Posts Tagged ‘discovery’

  • We love you Spotify, but wouldn’t it be great if…

    There’s no denying that Spotify has taken the world of online music to a new (legal) level. Their ever-increasing catalogue of artists and albums, coupled with the acceptance of their app into the iTunes store will, no doubt, alter the way millions of people both pay for and consume music.

    The application is a model of elegant simplicity. It easy to find your way around the service, listen to music and create and share playlists.
    However, there are times when you open the app and you have no idea where to start. To torture the metaphor – you have a potential fire hose of music content but turning the right tap can sometimes be difficult. What do I fancy? What’s my mood? What am I doing? Do I just want some background noise or some hum-along favourites?
    And there’s the problem. Discovering music on Spotify is difficult. They have the ‘Top Lists’ tab but that’s far too coarse to be an effective browsing or recommendations mechanism. And the chances of the ‘What’s new’ tab giving me a starting point for listening is remote at best.
    So I thought it would be a great idea to look at some ways to enhance Spotify. To devise some mechanisms which could effectively surface a wider range of their catalogue of tracks and help me decide both what to listen to and discover new music.
    I will caveat at this point that what I have done is by no means exhaustive. It’s not a usability review or an interface rework. It’s an idea, a sample of how we believe the Spotify experience could be extended and enhanced. Also, I have the luxury of not having to work out the technical feasibility of these ideas. Although it’s all ‘doable’ no account has been taken with regard to performance issues or other potential technical constraints.
    So, with that sorted out. Let’s take a look at some ideas. First, the ‘home’ screen.
    Instead of the ‘Artists you might like’ panel I feel there should be more of a ‘Your trends’ panel. This is an easy way to get back into music you’ve listened to. A 30 day (or longer) retrospective on your listening habits. Like what you put together last Wednesday. No problem. Click the chart to line up the tracks again and away you go. Also, should Spotify wish to develop more community features longer term this could be a great way of comparing tastes and habits with other users.
    <img>
    As I mentioned earlier, sometimes you just need to listen to music but have no idea where to start. Artist, genre or album just doesn’t always cut it. The creation of a mood browser would open up a whole range of music unconstrained by genre or artist. Clicking one of these links could take the user to a screen showing maybe 100 tracks as a starting point for listening.
    <img>
    One of the great things about digital music is that the old barriers and constraints of albums have disappeared. A playlist generator would work by allowing the user to input some simple criteria and generate a body of music. The way I envisage this working is that the user could input as many or as few criteria they like. Add some genres, add a couple of artist names to ensure they get added to the mix, set how long you want the playlist to last, hit generate.
    <img>
    Outside of your own experience (the middle of the screen) the right hand side could provide a glimpse into the activity of other listeners by location. Spark lines accompanying each track could click through to show listening trends for that piece of music. How has it’s popularity risen or declined over time compared to other tracks. Providing feeds from other services such as We Are Hunted also gives the opportunity to find out about emerging bands and tracks which may be popular for a very short space of time but may never appear on the ‘Top Lists’.
    <img>
    Once you’re off the home page and listening to music there are some other problems with the current experience. Related artists is, in my opinion, currently very inaccurate. And, once you have listened to an album from an artist there is no way to extend that experience. It’s back to search or one of the lists pages.
    Creating a side bar which could accommodate useful information such as artist influences and a ‘likeness’ finder. Influences could be real i.e. from the artists themselves as opposed to generated. And the ‘likeness’ slider could find artists or tracks which are either very related or completely different to what you have just listened to.
    <img>
    Surfacing the rich data that Spotify holds, there’s potential to expose data around an artist. Their maximum number of simultaneous plays, how often their tracks appear in playlists, how many total plays their tracks have had. In fact, there’s probably a whole data visualisation project right there.
    Anyway, back to music discovery.
    What I’ve done here is just a few quick ideas to release the potential for a much richer experience facilitated by opening up Spotify’s data. More useful music discover could also convince more people to sign up for the Premium account. Or even add a middle tier account which allows rich music discover coupled with some social features to allow easier sharing.
    Here are the full-size screen shots of the ideas.
    <img>
    <img>
    To stimulate the debate and because I, in no way ‘own’ the Spotify design, here is a link to the PSD files I created to do this exercise. If anyone wants to pick up the mantle, run with it, remix further then be my guest. Just be sure to post a comment with a link to your work.

    The application is a model of elegant simplicity. It easy to find your way around the service, listen to music and create and share playlists.

