Interview: Sean McManus on personalized music, customized books, and why he is using on-demand service Lulu -- and the background of his idea to write a novel featuring a mass customized service offering as its key element.
Sean McManus is the author of ‘University of Death’, a new novel satirizing the music industry. The book explores what happens when a major record label comes up with software for mass customizing music and uses spyware to sell it to customers, without telling them it’s all computer generated. Sean’s previous books include ‘Small Business Websites That Work’ and ‘The Customer Service Pocketbook’. As a journalist, Sean has written for Making Music, Melody Maker, Internet Magazine, Business 2.0, Internet Works and many more. And he has covered mass customization before: In May 2000 he wrote the mass customization essay ‘As you like it’ about for Personal Computer World magazine and in December 2005, he interviewed the company behind Erasure’s customized MP3s for his website at www.sean.co.uk.
Sean, Congratulations! You have written the first novel I know with strong references to personalization and matching-services in the music industry! What's the story?
It's a satire of the music industry, centred around one of the last surviving major record labels, Bigg Records. Clive Bigg is gobbling up independent labels and marketing lowest-common-denominator tosh made by boybands. It’s not enough, though, and like every other label, he’s seeing his business shrink away.
Then one day the solution arrives: a smooth-talking geek called Jonathan Harrington has spent ten years creating the perfect song: moving enough to make you laugh, cry, or dance on the first listen. The catch is that it’s computer generated and tailored for each listener after analysing his or her music collection. Together, Bigg and Harrington conspire to use hidden software to study what fans listen to, and then to automatically concoct and market their dream music to them.
While all this is going on, the story also follows the progress of Dove, who is burned out from touring for decades. He wants to break up his 'creatively bankrupt' band, University of Death, but he couldn't do a proper job. Now Bigg's bought up the indie label the band was on, he's about to make Dove an offer he can't refuse.
And the story also follows two of Dove’s biggest fans: Simon and Fred have a band called Goblin (performing a mix of rock and glam they call 'heavy tinsel'). Like many bands today, they can't get anyone to listen to them, and hope that Bigg will pluck their demo from the pile and launch their careers. As well as doing their own stuff, they cover University of Death in the hope that they'll catch someone's ear. As it turns out, their cover gets them into all kinds of trouble...
Dove, Simon, Fred, Jonathan and Bigg all collide in a finale that threatens the very existence of the music industry.
The story takes a slice through the music business: from the board room to the stage; from the studio to the record fair. It explores how fans relate to their favourite bands, how businesses use technology to manipulate consumers, and what would happen if the music industry disappeared overnight.
Where did you get the idea for this book?
In the 80s I remember typing in a program listing that created music on the Amstrad/Schneider home computer. It sounded a bit foreign and unstructured to me, but it started a fascination with computer generated music that I’ve had ever since.
In recent years, we’ve seen the internet become a channel for both marketing and market research. We’ve seen the rise of technologies that make mass personalisation possible. And we’ve seen record companies backed into a corner and taking desperate measures to prevent piracy, epitomised by Sony BMG putting software on music CDs which was widely considered to be spyware. We’ve seen the start of artificial intelligence as part of our e-commerce applications, with Amazon knowing my taste in books and music better than I do. And we’ve seen the rise of independent bands through communities like MySpace, where high quality music can be shared and sold outside the conventional music industry. All these threads came together in my plot. It’s a timely book. In fact, when the Sony BMG story broke, it felt like my plot was starting to come true!
‘University of Death’ is ultimately about why people love music, and where its soul is. The book explores the extent to which that can be automated or faked. As the story unfolds, it becomes clear that (in my novel at least), music needs to come from people, not machines. I listen to a lot of synthesiser music, but that works because there’s a creative person directing it and the computers are just being used as instruments. Even Brian Eno’s generative music, which is unique each time you listen to it, works because a creative musician has defined its parameters before it runs. The question is whether the software will one day be good enough that you couldn’t tell the difference between a computer inventing and performing a song, and real human creativity.
I know your early essay on mass customization (still a well linked source on the topic on the internet). Have you written any other books in the meantime?
I’ve written ‘Small Business Websites That Work’, published by Prentice Hall, and co-authored ‘The Customer Service Pocketbook’. There are free chapters to download from both at www.sean.co.uk.
Why did you want to write this book?
They say everyone has a novel in them. This is mine: it includes so many of the things I love - music, technology, record collecting, old computer games, jokes. But all of them in service of a story and bound together by a single theme. Everyone has something that they just know they have to do in their life, and writing a novel was one of mine.
It's been a long time since I've devoted that much energy to a single project, and it was extremely satisfying. I really enjoyed the writing sessions.
What are your observations on personalization of music in the real world? How often are you, as a consumer, using these services?
