Monday, December 03, 2012

Finding Digital Needles In Digital Haystacks

Remember those dark days before ONIX standards, search inside the jacket, or even picture of the jacket?

Contextual data , or as it in know, Metadata, is produced to help consumers find the book, validate it and also to promote it. When the physical bookshelf was the only option, then we relied heavily on the touchie, feely approach and the promotional sell was to the bookshop. After all,'if it isn’t on the shelf you can’t sell it' and bookstores often have finite shelf space at all levels. Then came the internet and it wasn’t so much about getting it on the shelf, but making it visible amongst the hundreds of thousand others. The onus shifted from supplying the retailer with basic information to supplying the consumer with rich information.  Now we are entering a new phase where the book starts to promote itself and it only the consumer that counts.

The new industry metadata standard (ONIX), has helped define the basic metadata and also its adoption across the supply chain. It served the intermediary world well , but is it enough in a consumer only world where titles are effectively on consignment and don’t have to be sold to the retailer? 

Many will argue that standards and agreed data structures are a must and without them books will just be lost in the new virtual space. Others will counter saying that they all too often act as a straight jacket and are still supply chain focused and not consumer orientated. Somewhere in between lies the reality, the challenge and the opportunity.

One of the greatest constraints the standards give us is on genre classification. We appear to have created the standard ‘tree, branch, twig’ hierarchical approach to genre classification. This was great when you had one book and only had one slot in a bookstore to display it, but is it really as relevant in a virtual bookstore with infinite shelf space where the book can sit in literally thousands of relevant slots? Is it relevant when digital content itself can define in which genres it belongs? Is it relevant to the consumer who is constantly finding and redefining genre and can’t wait for the standards bodies to endorse the name? After all, by the time new vocabulary is defined in the authoritative dictionaries, it already has been widely adopted and used.

So we now have a truly ‘mixed’ economy comprising, physical books sold through physical stores, physical books sold through digital stores and digital books sold through digital stores. Should the same construct serve all three channels, or should the physical model act as a loose base, but we adopt more appropriate and consumer facing methods for the digital world?

Amazon Kindle only allows two ‘dry’ classifications and seven keywords, Kobo is more tolerant and PubIt is only just about to finally go international.

Are genres truly hierarchical or now lateral? Are keywords now more important than the headline genre? Do we find ‘books like’ through sales or more sophisticated means? Do we enable keywords to be excluded as well as included? How do we express multiple demographic appeal? How do we rate relevance? How do we rate or harness genuine reviews? How do we use the content itself to define its appeal? Do new books always come before older ones?

We don’t have the answers, but we do realise that today’s cumbersome expressions are frustratingly restrictive and far from being engaging and in some cases relevant. They provide a base in the digital world, but sadly not much else.

Today, finding a new unknown title in these silos, is like finding a digital needle in a giant digital haystack.

No comments: