[mklopets]

We definitely don't need an API. Then again, we basically don't need most APIs. This doesn't estimate the article's complexity as of now, but its main point is not just getting the length of the entire site, but locating the most likely main content area and THEN doing the /250.

[fweespeech]

You can use something like:

https://github.com/grangier/python-goose

https://pypi.python.org/pypi/textstat/

+ using word counts that adjust for reading ease