I'm releasing BestSFBooks today:
http://www.bestsfbooks.com
I was inspired by this HN post from a few weeks ago:
http://news.ycombinator.com/item?id=2978027
BestSFBooks ranks science fiction/fantasy books according to how many awards they've won or been nominated for. I included all the big ones (Hugo, Nebula, etc.) and also more obscure awards that I like such as SF Site Editor's choice. Once I had all the awards in there it was easy to start creating a New Book list based on award winning authors.
I could tell the app was working as soon as I saw the two book lists on the home page. They're excellent!
The stack is virtually identical to the stuff I used to build PickHealthInsurance. BestSFBooks is built with Rails 3.1.1 (HAML, Sass, CoffeeScript). It's hosted on Heroku and MongoHQ. I used the Twitter Bootstrap CSS toolkit, which I continue to find hugely innovative and useful. As always, data acquisition and cleanup was the hardest part.
I've wanted to build something like this for quite a while. As my time becomes more valuable, I'm becoming less tolerant of bad books. For a laugh, check out the "prototype" that I created ten years ago - http://www.gurge.com/amd/top100
Please send feedback!
I'm most interested in this piece of the project. What were your particular tools and methodologies? How long did it take you, once you identified your data sources? Any interesting stumbling blocks or problems that were solved along the way?