I'm glad you asked this question ;-) The contest's purpose is not to replace real open data with scraping bad HTML, but to show people what kind of things woul

I'm glad you asked this question ;-) The contest's purpose is not to replace real open data with scraping bad HTML, but to show people what kind of things would be possible if government freed the data. The projects are just examples but not full featured, because the scraping approach has obvious limitations. In our case, all data is scraped from the INE website (http://www.ine.es/ http://www.ine.es/), but there's one single data which couldn't be obtained from any official source in a programmatical way, one that is essential: the mapping between official districts number and the district name. We had to type that data (http://github.com/valakirka/monquartier/blob/master/config/districts.yml http://github.com/valakirka/monquartier/blob/master/config/d...) and that's the only reason of the small scope. If we had had normalized and public data, our app would include the data for all cities, towns and villages in Spain, and not only at the district level but also at the neighbourhood level, which would make it a lot more interesting (and that's what we wanted to make clear to the people). When government does its part, we'll do ours ;)