial portion of their listening on digital devices. Obviously the iPod has been the major success story here – the definitive product that has been encouraging people to do the necessary work to transfer their music into more easily manipulatable digital files. But the increasing prevelance of broadband and wireless connectivity is helping too – becauase it’s the connection of these appliances to the internet that has created the explosion in interoperable, interconnected devices, applications and people. Clearly, the number of people listening to music through these channels is still tiny compared to the entire music-consuming public. There may be many people using iPods, but there’s still an adoption path for moving all your listening into digital jukeboxes and being perpetually connected to the internet (ubiquitous, always-on, non-computer-centric internet in the home is a bit of an obsession of mine at the moment).
But this small proportion looks like it is set to grow. One of the first questions you have to ask yourself in any organic R&D role (which is I think how I’d characterise what I do) is am I a freak or am I an early adopter? You have to have some sense of how much your instincts and excitements are in tune with real people in the world because otherwise you cannot possibly evaluate how those people might respond to the products, concepts or propositions that you think are exciting. In this case, it’s becoming fairly clear that people who are listening to digital music and in connected ways are very definitely more like early adopters than they are freaks. They’re pointing in roughly the right direction. And there are now enough of them that it’s becoming more and more worth people’s time to build little tools or widgets or applications or paradigms or appliances or business models around them. Which in turn appears to be making the whole area still more attractive, creating a feedback loop that is pulling more and more people towards new ways of listening. I don’t want to sound too cheesy but I’m afraid I can’t help myself – it’s pretty clear that we’ve reached a critical mass and that new musical functionality is about to explode. The only question now is what will be there when the smoke clears?
Over the next few days I’m going to write about some of the core trends that I’m seeing in people’s use of digital music, attempting to extrapolate from some current behaviours that we’re all observing around us – concentrating on how people wish to interact and use their music. I’m not going to spend too much time on the way some people may wish to legislate against these desires or build around them – because I believe for the most part that any attempt to do so will inevitably fail. Competing models that more adequately fulfil those needs will rise to take over in their place. The model that meets the most needs (while having the least obvious incumberences) will probably win in the really long-term, even if the market, commercial advantages or monopolist practices deform it in the short to medium term.
I’ll be talking about four major areas that seem to me to be indicative of the unevenly-distributed musical functionality of the future – (1) portability and access, (2) navigation, (3) self-presentation and social uses of music and (4) data use and privacy. These trends within these areas are – I believe – representative of much larger trends across the consumption of all text-based, audio-based and video-based media and so it might be possible to draw conclusions beyond the consumption of music. I am however not planning to do so. And I make no claims that these areas of enquiry are absolute or canonical, or that there are no other areas that I should also be investigating. All I’ll argue is that these four areas are core to the movements that we’re currently seeing and that they are each likely to play themselves out in the product designs, interface designs and business models of the near future.
Of course what comes after that remains to be seen…
Tomorrow: The New Musical Functionality, Portability and access
20 replies on “The New Musical Functionality…”
A good (better imho) alternative to Audioscrobbler is Musicmobs (http://musicmobs.com/). While Audioscrobbler only monitors what one is playing while a software like iScrobbler is running in the background (and using resources), Musicmobs analyses the Playcount of tracks in your whole library, meaning songs played on, for instance, the iPod are considered aswell.
Tom – I’m looking forward to your notes. Its of particular interest to me having founded a music recommendation company 5 years ago and having been through the nuclear winter of tech startups along with the recent interest in music stores.
From a business perspective, it is interesting to note that music of the online stores have very little money left after they divide up the pie to pay for commercial music recommendation services. Take for example, Media Unbound which reportedly got 0$ for Pressplay. Other companies, such as Mood logic and Savage Beast, Kick, Gigabeat etc. have come and gone or are struggling by focusing on additional revenue streams hoping to come out of this environment in a few years.
Services such as audioscobbler and other home built community resources may well be the future as they are run on a shoe string rather than funded businesses….but time will tell. With such a small amount of global music sales being done online it will be very interesting to see how the alpha-geek (not freak) early adopter style translates to the general population.
My belief is that the user experience will need to have personalisation built seemlessly into applications and that implicit rating and profiling systems will be mandatory, versus explicit rating of music from 1 -5 for example.
We have also experienced that people looking at music recommendation systems are significantly different from the general music listening population, in so far as our experience shows that a simple collaborative filtering style people who bought this also bought this is 80% perfect and for a 1 line sqal statement the ROI on this is VERY compelling, making it difficuly to justify additional expense for the 20% for system xyz which uses technique a or b or c (each of which are perfectly valid).
