A key problem with RSS as deployed currently is that there is no way to tell from the feed whether each entry is the full content, a human generated summary or abstract, a machine-generated truncation, or just the bald title.

Even for the human reader, knowing whether one is seeing the author's summary or the complete entry requires remembering which style that author has chosen, and when entries from different sources are interleaved in a time- or subject- based aggregator, this places extra load on the reader to realise which ones are meant to be clicked through for more.

My understanding below was mistaken - Atom's idea is to ensure that <content> is always complete and <summary> is used for the various truncated forms. This should be explicitly stated in the spec.

As I understand it, Atom's solution to this is having alternative content elements under the entry, which can be labelled with 'rel' attributes to indicate their kind. The [WWW]draft makes no mention of this, but content describes some 'possible relations'.

In order to avoid the ambiguity present with RSS, I would like to suggest that each content element MUST have a 'rel' attribute indicating one of: 'complete', 'summary', 'truncation'

This would avoid the current need to check the linked HTML page to be sure that one has not missed part of the expressed content.

All content elements MUST have a rel attribute, which is one of 'complete', 'summary', 'truncation'

written by KevinMarks