First there was Amazon. Now, there’s a flourishing community of new sites, apps, and organizations that want to provide suggestions for the next books that we’ll love (and help us avoid the ones we’ll hate).
Some offer bespoke suggestions from real people; others are driven by algorithms. This recent piece on Co.Exist looked at the advantages and disadvantages of the two approaches.
A new site, Readgeek, is entering the mix and has already gotten an enthusiastic response from Reddit last month. To use it, you rate as many books as you would like, and the service sends back a long list of suggestions, similarly to the Amazon-owned site Goodreads.
Launched by a developer in Germany, Readgeek’s aim is to provide unexpected suggestions and help people “avoid the filter bubble.” Rather than suggesting books that are similar to books a user has rated highly, it offers up books liked by other users who have similar tastes. The main difference between Readgeek and Goodreads is that Readgeek offers a prediction for how much a user will like almost any book he or she looks up. It also offers much more nuanced ratings, via a 20-step rating on a scale from 1 to 10 (rather than only 5 stars). The last difference, as the developer Uwe Pilz put it: “Well, and it is not owned by Amazon :)”
Readgeek performed only so-so when I tried it out after rating the first 25 books I’ve read that the site happened to show me. It suggested a collection of the cartoons, The Far Side by Gary Larson, without even knowing I’m already huge fan. It suggested another book I love, Labyrinths, by the writer Jorge Luis Borges, but that wasn’t a huge stretch given that I had given already given a 10/10 rating to the book, 100 Years Of Solitude of Solitude, which is the most popular book by Gabriel Garcia Marquez, an author often mentioned in the same sentence as Borges.
It also suggested several books that I do like, and I definitely don’t hate–but all are the kind of books many people read in high school English class (or at least I did), such as A Tree Grows in Brooklyn, The Things They Carried, and Darkness at Noon. These are the kind of books that people probably have in common for this very reason.
All in all, it seems like a promising service that could use some improvement. But if you’re looking for more human personalized suggestions, you can try your local library.