    However, there are times when you open the app and you have no idea where to start. To torture the metaphor – you have a potential fire hose of music content but turning the right tap can sometimes be difficult. What do I fancy? What’s my mood? What am I doing? Do I just want some background noise or some hum-along favourites?

    And there’s the problem. Discovering music on Spotify is difficult. They have the ‘Top Lists’ tab but that’s far too coarse to be an effective browsing or recommendations mechanism. And the chances of the ‘What’s new’ tab giving me a starting point for listening is remote at best.

    So I thought it would be a great idea to look at some ways to enhance Spotify. To devise some mechanisms which could more effectively surface a wider range of their catalogue and help me both decide what to listen to and discover new music.

    I will caveat at this point that what I have done is by no means exhaustive. It’s not a usability review or an interface rework. It’s an idea, a sample of how we believe the Spotify experience could be extended and enhanced. Also, I have the luxury of not having to work out the technical feasibility of these ideas. Although it’s all ‘doable’ no account has been taken with regard to performance issues or other potential technical constraints.

    With that sorted out, let’s take a look at some ideas. First, the ‘home’ screen. For a start, you’ll notice that I’ve relegated the ‘What’s new’ to a different tab. I’m not sure if there is record label pressure commercial reasoning for this list but it seems like a bit of a space filler to me. Anyway, for the purposes of this exercise, it’s gone.

    Your activity
    Instead of the ‘Artists you might like’ panel I feel there should be more of a ‘Your trends’ space. This is an easy way to get back into music you’ve listened to. A 30 day (or longer) retrospective on your listening habits. Like what you put together last Wednesday? No problem. Click the chart to line up the tracks again and away you go. Also, should Spotify wish to develop more community features longer term this could be a great way of comparing tastes and habits with other users.

    activity

    Play by mood
    As I mentioned earlier, sometimes you just need to listen to music but have no idea where to start. Artist, genre or album don’t always cut it. The creation of a mood browser would open up a whole range of music unconstrained by genre or artist. Clicking one of these links could take the user to a screen showing maybe 100 tracks as a starting point for listening.

    moods

    Playlist generator
    One of the great things about digital music is that the old barriers and constraints of albums have disappeared. A playlist generator would work by allowing the user to input some simple criteria and generate a body of music. The way I envisage this working is that the user could input as many or as few criteria as they like. Add some genres, add a couple of artist names to ensure they get added to the mix, set how long you want the playlist to last, hit generate.

    playlist

    Outside of your own experience (the middle of the screen) the right hand side could provide a glimpse into the activity of other listeners by location. Sparklines accompanying each track could click through to show listening trends for that piece of music. How has its popularity risen or declined over time compared to other tracks? Providing feeds from other services such as We Are Hunted also gives the opportunity to find out about emerging bands and tracks which may be popular right now but may never appear on the ‘Top Lists’ longer term.

    SpotifyHome_d01

    A reworked Spotify Home Page

    Music related to an artist or band
    Once you’re off the home page and listening to music there are some other problems with the current experience. Related artists is, in my opinion, currently very inaccurate. And, once you have listened to an album from an artist there is no way to extend that experience. It’s back to search or one of the list pages.

    Creating a side bar which could accommodate useful information such as artist influences and a ‘likeness’ finder would help prevent the current staccato experience. Influences could be real i.e. from the artists themselves as opposed to generated. And the ‘likeness’ slider could find artists or tracks which are either very related or completely different to what you have just listened to – updated in real time as you move up and down the scale.

    artist

    In surfacing the rich data that Spotify holds there’s potential to expose some very interesting content associated with an artist. Their maximum number of simultaneous plays, how often their tracks appear in playlists, how many total plays their tracks have had. In fact, there’s probably a whole data visualisation project right there.

    But that’s a different blog post. Back to music discovery.

    What I’ve done here is just a few quick ideas to release the potential for a much richer experience facilitated by opening up Spotify’s database and manipulating the content in more interesting ways. More useful music discovery could also convince a greater number of people to sign up for the Premium account. Alternatively, it could provide a case for a fruitful middle tier account which allows rich music discovery coupled with some social features to allow easier sharing but is still largely display-ad supported.

    What I’ve discussed here is just a starting point. A catalyst for ideas to enhance what is already a great service.

    SpotifyArtist_d01

    A reworked Spotify Artist page

    To stimulate the debate and because I in no way own the Spotify design, here is a link to the PSD files I created to do this exercise. If anyone wants to pick up the mantle, run with it, remix further then be my guest. Just be sure to post a comment with a link to your work.

