JD_Toims - 12:38 am on Nov 14, 2013 (gmt 0) [edited by: JD_Toims at 1:04 am (utc) on Nov 14, 2013]
I think I have an example that will make sense of the differences, so hopefully my triple reply will be okay.
Think of HTML as pie...
Document = The whole pie.
Section = The portion people are given.
Pieces [groupings] = The ingredients people eat.
In trying to "get" HTML you, imo, should be able to "start with the pie [document]", then "cut the portions [section it]", and let people "taste the ingredients [read the pieces]" -- In other words, the "ingredients" [pieces / grouping(s)] of a "portion" [section] of a "pie" [document] are usually what people "feast on", and if you can't identify those as a search engine, you likely can't "serve the best results", but the responsibility does not completely fall on search engines for "understanding" what's being "fed to people" on any given website, because if someone "just codes whatever where ever" and a search engine can't "put the ingredients together" then the "pie" [website] doesn't ever get "served".
[edited by: JD_Toims at 1:04 am (utc) on Nov 14, 2013]