Recognize when the same content appears in several contexts

[zackw]

(Related to, possibly the same as, #12)

Many blog-type sites have the same material appear both in a canonical location, and in a variety of "index" pages which aggregate material that has something in common. For instance, using my own site: the article https://readings.owlfolio.org/2013/analysis-https-cert-ecosystem/ appears at that canonical URL, but is also embedded in nine other pages:

• https://readings.owlfolio.org/a/zakir-durumeric/
• https://readings.owlfolio.org/a/james-kasten/
• https://readings.owlfolio.org/a/michael-bailey/
• https://readings.owlfolio.org/a/j-alex-halderman/
• https://readings.owlfolio.org/t/field-observations/
• https://readings.owlfolio.org/t/https/
• https://readings.owlfolio.org/t/pki/
• https://readings.owlfolio.org/2013/
• https://readings.owlfolio.org/2/ <-- unstable

There's an existing microformat (hEntry) that lets sites annotate such index pages so that search engines can figure out what the canonical URL of any given chunk of material is.

hypothesis should recognize this microformat, and

• record annotations under the canonical URL when material is annotated via one of the index pages;
• reveal annotations for the canonical URL when viewing an index page.

[tilgovi]

Closer to #119