Interaction designer – Hannah Donovan

AUTHOR:  
Published:  August 23, 2011
Interaction designer – Hannah Donovan

Hannah Donovan is a Canadian interaction designer living in London. She led design at Last.fm for five years, and these days works as an independent product designer. Hannah is speaking at the Web Directions South conference this October in Sydney. She took time out to speak with desktop.

What was the state of music on the web when you joined Last.fm and how does that compare to when you left?
Well, music on the web in 2006 was a pretty crazy place. It was hot on the heels of the wild west P2P days of the early naughties, and only a handful of companies were experimenting with how to stream music online. Streaming music was (and still is, to a much lesser degree) fraught with legal and technical difficulties. At the time I remember debating a lot as to whether people would ever be able to ditch their ideas of ownership for a music collection in the cloud. Now it seems it’s finally happening – my early adopter friends seem happy to wipe their iPhones of MP3s and listen to Rdio or Spotify playlists. Compared to this behaviour, CD collections seem almost quaint.

But streaming wasn’t ever the most interesting thing to me. I’ve always been much more fascinated with the things you can do around music besides just listening to it. The social stuff: discovery and recommendations. Back then, Web 2.0 technologies were just taking off, and that was really exciting for music specifically. Things like tags and other user generated content let people interact with, and explain music in their own words. For the first time (since the birth of the recording industry), music felt like it was in the hands of the people. And for the first time on the web, music was social. It all felt so right.

Apparently it made sense to a lot of other people, because around 2007, the space absolutely exploded with competition, and everyone wanted in on music, and that was when the biggest shift happened – MySpace. They didn’t just let you socialise around music, they let you make music a central part of your identity. Direct relationships between these identities formed and another big first happened: relationships between artists and fans. Now, all these features are commonplace for even the tiniest music blog, and come out-of-the-box with most services.

The one thing that hasn’t changed much in terms of mainstream perception is recommendations. Technically, it absolutely has – the discipline of Music Information Retrieval (MIR) really came into its own right over the last few years, with scientists exploring how to retrieve information from music. At Last.fm, the building block for all this is a “scrobble.” Every time you listen to a track, it “scrobbles” to Last.fm – a bit like an automatic status update for what you’re currently listening to. Based on the scrobbles you build up over time, Last.fm recommends you more music you might like, similar to how Amazon recommends products.

Scrobbling was always a challenging concept to explain, and I really don’t think the majority of people were quite ready for this back in 2006. Now, however, I do. Services like Facebook, Twitter and Foursquare have popularised the idea of broadcasting what you’re doing, and a product that broadcasts without you having to even do anything is something I can see people using now. Half of me wishes I could get in a time machine and redesign the experience of scrobbling for 2011, because I think we’re ripe for it!

Can you talk about some of the design challenges you faced with that service?
One of the very first design problems I was tasked with when I joined was ‘how to get people to download the software’. Especially one that tracked what you listen to! People would worry it was spyware, they would wonder it was going to upload their files to the web, or they simply wouldn’t know how. Back then, downloading software was something people were still a bit suspicious and/or unfamiliar with. Transitioning between installing something on your computer and putting a password into a browser window was an excruciatingly difficult user experience to design for – especially on a Mac, where up until recently, you were required to have some knowledge of where to put the application when it downloaded.

This was made even more difficult by the fact that people didn’t really understand what they were downloading until they saw a scrobble happen. We called this the magic moment – when users saw the name of what they were listening to on the web, their faces would light up. It was always amazing to watch. But getting them to that moment was very, very hard.

Due to its organic growth, Last.fm had a lot of unique terminology and concepts too – scrobbling, radio stations, tags, friends, neighbours – there was a lot of explaining to do. If nothing else, my copywriting skills improved a lot over those five years! My recommendation for anyone in a similar situation would be to only have one conceptual model you need to explain. More than that is too many.

Visually, music is one of the hardest things to design for, because it’s an inherently auditory medium. Almost any style or aesthetic you can think of that looks and feels like music actually feels like a genre of music, which was something I was adamant about staying away from. Last.fm is a genre agnostic service – the graphic design of the site has to look comfortable supporting the content of a death metal band, a folk singer or a grime artist. Unless you’re targeting a particular genre, I think this is an incredibly important detail that is hard to execute with personality and warmth. Of course, factor in the localisation in 13 different languages, the variation in amount of user generated content (no content versus lots of content) and executing this brand with a native feel across the web, desktop, mobile, XBox and other devices and well, it keeps you on your toes.

Image by Jeremy Keith, taken from Hannah's blog – blog.hannahdonovan.com

What do you think will be the next big revolution in the world of music?
On the technical side, I think that cloud storage and on-demand streaming (even in offline contexts) will become commonplace. I also think that we’ll see the mainstream adopt some kind of automatic ‘status update’ concept around music for you to share what you’re listening to with friends. I think we’ll see more charts and graphs comparing you to your friends, city, country and the world – because loading your music taste profile in a faction of seconds and comparing it to others is now a much easier thing to do.

Culturally, with music becoming more intangible than it already is, concerts and concert tracking and ‘check ins’ are only going to become more important. Concerts (as well as collectable merchandise and vinyl) are the only ways we can still ‘touch’ the music.

But most of all, I think what we’re seriously missing right now is a way to express ourselves musically. I still don’t have a good way to share the music I love with the people I love, in a way that isn’t another sterile feed of content. People trade YouTube links on Facebook, tweet what they’re listening to, post MP3s on Tumblr, but in terms of self-expression, these methods really pale in comparison to what we experienced when MySpace launched. Infamous for the barrage of sensory overload emitted from your browser while you attempted to load a page, it’s also worth remembering that MySpace was a site that created culture; musicians became famous, whole new trends in graphic design and fashion emerged. With all the pre-fab tick-the-box profiles we mindlessly go through the motions of setting up (or ‘connecting’ to import our profile data from an existing service), our musical identities have been reduced to a set of form fields.

