Unrelated, but should we also be searching extractedUrl, extractedProviderUrl, and/or title?
Basically, "title" because the extractedTitle will be less useful if the scraped page was behind a login (ie: facebook, or something on localhost that embedly can't touch).
@nchapman said "Good points. Let's do a follow up PR for these. @pdehaan will you create an issue?"
Per https://github.com/mozilla/chronicle/pull/329#discussion_r24628825 and #324:
@nchapman said "Good points. Let's do a follow up PR for these. @pdehaan will you create an issue?"
THIS IS THAT ISSUE!!!