Here's some things with more substance (I don't think the blog author is doing any SEO machinations, but there's just not much to learn from his last few posts. Essentially he's written Data Science advice similar to many other authors, but he's substituted the words "Machine Learning"

I could dig up other ML gotchas/guidelines posts, need to dig thru bookmarks)

http://homes.cs.washington.edu/~pedrod/papers/cacm12.pdf

http://www.inf.ed.ac.uk/teaching/courses/dme/html/datasets04... (read Challenges in each dataset)

http://alpinenow.com/blog/machine-learning-is-not-black-box-...

http://research.microsoft.com/en-us/um/people/minka/papers/n...