A bit of a dump but its early Sunday morning 😉
This is at a bit of a tangent, but, since I’m certain I’ll never move to listening to music by digital means alone – I have too much vinyl – one thing I find incredibly frustrating is that services like Audioscrobbler only represent a small corner of my listening habits. I use an iPod constantly when out of the house, but tend to listen to vinyl at home almost exclusively, ie. my out-and-about listening is all downloaded and digital, my at-home listening is all analogue and paid for. And there’s a big difference between the two collections, which makes taste-tracking services like Audioscrobbler next to useless, unless it’s somehow made absolutely clear that they represent a particular subset of my musical tastes (an odd combination of ‘disposable’ music and rare or hard-to-find music).
I doubt anyone will be making a device to scan the labels of 12″s on my turntable as they play and upload the record’s details to the web, but that’s the sort of thing we need, I think, in this interim period when people consume both digital music files through web-enabled players as well as, erm, real albums and singles. And I suppose you could take that further – it would be great to have a mobile device hooked into a service like Shazzam that would not only track and publish music I consciously choose to listen to, but music played for me by other people, in nightclubs, etc.
Not sure if that’s any use to you, Tom, but this is certainly a fascinating area for a bit of research – music is curious in that folk seem to wear their music tastes as a badge in a way that they don’t with film, telly, even literature; so attempts to track listening habits and use that information in a social software-ish way should start throwing up some interesting possibilities.
some of the core trends that I’m seeing in people’s use of digital music, attempting to extrapolate from some current behaviours that we’re all observing around us
On a band’s fansite bulletin board that I frequent, there was a discussion a few months ago about digital music meaning the death of the album.
At least, that’s how I interpret the discussion. The album as we know it sprang directly from the invention of the long playing record (LP) — before that, it was all singles. CDs mimic that form, but not entirely (the idea of “sides” has gone out the window). And the iTunes Music Store pretty much uses albums only as a time-based filing device, if you think about it. Oh, this song goes here, in this period of this artist’s output. The fact that singles from tribute albums & samplers show up on the screen in chronological order between singles, EPs and “proper” albums just kind of reinforces that way of seeing songs as separate songs, rather than as being of-a-piece.
In the indie crowd, there’s been a trend lately for semi-concept albums, I think (Mountain Goats, Wilco), that’s possibly a preconscious, nostalgic reaction against this… people thinking “Hey, they don’t make albums like that any more.”
The New Musical Functionality
Tom Coates has the first of what looks like a fantastic series of posts on the new musical functionality, an extended musing on the distribution of production, reproduction, and filtering of music, covering especially the newly social context. Over the…
verbal be mine
Hey flooz kittens whats happening? I’ve just returned from a three day ottawa bonanza research trip. It was good.I’m sitting here compiling some fresh tracks for my gym workout experience and listening to Verbal here from Amon Tobin.. and y…
The New Musical Functionality
Tom Coates has the first of what looks like a fantastic series of posts on the new musical functionality, an extended musing on the distribution of production, reproduction, and filtering of music, covering especially the newly social context. Over the…
Am I freak, or…?
Tom Coates, writing about the new musical functionality of the Internet, says: One of the first questions you have to ask yourself in any organic R&D role (which is I think how I’d characterise what I do) is am I…
I lamented last year of the ìmissed opportunityî the music industry had in failing ot create a ìcooption modelî via Napster as a centralized downloading center.
This involved a “jiu jitsu” of the Napster sharing framework.
The goal of the model would be to turn the Napster network into a massive data mining / advertising promotion / sales machine.
Each aspect of this model emphasizes the accumulation, analysis and application of consumer music data. The first goal would be to find out a) what people are listening to; b) what else they might want to listen to; and c) extending their relationships with the artists whose music they appreciate.
Of course, any industry as technophobic and and paranoid as the music biz could never see through to a system such as that.
No matter. they are slowing mismanaging themselves out of existence. Lets just hope that whatever replaces them is more consumer friendly than the RIAA is (was) . . .
This is sort of a tangent as well, but at Zerophase we are working on a method that lets the artists get 100% of the sales of digital music downloads from our site, no matter how much they happen to sell. We will derive our revenue in other ways, such as merchandising, premium artist pages, sales of comp discs, etc… We will only offer it to our premium (paying) artists, so we can pay the bills. It’s looking good from the theoretical side, and hopefully the implementation will work well too. Bookmark us and check in from time to time 🙂
“Jetpacks with streaming audio” was my facetious suggestion on Slashdot a few weeks ago when someone asked “Whatever next?”
I’m not sure about the jetpacks bit, but I definitely like the idea of walking around like I’m my own personal radio station. I guess I would have a broadcast radius of about 10 metres (maybe that’s a bit too much), and as long as you’ve got a listening device that can tune in to my stream then you can listen to the tunes that I’m listening to. And if you like my choices, you might want to stick around me for a bit. But if all I play is teh suck then you can just head off and find some other mobile radio show/person.
And of course no need to ask what songs I’m playing – that info will just show up on your player.
