#host configuration should be http://www.neh.gov/news/humanities/


#meta data 
title:substring-after(substring-after(//title,':'),':')
author:substring-after(//h2[@class = 'subHead'],'By')
date:substring-before(substring-after(//title,':'),':')

#img and caption handling
wrap_in(small)://div[@id = 'mainContent']/table/descendant::p/descendant::text()
wrap_in(fieldset)://div[@id = 'mainContent']/table

# clean up
strip: //table[@class = 'marginpaddingTop']
strip: //h2[@class = 'subHead']

test_url: http://www.neh.gov/news/humanities/2011-11/IslamicScholar.html