Increasingly, governments and businesses will need to recognize that privacy is a two-way street. Consumers may be willing to share more of their personal information (in order to gain benefits) but they expect a greater degree of transparency in return.
This study didn’t address libraries directly, but the results are interesting, especially in light of recent issues with Amazon Kindles.
You don’t want a book recommendation from someone who hasn’t read the book. So why, asks a new project called BookLamp, would you want to rely on an electronic recommendation based on marketing data rather than book content?
BookLamp is launching a new kind of book recommendation engine today that scans the texts of its partner publishers to establish what it calls “Book DNA.”
Much like Pandora assigns specific qualities to music, BookLamp measures the story components of a book (characteristics like history, domestic environments, physical injury) and how it’s written (density, pacing, dialog, description, motion).
It uses these descriptions to suggest books you might like based on a book you’ve liked in the past, turning up books that match the actual style and content of the text rather than books people like you have purchased in the past. The objective is an improved online browsing experience.
For a more detailed look at this book recommendation system, check out the article at salon.com.
But how did customers respond to this pricing decision? They were outraged! As you can see on the product page, the book has been overwhelmed with one-star reviews based not on the quality of the book itself but instead on the perception of greed and unfairness on behalf of the publisher. “junk,” writes Amazon reviewer Juan M. “It’s ridiculous that the E-BOOK is as much as the physical copy. Greed indeed.”
What I was hoping for from this article was a discussion of Amazon’s ambiguous rating system and how it can affect potential readers’ perception of the content itself. Unfortunately, the author didn’t go that direction, but rather discussed why he feels perfectly comfortable paying more for an ebook than a hardcover. That’s fine for him, but clearly a lot of people don’t agree.
“From a utility point of view,” .mp3s are a lot more convenient than CDs. You can get them whenever you want them. You can sample music with very little commitment. You can have the same music at home, in your car, on your phone, and with you at the gym. You can make playlists — and you don’t even need piles of cassette tapes and a boombox to do it! But if a record label tried to charge more than a CD for an .mp3 album, no one would buy it. Record labels are having enough trouble getting people to buy downloaded albums that cost less than CDs. This illustrates pretty clearly that consumers aren’t used to paying for convenience. They think they’re paying for plastic and laser encoding and album art. The more stuff they have in their hand, the more it should cost.
It’s obvious from situations like this with Michael Connelly’s latest book that Kindle users do not think they should be forced to pay more for convenience. Not everyone is as zen about it as the original poster. That’s a fact.
The question that I find more interesting is whether or not people should be leaving negative reviews in a situation like this. After all, what is being graded? Do those stars at the top of the page reflect the content of the book, as written by Connelly himself? Do they represent Amazon’s delivery of the product? What if some reviewers take off stars for poor cover design, or an unpleasant typeface?
Because of the poor reviews, someone might glance at the listing for the book and decide it’s not worth reading. Not everyone is going to read enough reviews to discover most people have a problem with the publishing model. I’m really curious how Amazon’s collaborative filtering handles such situations, too. Is this going to stop the book from appearing as a recommendation for people who liked other similar products? Even if the price gets lowered, which appears to have happened in this situation and could spark a whole other discussion, those reviews are still there. Not everyone is going to go back and change them, especially not the unhappy people who already paid more.
Netflix uses a software algorithm to recommend movies and Zappos uses one to recommend shoes. Now Goodreads, the social network for book lovers, is introducing an algorithm to recommend books.
Goodreads was started in 2006 for people who wanted to talk about books online. Its 4.6 million members load their virtual bookshelves with books they’ve read, are reading and want to read, rate and review them, and discuss them with friends and others on the site.
On Thursday, Goodreads will announce that it has acquired another start-up,Discovereads.com. It uses machine learning algorithms to analyze which books people might like, based on books they’ve liked in the past and books that people with similar tastes have liked.
I find Goodreads to be a little cumbersome, so I don’t use it as much as I’d like. I love recommendation systems, though, so maybe I should give it another try. The social recommendation idea behind Goodreads is actually great for serendipitous discovery, but the addition of collaborative filtering should provide another helpful angle for users.