This is what I got from the article. data.gov has LOADS of datasets for the USA, as well as hundreds of other APIs from different departments that can be used to retrieve that data in real time.

The problem comes just after you click "download" and realize you have a 600MB XML file listing the Ph levels of groundwater sites around the country for the past 25 years. How in the hell are you going to make that data an interesting draw that users want to read or learn about?

I have no skill for frontend design, but I've recently begun to learn how to display data in an intelligent manner. If anyone is interested, these books have helped me:

Information Dashboard Design: http://oreilly.com/catalog/9780596100162

Visualizing Data: http://www.amazon.com/Visualizing-Data-Explaining-Processing...

Beautiful Data: http://www.amazon.com/Beautiful-Data-Stories-Elegant-Solutio...

> How in the hell are you going to make that data an interesting draw that users want to read or learn about?

People are probably looking for that data. Find out exactly what they're searching for, and tailor your content/site presentation around those keywords. You can use Google Keyword Tool to find this information:

https://adwords.google.com/select/KeywordToolExternal

Right, but you still need to present it in a way that is easy for people to understand. XML is a fantastic way to deliver a large amount of data, but it is hardly parse-able to anyone who doesn't know what a markup language is. JSON as well. My point was that "take data, throw adsense on it" is hardly a simple plan.

Developers need to know how to display data in an easy to understand, easy to follow, and easy to consume manner in order to get people to the site. Keywords and SEO only go so far. You still need readable content for humans.

Right, you need to curate the data as you import it. I assumed that was assumed. You need to make it readable for humans and for search engines (SEO).