I think we’re hungry for tools around self-expression, curation and editorial. Not in the way we shudder to remember them from the early days of the web, but done intelligently – in a way that satisfies people emotionally and computers technically. People crave beautiful experiences, and computer brains need to be able to parse that beauty in ways that let us continue to do exciting things with the data we create. The next music revolution will be one that creates culture, not just spreads it around more easily.

You’ve done a bit of work with the Kinect interface. What was that like, what was the biggest challenge in porting a web service to that medium, and do you think we’ll all be browsing the web in the future using a similar interface to the Kinect, Minority Report-style?
Absolutely not! All you have to do is try holding your arms out in front of you for a minute to see why. I do think the web will be something we increasingly interact with through touch screens and other gestural interfaces, but if we’re doing anything with our hands that’s more complicated than paying for our groceries it will be on a horizontal surface. Working on the Kinect was interesting for other reasons – the voice commands for instance, were easily the coolest feature – we love technology that lets us be lazy! I’m really interested in design experiments around more natural full-body gestures we already do, such as nodding / shaking one’s head, the use of sound-based controls and haptics.

Any tips for designers who are finding themselves faced with having to design for a device that is new and without established conventions?
You really have to trust your gut – but your gut feel should be informed by specialised knowledge about how your users behave. It’s your responsibility to find out as much as you can about them. There’s loads of great reading material on the subject of user research and user observation, so I won’t go into detail about that here, but if you’re new to this area, I find myself constantly going back to these four simple skills:

  • Notice everything. I watch people all the time. I listen to how communicate, I watch how they’re interacting with their devices, I notice what they wear and what kind of phone they have. Yes, it’s probably borderline creepy, but it helps me do my job. It’s this anecdotal observation that will come in handy when you’re racking your brain on a design problem.
  • Be curious. It’s not enough to just watch how people do things, you need to be an active, engaged observer. Ask questions: why do they use that phone? Why do they communicate in that way? What statement are they making with those clothes? Become a sharp critical thinker as you watch.
  • Be a good listener. People would often tell me about their problems and experiences with Last.fm – it was fantastic, I would just sit back and listen, and then scribble down notes later when they weren’t watching. If you’re lucky enough to be in a situation like that, solicit that kind of feedback. If you’re not in a situation like that, then you have to go find people to interview – but I like to think of this more as an opportunity for them to tell you their stories. Make them feel comfortable, ask a few leading questions about whatever it is you’re curious about and sit back and listen.
  • Try things out on people. Nothing beats someone actually using your product. Do this early and often, and while doing this, remember the first three skills: notice everything, be curious, listen well.

But when all else fails, trust your gut and stick to it. Your gut will tell you to take risks and do bold things, and you should listen to it. More often than not, it’s right. It’s always better to make something that elicits a strong, “I love it / I hate it” response than a neutral, “Meh, it’s OK” response. Don’t do anything half-way; mediocrity doesn’t teach you anything.

You’ve made it known that you like to touch things! Aren’t you in the wrong job? I read somewhere that you got into web design because working with print scared you, yet working with print would allow you to explore different types of stock and texture … how do you go about bringing these tactile yearnings to your work in the online space?
Yes, I do like to touch things! My friends jokingly call these moments when I stop to inspect a furry plant or coat “Rain Han” moments… but I’ve always loved textures and they’re a big influence on my design. It’s been really interesting seeing Apple (and the rest of the world) turn to paper, wood and fabric for inspiration lately; it makes sense. The web is becoming more tactile. Designers are looking for ways to represent that.

Understanding how textures look and feel in real life and recreating them to make virtual spaces easier or more fun to navigate is something I get way too much enjoyment out of. I’m a stickler for consistency across textures and lighting effects in an interface. Also, the web is becoming more physical too – embedded into everyday objects, and the lines between industrial design, interface design and user experience are becoming increasingly blurred.

I’m always fascinated by hearing about people’s individual processes. You recently shared your design process using a ‘texture garden’ analogy. Could you talk about that a bit? Any tips for budding designers who may feel like their own process is still up-in-the-air or evolving?
Use the right tool for the job. There is no single formula for process, and mine varies wildly for each project I do. I think the idea of a single set of actions that gets applied to every project is crazy. Sure, there are some broad categories that are almost applicable, but beyond that, it’s about looking in your toolbox and choosing the best tool for the job. Every time you stumble upon a tool that worked well for something, no matter how small, write it down. Over time, you’ll have a toolkit you can refer to when you’re unsure where to turn with the next step in the process.

If you’re ever in a situation where someone is telling you what they’d like to see next in terms of process (e.g. “Next I’d like to see some wireframes”), ask yourself if this is the right tool for the job. Sometimes the answer will be no and you’ll need to pipe up. Projects don’t usually fail because of the people, they fail because of the process. Nobody knows what you need to design better than you, so look in your toolkit and propose another tool.

I have one rule for working with clients and stakeholders: Tell me what you want and why you want it. It’s my job to figure out how to do it. I don’t work with people who can’t follow that rule!

You could save $100 off your ticket to see Hannah speak at Web Directions South on the topic of Designing Without The Browser, but there’s only a week to go! Now is the time to convince your boss that attending the web conference of the year will help you and your organisation. Register here.

One Response

  1. I saw Hannah speaking at the dConstruct conference in Brighton last year and am looking forward to catching her at this years Web Directions South. Highly recommend attending if you can.

Leave a Reply

Your email address will not be published. Required fields are marked *