For the Love of Metadata

In Laura Dawson’s chapter in Book: A Futurist’s Manifesto (2012), she addresses “What We Talk About When We Talk About Metadata”.

Beginning with a brief history of metadata, Dawson notes that, “because a book is no longer a physical object, discoverability via metadata is only just now becoming a front-office problem.” She explains how this coincides with Brian O’Leary’s chapter about the “book” no longer being the physical container that holds the content. I think this also integrates with the marketing and discoverability of a title has become more of a front-office problem. For example, if a potential author doesn’t have an existing web presence, that is considered a huge problem for a publisher. The author is expected to take on a marketing role, and without having a platform to do so, they are facing a steeper climb to discoverability.

… most book sales are now happening online. …if there is insufficient (or inaccurate) metadata for those books, consumers simply will not find them. The publisher (and retailer) with the best, most complete metadata offers the greatest chance for consumers to buy books. The publisher with poor metadata risks poor sales—because no one can find those books.

Aye, there’s the rub. Dawson hit the nail on the head with this statement, and if I had to highlight a thesis for her chapter, this is it. With online book sales becoming increasingly common, as well as discoverability being primarily online, and good metadata is at the root of that.

Dawson’s statement raises all sorts of questions for me, the biggest of which is regarding the work required to create complete metadata, and how that factors into the workflow of a publisher. These are often considered “boring” data entry jobs, shunted off to interns or junior staff, which may mean they don’t fully understand the importance or the system.

At the same time, Solomon notes, “lots of metadata” is not the same as “good metadata.”

This is so key. If you just stuff your metadata full of keywords and every possible detail, the value decreases.

First of all, the end consumer is the person that matters, not some search engine that is crawling your metadata for keywords. If you put in keywords that attract a user, but don’t deliver, they will leave.

Second, if you stuff your metadata so full of information—tiny details, every review quote, all the praise and blurbs for the author’s previous titles—the user is going to get bogged down. We are frequently being told that a reader’s attention is grabbed within the first 30 seconds; this is even shorter on the web. If you laden your metadata with unnecessary details, then you may overwhelm the reader.

I think a lot of people forget this when talking about metadata and book “discovery”: the reader is your ultimate audience. They talk about algorithms and search engine optimization, and keyword rich content, but they forget that it’s ultimately the user/reader/consumer who will determine if the content is valuable to them. This value will be determined by a book sale, or perhaps by eyeballs and time spent on a page.

Also, there are things that readers need to assess the value of a book online that they would be able to do in-person, such as table of contents. This is where Amazon’s “Preview” and GoogleBooks come in really handy for users—they can “flip through” the book like they were in the store in order to determine its usefulness and value. I’ve used these functions (particularly GoogleBooks) when writing essays for class. I can search for a term, or look at the table of contents, find out if the topic I need is discussed in detail, and then acquire the book if it is.

I really appreciate how Dawson seems optimistic about the fact that there is no universal standard for metadata. I am really tired of all the “doom and gloom” in publishing. Perhaps I’m defensive of traditional publishing, or perhaps I just don’t think that bemoaning the fact that things are changing is going to solve anything. It’s possible to acknowledge that the marketplace is changing without sounding so dire.

This digital flexibility makes it possible for a single ePUB file to contain a MARC metadata set, an ONIX metadata set, and a proprietary, consumer-friendly taxonomy that only the publisher’s website can render.

This makes me question about the amount of work required to create all these metadata files. If I understand correctly, by majority ONIX is the standard for metadata. So why have so many other types of metadata sets proliferated? Was ONIX not meeting all the requirements for users? I realize that Dawson addresses that concern to some degree by stating that no single metadata schema describes a book to the full satisfaction of all parties, as it would be bloated, ineffective, and would require frequent updates.

But I’m still curious about the amount of work required for the publishers (traditional or otherwise). Do they have one in-house system that has fields for all the potential outputs requested? What about smaller publishers? Do you have to complete metadata as a task off the side of your desk, or do you assign it to an intern who doesn’t fully understand its value? And doesn’t that create a further inequality where publishers who can’t spend a lot of time inputting the metadata into a schema will ultimately lose out on discoverability? I think these questions are really valuable to consider, and perhaps there is no clear answer yet. Overall, Dawson gave a great overview, addressing a lot of key points, and raised a lot of salient questions for me.

One Reply to “For the Love of Metadata”

  1. I’m glad you found the reading helpful, as I agree it does provide a good overview. Thanks for the summary of the key take aways, as well as for adding in your own thoughts.

Leave a Reply