The most exciting thing for me has been Trust Media’s customised MP3s, made on-demand using a Flash interface. Erasure made best use of the concept: you could define what kind of beats, vocals, basslines and synth lines you wanted, as you heard the track looping. When you were done, you paid and downloaded your track. Each combination was limited to a single copy, and had unique artwork. The music industry’s been marketing so-called ‘limited edition’ CDs for years, with serial numbers on them often running into hundreds of thousands. This really subverts that: Having the only copy of my favourite version of a particular song and knowing nobody else can buy it is truly a ‘limited edition’. Erasure really appreciated what they could achieve creatively with this technology, and it would be good to see more musicians adopt it. Trust Media is pushing the antipiracy aspect at the moment: people are less inclined to share something that’s unique to them (and traceable), and others are more likely to want their own unique version than someone else’s copy. When the music industry is suffering a decline, it makes more sense for the company to sell antipiracy software than an experimental music format, even if they’re the same thing.
Brian Eno’s done some interesting work with generative music, where he sets the parameters of the work and then each performance is unique. There’s no computer creativity involved in this: it’s still very much his work, with the computer randomly generating each performance of what is essentially one work. His first release of generative music ran on floppy disk and the software is obsolete now, but his 77 Million Paintings software brings the idea up to date and combines it with visuals. It’s not really personalised, though, even though each performance is unique, because I have no control over it.
I enjoyed the music recommendation engine Pandora while that was available [in Europe], but that’s been closed to people outside the US now because they can’t afford to pay international license fees. Last.fm is a nice recommendation engine, but I haven’t used it too much. I still tend to find new music through magazines, reviews online, friends and gigs.
As with publishing, mass customisation has made it viable for bands to sell their own music on CD from the very start. I’ve bought a few CDs by unsigned bands which probably wouldn’t have existed without the mass customisation and ecommerce technology that was used to create and sell them.
And your book is not just on personalization and customization, but I saw on your website that you also are using a print-on-demand service (Lulu.com) to publish it. So why are you self-publishing 'University of Death', and why are you using print-on-demand?
The main reason for using Lulu as my publishing platform is that it enables me to get the book out there much more quickly. I have friends who have written great books and then spent years trying to get interest from a major publisher, while their books have quietly gone stale. I spent two years writing my novel, and I didn’t want to spend another two traipsing it around publishers who are already inundated with other good books. By self-publishing, I can ensure the book reaches readers much more quickly. Because the book deals with many contemporary themes in the music industry and technology, this was important to me.
For a venture like mine, it makes good business sense. There’s no up-front cost working with Lulu, and I don’t have to store hundreds of copies of the book under my bed or in my garage. The downside is that it’s massively more expensive per copy than it would be to do a conventional print run, but it’s an ideal way to test the market for new creative products. I particularly like Lulu because it takes care of the retail side of things too – it handles the credit card or paypal orders, customer service and support. It helps that Lulu tends to rank well in search engines too. Working with Lulu means I don’t have to be involved in handling individual book sales, don’t have to spend up-front, and don’t have to carry stock. It also means customers can have a smooth and fully supported buying experience.
And where can we buy your book?
Thanks for asking! This book is not available in the shops. You can only buy the book at Lulu.com.
When you place your order at Lulu, they'll print your copy, perfect bind it, stick it in a sturdy cardboard wrapper and post it out to you. This book is not available anywhere else because copies don't exist until they're ordered.
You can download the first two chapters for free through www.universityofdeath.co.uk.
To conclude: What is, in general and beyond your industry, the greatest mass customization offering ever – either one that is already existing or that you would like to get in the future?
I’m not sure whether it counts as my industry or not, but I’d like to see more done on books. Wouldn’t it be great if I could instruct an intelligent agent to create a book about ‘Pink Floyd’, or even ‘Dark Side of the Moon’, and have it deliver a unique printed artefact to my door? The software could source newspaper clippings and reviews from leading publishers, maybe some blog posts from well-respected fans too. It could sort them into chronological order, and source images from leading photo libraries. It wouldn’t be easy: there’s a whole rights nightmare to resolve, and the micropayments could prove tricky to administer, particularly once you get down to the level of paying freelance journalists. But if the infrastructure was there, the content would follow. And you could create an interface for narrowing the search to something useful (eg, let users specify publication dates, proportion of blog content to newspaper content, number of images etc). Books are still the best way to communicate and digest large chunks of information, but at the moment, there needs to be a significant market for each book to make it commercially viable. That’s because somebody has to do the leg-work of writing each one, and someone else has to market and distribute it. If you want a book about 90s band ‘Kenickie’ (as I do), you’re probably the only one, so you’re out of luck.
We can already do much of the stuff required: we have good search algorithms, there is a lot of tagged content out there, and there are applications that create PDFs on demand, and others that print them in books. We already trust search engines to decide what content we should see online, so this would be an extension of that and would probably work best if restricted to trusted content providers named up-front. It could be a great way for rights owners to make money from archive material and for researchers or enthusiasts to access original reports from the archives.
This is all probably some way off. Still, I can recommend a nice novel to read in the meantime… ;-)