Book Reviews: How The Web Reads 'Em, via ReadWriteWeb

Why Google and Other Humans Don't Read Your Book Reviews

The book and media industries are going through interesting times, to put it mildly. As physical books prepare for their demise, the confusion around pricing of digital ones grows. Yet, whether physical or digital, to sell books you need marketing. People need to hear about a book before they buy it.

Alex Iskold of ReadWriteWeb haz most of teh bomb on online book reviews.

In short, most reviews published online as part of large media outlets' contribution to book reviewing ARE NOT TAGGED SO THAT GOOGLE CAN FIND THEM (and I want to shout this from the rooftops, I do). Which partly explains, as you'll see if you read further into that post, why all the reviews I ran here last year carrying 'review' in the title did quite well in search results, thanks.

A Simple Way to Please Google

So how should the book reviews be tagged?

To start with, the title needs to make it clear, that this is a book review. Of course humans may find a more subtle title more enticing, but for the sake of machine: Book Review: Manhood by Mels has to be present. It would be even better to mark up that this is a book review, and here is the book title and here is the author.

Next, the post needs to be adorned with the right tags and keywords. L.A. Times' reviews are certainly very clever, but again, Google does not get humor. A better tag would the title of the book, the name of the author and the non-conspicuous phrase "book review".

Iskold obligingly runs through how this important task might be done better through the use of structured formats or markup, of which there are now quite a few breeds around. This is of course so damn important I had to stop my self-imposed Internet weaning exercises to tell you all.

I'm assuming you've looked at Iskold's example of tagging, but I will go further and tell you why Zadie Smith's review of E.M. Forster's radio show scripts in the New York Review of Books, the link to which I fluffed a while back, can be easily found online, even without review in the title: because the NYRB in their wisdom used a structured markup known to all good cataloguers and digital archivists as Dublin Core, a subset of RDF (see the comments in Iskold's post for further technical discussion on these matters.) This is an EVEN BETTER WAY, particularly if your subbies don't want to see a page full of titles screaming REVIEW. It works well, however my little reviews also work with none of this fancy metadata (to play with that I have to pay Typepad more dollars), only with review in the title.

(If you want to see the tags, go to View in the browser toolbar, and click on Page Source for a similar view to that shown in Iskold's post. Typepad doesn't like the long lines of markup text, so I won't waste time producing it here.)

What, then, are newspaper owners thinking while they skinny up our Saturday papers and threaten to lock up their book reviews behind paywalls, all for the lack of a little measly markup? "Hooray, no more book reviews, nobody has complained so they obviously weren't reading them anyway"??? Don't know why = don't want to know....!!

Do they care? One has to wonder, seriously. The small amount of research I have done into this across several sites, including Book Forum (where not all metadata in reviews is equal) and Flavorwire (which often has 'review' in titles, though they do lose some findability in a 'Daily Dose' reviews segment) suggest that the very simple act of including book review or review in the title can make all the difference, and structured markup is teh bomb.

The first of these two practices is very common web knowledge, and surely known to most sub-editors by now - or are we finding out the hard way, watching book reviews in papers die online so that media bosses can pretend they could never afford them anyway?

Bear in mind that the search online for a review does not take into account the browsing aspect of review reading which is such an inviting aspect of  the 'paper' newspaper (although a featured reviews webpage could assist with that).

The Age has it right, for the reviews that are put online anyway, following Iskold's suggestion that Book Review be in the title tag:

<title>A Woman of Seville - Book Reviews - Books - Entertainment - theage.com.au</title>

returns Carmel Bird's recent review of Sallie Muirden's A Woman Of Seville at the top of world and Oz searches.

The Australian failed this test on one example I tested - consequently the review of Lenny Bartulin's The Black Russian which was put online on the weekend is practically invisible. Here's its metadata:

<title>The art of being a bad luck magnet | The Australian</title>

<meta name="description" content="The Black Russian By Lenny Bartulin Scribe, 261pp, $27.95" />

<meta name="keywords" content=" crime thriller, Phillip Marlowe" />

Venero Armanno's November 2008 review of The Slap, which did have 'Book Review' in the keywords tag, though not the crucial title tag, has disappeared without trace, while my review still rides the Google waves and continues to draw traffic. This is a great pity, and I have linked to it from my review today.

If you're a reviewer or an editor, get thee to the IT department, go, go, and get it sorted. While you're there, tell them to put all the weekend reviews online very early on Saturday morning, too (this, the Australian does, and the Age dawdles unconscionably over). So bloggers can link to those fine examples of reviewing in the print pubs and go on doing what we prefer to do, engaging in conversation and chatty recommendation, hopefully a la Forster, instead of being blamed unfairly for the nonfindability of newspaper reviews. You have librarians, guys - clearly The Age is using theirs.

As you were. Transmission will probably resume in June. May not talk much more about this in the comments as there is a lot of stuff going on here. But I did want to share these words of vindication.