krawlr is a toy I'm making to improve my Python and learn more about scraping and how to do it in a courteous manner while adding value to the fetched information. I'm reading a few books on the matter, looking at the code, and trying to write nice code.
The crawling/scraping part will serve for another purpose though: statistical learning and data visualization on data from a local ecommerce website. People here are always talking about cars (prices, events that took while driving, etc). There are a lot of cars being bought and sold and I thought about a graph that'd show a good deal as an outlier. A scatter plot with a regression curve where the dots are the site postings (price vs mileage, etc). Maybe PCA. Who knows.
I also liked what MIT's The Observatory of Economic Complexity does. I learned a lot of things looking at the treemap. So I want to do something similar with D3.js.