Oh, but what if there are a lot of people all broadcasting at once, like on a bus or train? I guess maybe there would be signal strength issues, and maybe also there could be a way that the devices could negotiate frequency or something. And there’d be a note on the display telling you how many other streams are currently active in your area.
And finally – there would be a class of music players that would have Uber-narrowcast Status, so that if they sensed that everyone needed to get funky now! they could just override everyone else’s streams with one song guaranteed to get everyone dancing. This feature could be especially useful in times of trouble.
Hi Tom, sorry for using comments but I don’t know what I have done with your email address. FYI I answered Stephanie’s post (Climbtothestars.org) on my blog an hour ago if you are interested.
MusicMobs
In reading the comments to Tom Coates’ The New Musical Functionality…, I found a link to a site called MusicMobs. MusicMobs is a music information site which is driven by individual users uploading their exported iTunes music libraries. That means th…
MusicMobs
In reading the comments to Tom Coates’ The New Musical Functionality…, I found a link to a site called MusicMobs. MusicMobs is a music information site which is driven by individual users uploading their exported iTunes music libraries. That means th…
Notes on Tim O’Reilly’s Oscon 2004 speech
One good thing about audio on the Web is that I can listen to things while I’m working. Which is precisely what I did this morning with Tim O’Reilly’s keynote speech at the Open Source Convention currently being held in…
P2P killed the Radio Star
One of the biggest problems with ITunes is that despite easy downloads, one finds it difficult to get recommendations on songs, especially if you don’t like mainstream crap. But no more, thanks to new P2P tools like AudioScrobbler. Song Buddy,…
The New Muscial Functionality
Clay raps it out – some more….. Tom Coates has the first of what looks like a fantastic series of posts on the new musical functionality, an extended musing on the distribution of production, reproduction, and filtering of music, covering especially …
Audioscrobbler
Tom Coates mentioned the Audioscrobbler music recommendation service last month in the first of his series of posts about what he terms “the new music functionality”. The service was temporarily closed to new accounts while they upgraded their hardware…
I’m a total latecomer to this thread, but this is something that I began researching before I left Ivrea. (In fact, I’ve just posted much of what I’ve written here at Social Beasts, since my comment got really long!)
I’ve been very curious about how we build our identities via music, and how we learn about other people through their musical tastes. I’ve been in love with iTunes, Audioscrobbler and Last.FM since i encountered them. I’m really curious about how these applications will change our relationships in smaller, closed networks, and how this transfers onto cities and neighborhoods.
Ivrea is a small, closed community where people know each other face-to-face, and initiating contact with somebody is easy. Outside of the permanent residents, a variety of professors and visitors pass through the Institute. I realized that thanks to iTunes and the closed network, I was learning about my students, colleagues and school visitors in a way that I wouldn’t have an opportunity to do previously. It demonstrated that I had commonalities with people I would not have guessed, and gave me a basis for conversation about something else other than our typical relationships would have suggested (interaction design, teaching, what do you think of Italy, and so on).
Being a huge music fan, I’ve always done this by looking at people’s CD collections. But that requires a level of intimacy — you need to be invited into someone’s home, or be able to pick up their Case Logic, or ride in their car, or run into them at a record store or show.
As I interviewed people, I learned that for some, music is incredibly personal, “like family pictures,” said one student from India. A researcher from Venezuela said he kept his CDs in his bedroom because that’s how intimate music was to him: maybe (and just maybe) you’d get to see his music collection. A Japanese student said that when she first got to Ivrea, she would only share some of her music on her iTunes — she didn’t want to give people the wrong idea about who she was.
There were a few pivotal people that always came up in the interviews. Two British students had expansive musical tastes. One was known for the scope and size of his music collections (he had different ones: one at home, the other portable). The other, a very quiet guy, had a very eclectic collection — one that had many things that I had or would want to have. For many people in the community, they got to know a side of this student that they would have waited for longer in conversation, since he was rather shy.
A Stephen Aubrey and Katie Brown, students at Wesleyan, refer to this as Playlistism , where you make judgments about people based on what’s in their playlists. The Wesleyan Argus article looks at the pop-culture and negative sides of this, but I found it very positive and downright fascinating at Ivrea. iTunes was the killer community app, when it was possible to generate face-to-face context.
Music is, after all, a high-context thing. Looking at and judging someone’s musical collection requires knowledge about the music, the band, the label, and accessing one’s own connotations. Two students I interviewed, an Italian man and a Turkish woman, were similar to me in that regard: they are huge musical fans, and look at a collection as a whole to determine someone’s taste. (Which is to say a little BeyoncÈ or Justin Timberlake wouldn’t derail a collection, and would even indicate ironic pop sensibility).
Sharing music and “playlistism”
I’m a total latecomer to Tom Coate’s thread on the new musical functionality . This is something that I began researching before I left Ivrea. I’ve been very curious about how we build our identities via music, and how we…