  • Serendipity… WTF?

    A tweet randomly flew through my cascading deck over the weekend, containing a link to a blog I’d never visited before.

    Contra the NY Times today, why the web is the greatest serendipity engine in the history of culture: http://bit.ly/yZNa2 about 16 hours ago

    The blog post is by Stephen B Johnson and articulates a frustration I’ve been struggling with for a long time. I don’t get this idea Web technologies are responsible for a decline in ‘natural’ serendipity.

    My uncle Ben described the declinist position in a blog post he wrote some time ago at BBH Labs:

    It’s a world of perfect targeting. Optimization. Zero wastage. Absolute utility. Total accountability.

    More and more of what I see, hear, read and even taste seems exceptionally cunningly targeted at me. My RSS feeds me handpicked news streams. I get perfect movie recommendations via Netflix, books I’ll enjoy via Amazon, uncannily relevant advertising when using Gmail, weirdly familiar music from Last fm. Satnav keeps me resolutely on the data-derived optimum track. And so on.

    Well, these shifts are triggering a smoothing out in our experiences, prompting a reduction in serendipity and introducing a spooky predictability to many facets of our lives. It’s becoming clear that ultra relevance comes with a hidden price. Because if everything’s relevant, then nothing’s unexpected, and if nothing’s unexpected, then nothing surprises you, and if nothing surprises you, then that’s a strange, neutralized, vanilla kind of life to lead. Think John Anderton meets Truman Burbank.

    We’re talking about the end of surprise.

    Of course, this is all true.

    We *do* live in a world that’s increasingly like the one Ben describes, but I haven’t noticed any decline in ’surprises’ yet, and like Stephen I have absolutely no difficulty in discovering weird new shit ALL THE TIME to look at online. In fact, my biggest problems right now *by far* are:

    • A super-abundance of weird new shit to discover
    • Not enough time to do so, and a resulting daily sense of crushing regret
    • Poor attention literacy, manifested by a willingness to be distracted by new stuff discovered by accident through practically every interaction I have online – this is especially serious now that I have a torrent of links pouring into my eyes through my various iPhone apps and desktop, and emanating from a network of 483 ‘friends’ on Twitter

    The declinists argue that our ‘human’ need to discover things through serendipity is thwarted by emerging recommendation and discovery technologies but, like Stephen, I feel that it’s completely the other way around.

    The declinists point to a mythical golden pre-Web era of serendipity. They say that the way people read newspapers in the old days supported serendipitous discovery far better than a website can. They claim that the experience of discovering music through radio and club DJs was more serendipitous than the experience provided by online music sites. They seriously believe that bookshops and libraries made it easier to discover knowledge by accident than the Web can.

    I say that’s rubbish and the declinists are 100% wrong. Serendipity is boosted in the age of super-abundance

    The word itself is fascinating. It’s a lovely word isn’t it? It feels good saying it. It’s one of those graceful, slightly mysterious words that invests the one who utters it with depth and a sense of sophistication. The word is unnaturally powerful, magical – almost totemic to its acolytes and disciples, let’s call them the serendipitards.

    Its etymology is fascinating and perhaps explains its mythic overtones. The English author Horace Walpole introduced it into the language in a letter in 1754 with the following explanation:

    “this discovery, indeed, is almost of that kind which I call Serendipity, a very expressive word.”

    Walpole created the word from an old name for Sri Lanka, Serendip and explained that this name was part of the title of:

    “a silly fairy tale, called The Three Princes of Serendip: as their highnesses traveled, they were always making discoveries, by accidents and sagacity, of things which they were not in quest of….”

    I met quite a few serendipitards earlier this year at SxSW at a talk billed as ‘Music 2.0′ by Elliott and Sandy Hurst of Supernova.com. The talk’s premise was that the new era of unlimited catalogue and algorithmic recommendation simultaneously provides too much choice AND makes discovery less random than it was in the age of scarcity, and life is therefore less pleasurable.

    As this blog post relates, I became involved in an  argument in which several of the audience – including me – attacked the panel.

    Firstly, I made the point that algorithms are tools: you can use them for good or for evil. In fact, Last.fm had (maybe they still do) a recommendation engine that you could crank up from ‘just like yours…’ to ‘almost totally random’ (the recommendation equivalent of Spinal Tap’s ‘volume 11′ setting).

    Secondly, I tried to argue that the new super-abundance of unlimited catalogues cancels out the predictiveness of algorithmic recommendation, and that far from being a paralysing nuisance, abundance is a good thing. In evolutionary terms surplus is definitely “a good thing”. The myths and stories of “the ancient ones” demonstrate that pretty much every global culture (other than ascetics, puritans, Taleban and other self-punishing guilt freaks) fantasise about treasure caves, El Dorado, overflowing cups, feasting, milk and honey and cornucopias. The afterlife envisaged by pretty much every world religion is characterised by abundance – and that’s because for most of human history life has been lean, mean, brutal and short.

    So, what’s with this “too much choice” whinging? I don’t get it. At all.

    Not only that, but it’s a really boring, college debating society type argument. Ultimately, it comes down to a philosophical point about how random anything engineered by mankind – in this case an algorithm – can really be. That’s about as interesting as debating how many angels you can dance with on the head of a pin.

    Lastly, I would just point out that this is not an issue we hear real users complaining about. Ever.

    I challenge the serendipitards to go out and discover some real users who are worried about super-abundance and the decline of serendipity. It is simply not a real part of any normal person’s experience. Normal people want more of everything and some tools to help them filter the stuff they might like.

  • Follow me: Seven online communities I’m on right now

    If you’re going to have any success building healthy, active and engaging social communities online, you’ve got to be endlessly curious about how they work. Why is it that some communities gain new users every day while some dry up like a Texan graveyard? Why do you suddenly feel like making movies when you’re on Vimeo, while you just feel daunted at the task when you’re over at YouTube? Why do I feel so alone on a travel site like Dopplr, while I’m constantly meeting people all over the world at a photo site like Flickr?

    It’s not enough to read case studies or understand the technology behind the multitude of online communities out there to find the answer to those questions – the social interactions, exchanges and content creation will always form the core experience of any community. So the only way to really understand what makes people tick when they join a community is to hit the sign up button and get involved…

    Here’s a mix of some of my older and recent discoveries that you might want to explore, too…

    7: Second Brain

    Hands up if you already broke that new years resolution on getting your online presence organized… (I did). Second Brain promises to get you sorted by managing your files, bookmarks and feeds from the content you produce on other sites. It’s also great for discovering stuff other people collect. Feels a bit like a combo of Twine, Del.icio.us and Stumble Upon all at once…

    6: Shopstyle

    Ever had an amazing idea – only to discover that someone else beat you to it…? When it comes to fashion, Shopstyle got there before I could. I’m not very sad about it though – I could spend hours here!

    ShopStyle lets you put together entire outfits with just a few clicks – mix and match between your favourite brands. You also learn a lot from looking at how other women combine their looks. I’ve only got two faves on my outfit, though (hint, hint!).

    Dorothy Perkins at ShopStyle

    5: JustMeans

    This is a site I joined before I found my dream job here at MadebyMany. JustMean’s mission is to “provide best in class services that help companies communicate and implement their socially responsible efforts on-line.” They’ve got plenty of interesting job listings if you’re looking for work and you can network directly with companies and other job applicants who all publish streams of great editorials, news and videos.

    4: Brightkite

    Brightkite lets you share your whereabouts with friends. If you’re looking to meet new people at your current location, Brightkite will let you do that too… Feels a bit like Twitter with a post code locator.. Needs new users to become truly useful. Whenever I try to spot new people near my location, no matter where I am, I end up with Stuart, Tim, Matt or Oli on my radar…. (always glad to see them though!)

    3: Trendhunter

    Trendhunter is a bit of a mess for my taste. But there are so many interesting things to discover there. For the record, I don’t consider myself a trendhunter – I’m just lurking around, waiting for other people’s finds… such as this fairytale bedcover...

    2: Imagekind

    Imagekind is an online community for selling and buying art. I’ve stuck some of my photography from Flickr on there. Haven’t sold a single print yet but already gotten 41 fans… hum!

    1:  Vimeo

    If you’re into making videos, this is the community for you! What makes this community so great is the groups you can join – people create groups based on their interests – such as animation, conferences, random projects .
    I’m writing a separate blogpost on Vimeo in the near future – so check back for that if you’re interested – in the meantime, enjoy this poetic piece…

    Gorge Fog from Andrew Curtis on Vimeo.

    As always – please leave a comment if you’ve got any communities or anything else to share…

Our latest tweets

Categories

Archives

Find us on the web